Paperlane

no. 4 · from the archive← Tue 27 Oct archive Fri 30 Oct →

Wednesday, 28 October 2026

719 papers in 32 fields

Analysis of PDEs45

For $p:[0,1]\to(0,1)$ a continuous function, we consider the Luxemburg functional \[ Λ_{p(\cdot)}(u):=\inf\left\{δ>0:\int_0^1\left|\frac{u(t)}δ\right|^{p(t)}\,dt\leq 1\right\}. \] We develop a suitable functional analytic framework for $Λ_{p(\cdot)}$ so that it can be incorporated into a positive order cone, which can then be used in the analysis of nonlocal differential equations having the general form \[ -A\left(\int_0^1 b(1-s)(u(s))^{p(s)}\,ds\right)u''(t) =λf(t,u(t)),\qquad t\in(0,1). \] By means of a specific numerical example, we demonstrate that, as opposed to a methodology relying only on the supremum norm, our methodology via the Luxemburg functional yields an appreciably sharper localisation of any solutions of the nonlocal equation.

comments
48 pages, 2 figures
fields
Analysis of PDEs, Functional Analysis
pdf ↗ page arXiv:2610.35878 ↗

We study the simultaneous recovery of an unknown initial signal and unknown vertex wave speeds for the wave equation on a finite graph from time-dependent measurements on a prescribed set of vertices. With the graph geometry known, we first show that a single experiment determines both quantities under a new two-witness zero forcing condition and generic nondegeneracy assumptions. We then develop a more general frontier-propagation framework in which the redundancy required for simultaneous recovery is distributed across successive graph contours rather than localized at individual forcing vertices. We derive Lipschitz stability estimates for the frontier propagation problem. We also show the recovery and stability results extend to positively weighted graphs.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36106 ↗

\[ u_t = \nabla \cdot (uv\nabla u) - χ\nabla\cdot (u^2 v\nabla v) + \ell uv, \qquad \qquad v_t = Δv - uv, \] is considered in a smoothly bounded domain $Ω\subset {\mathbb{R}}^n$ and with $χ\ge 0$ and $\ell\ge 0$. Within a suitable concept of solvability, a statement on uniqueness of solutions emanating from initial data $(u_0,v_0)$ is derived under the mere assumption that $u_0\in L^\infty(Ω)$ and $v_0\in W^{1,\infty}(Ω)$ are such that $u_0\ge 0$ and $v_0>0$ in $\overlineΩ$; in particular, no condition on continuity of $u_0$ is imposed. This result is thereafter used to reveal that in physcally meaningful scenarios, for any initial data which are additionally assumed to be continuous and nontrivial, such solutions can always be found, and that these solutions depend continuously on the initial distributions not only in bounded sets of the form $Ω\times (0,T)$, but actually throughout $Ω\times (0,\infty)$, in a suitable sense.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36117 ↗

We study the Keller problem for the Dirac operator $\mathcal{D}_m - Vβ$ in $\mathbb{R}^d$, and minimize its lowest positive eigenvalue over all potentials $V$ with prescribed $L^p$ norm. We prove that optimal potentials always exist and that the corresponding eigenspinors solve the nonlinear Soler equation. This provides a variational proof of existence of solitary waves for this equation. We prove that there are no zero modes for such operators in dimensions $1$ and $2$, and conjecture that zero mode may exist in dimension $3$. In dimension $d = 2$, we provide numerical evidence that the optimal potentials concentrate on flower-shape regions, which are not necessarily radial, showing that Faber–Krahn type results may fail for such Dirac operators. In dimension $1$, all quantities are explicit.

comments
23 pages, 4 figures
fields
Analysis of PDEs, Mathematical Physics, Spectral Theory
pdf ↗ page arXiv:2610.36240 ↗

We investigate conjugate flows in a two-layer fluid system with a flat bottom and a free surface. These are up- and downstream states consistent with simultaneous conservation of mass, momentum, and energy; and therefore candidate end-states of one-dimensional dissipationless bores. By choosing variables in a specific way, we are able to explicitly parameterize all physical non-trivial conjugate flows in the system, using only rational functions. This two-parameter family can be categorized according to whether the states are compatible with (1) a rising surface, and falling internal interface; or (2) vice versa. In particular, we rule out the existence of monotone bore solutions, consistent with an earlier result due to Chen, Walsh, and Wheeler. However, here we do so already at the level of conjugate flows. Separating these two distinct categories is a one-parameter family of critical trivial conjugate flows. The complete description obtained in the present paper serves as an important stepping stone towards finding actual realizing heteroclinic orbits connecting such conjugate states.

comments
15 pages, 4 figures
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36242 ↗

We revisit the Calderón problem for the fractional Schrödinger equation and prove that the potential is determined by Dirichlet-to-Neumann maps at a single exterior point, with exterior inputs varying in a disjoint open set. This holds for continuous potentials for all $0<s<1$ and for bounded potentials when $s\geq 1/4$. For ordered nonnegative bounded potentials, one fixed nonzero nonnegative input and one point measurement suffice for every $0<s<1$.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36253 ↗

Let $Ω\subset\mathbb{R}^n$ be a bounded Lipschitz domain and let $p:Ω\to[1,\infty)$ be a measurable, essentially bounded exponent; no regularity of $p$ -- in particular, no log-Hölder continuity -- is assumed. For a general non-negative measurable kernel $\mathcal{K}(x,y)$ admitting a radial bounding function $κ(x,r)$ in the second variable, we establish the fractional modular Poincaré–Wirtinger-type inequality $$ \iint_{Ω\timesΩ}\mathcal{K}(x,y)\,|u(x)-u(y)|^{p(x)} d y\, d x \leqslant C\int_Ωc(x)\left(\int_Ω \frac{|\nabla u(z)|^{p(x)}}{|z-x|^{n-1+θ}}\, d z\right) d x, $$ valid for every $u\in W^{1,1}(Ω)$, in particular every $u\in W^{1,p(\cdot)}(Ω)$, and every $θ\in[0,1)$. Here $$ c(x)=\int_0^{D}r^{n+p(x)-2+θ}κ(x,r)\,d r, $$ $D$ is the diameter of any fixed ball containing $\overlineΩ$, and the constant $C$ depends only on $n,p^+,θ$ and the fixed domain $Ω$. For the fractional kernel $\mathcal{K}(x,y)=|x-y|^{-n-λp(x)}$ we obtain a variable-exponent Gagliardo-modular estimate. The proof replaces the starshaped chain decompositions used there by a combination of a Morrey-type estimate on starshaped domains, a positive-kernel proof of weighted Poincaré inequalities, and one reflection extension for the weights $|z-x|^{-(n-1+θ)}$. The constants are uniform down to exponent $1$, and the inequalities are understood in the extended nonnegative sense when either side is infinite.

fields
Analysis of PDEs, Classical Analysis and ODEs, Functional Analysis
pdf ↗ page arXiv:2610.36293 ↗

The Kotani-Last problem and the Deift conjecture are two well known statements, which both concern the almost periodicity properties of the functions under consideration, and which have both been disproved in their original formulation, which involves Bohr/Bochner almost periodicity. To contrast these disproofs, and to further elucidate the essence of these two challenging problems, we establish positive results regarding each of them by showing that Weyl almost periodicity can be deduced whenever the Sodin-Yuditskii setup can be implemented. The latter requires that the Schrödinger operator associated with the function in question has purely absolutely continuous spectral measure and a spectrum with finite total gap length and the Parreau-Widom property. We also show that the known examples for which Bohr/Bochner almost periodicity has been shown or suggested to fail in the context of the Deift conjecture are in fact Weyl almost periodic. As a result, an accordingly modified version of the Deift conjecture remains a tantalizing open problem.

comments
46 pages
fields
Analysis of PDEs, Mathematical Physics, Dynamical Systems, Spectral Theory
pdf ↗ page arXiv:2610.36403 ↗

In this paper, we are going to deliver a completely self-contained proof of the variational characterization of the principal eigenvalue of $-Δ+q$ in $Ω$ under Robin boundary conditions of non-classical type, $\frac{\partial u}{\partial n}+β(x)u=0$ on $\partialΩ$, where $β(x)$ is allowed to change sign on $\partialΩ$. The classical case occurs in the very special case when $β\geq 0$ on $\partialΩ$.

comments
15 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36457 ↗

This work is devoted to study the Cauchy problem for the quantum Navier-Stokes equations in $\mathbb R^N$, $N=2,3$, with density dependent viscosities $μ(ρ)=ρ,λ(ρ)=0$, quantum capillarity $κ(ρ)=1/ρ$, which was derived by Dunn and Serrin (see Arch. Ration. Mech. Anal. 88(2):95-133, 1985) as a phase transition model. For arbitrarily large initial data satisfying suitable regularity and decay assumptions, with density positive at every finite point and vanishing at infinity, we prove the existence and uniqueness of global strong solutions without symmetry assumptions. The admissible pressure range is $γ\ge 1$ in two dimensions and $1\leγ< 8/3$ in three dimensions. On every finite time interval, the density remains pointwise comparable to its initial profile, and both the effective velocity and the logarithmic density gradient remain bounded. The central estimates require no uniform positive lower bound for the density. When $γ>1$, a suitable combination of the density and effective velocity equations allows the maximum principle to yield an $L^\infty$ bound for the effective velocity using only an upper bound for the density. We then introduce the relative density with respect to a fixed profile determined by the initial data. In the isothermal case $γ=1$, we combine estimates for this relative density with a scalar inequality involving the effective velocity and the logarithmic density to close the coupled estimates and obtain the effective velocity bound. Combining the relative-density bounds with weighted gradient estimates and localized De Giorgi iteration, we control the logarithmic density gradient and close the higher-order estimates.

comments
61 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36480 ↗

It is well known that solutions to deterministic Keller–Segel chemotaxis systems can blow up in finite time under certain conditions. To investigate whether noise can prevent such blow-up, we study parabolic–elliptic and fully parabolic Keller–Segel systems on smooth bounded domains with pointwise superlinear multiplicative Itô noise $σ_0u^r\,\mathrm dW_t$, where $W$ is a real Brownian motion. Our results show that superlinear multiplicative noise can prevent finite-time blow-up in Keller–Segel systems. For $r>3/2$, we prove global existence and pathwise uniqueness of nonnegative mild solutions in arbitrary spatial dimension, for every $σ_0>0$. At $r=3/2$, both systems are globally well-posed in two dimensions for every positive noise intensity; in arbitrary dimension, the parabolic–elliptic system is global under $σ_0^2\ge2χ$. For $r>3/2$ in the fully parabolic case, the proof combines a logarithmic estimate for the time derivative of the chemical signal with comparison of the local solution to the solution of a scalar stochastic differential equation. The two-dimensional case follows from mass quadratic variation, deterministic maximal regularity, and localized energy estimates. We also establish almost-sure $L^1$ extinction and algebraic decay of fractional mass moments.

comments
25 pages
fields
Analysis of PDEs, Probability
pdf ↗ page arXiv:2610.36485 ↗

This paper studies a singular Toda system with linear source terms on a compact surface $Σ$ with boundary $\partialΣ$. The aim is twofold. First, using isothermal coordinates and reflection arguments near the boundary, we perform blow-up analysis obtaining a compactness result. Second, by assuming that $\partialΣ$ has at least two connected components, we prove a general existence result via a min-max scheme combined with a retraction argument.

comments
28 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36548 ↗

We study weak solutions of the incompressible and the compressible Navier–Stokes–Vlasov systems on a bounded domain, where a viscous fluid is coupled to a dispersed particle phase through a drag force; the fluid obeys the no-slip boundary condition and the particles are subject to specular reflection. For solutions with Besov regularity, we show that the Vlasov equation is renormalizable in the sense of DiPerna–Lions, and we give regularity thresholds for the validity of the global energy equality. If the distribution function $f$ belongs to $B_p^{α,\infty}$ in space and velocity with time integrability $p$, and the fluid velocity $\mathbf{u}$ belongs to $B_q^{β,\infty}$ jointly in space and time with integrability $q$, then the energy equality holds provided $α+β>1$ and $β$ exceeds an explicit bound depending on $p$, $q$ and the dimension, under a natural integrability assumption on the pressure; an analogous criterion holds for the compressible system when the density is bounded from above and below. The condition $αp>1$ imposed on $f$ is not merely technical: it is exactly the range in which the macroscopic momentum flux possesses a boundary trace, so that the specular reflection condition passes to the macroscopic energy balance. When the particle phase is absent, our thresholds coincide with the criteria $β>\frac12+\frac1q$ and $β>\frac3q$ obtained by Yu for the incompressible Navier–Stokes equations alone. The proofs combine a boundary cutoff argument with commutator estimates based on a scale-localized Besov modulus, together with a boundary-strip analysis.

comments
Submitted to Arch. Ration. Mech. Anal
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36579 ↗

We study time-periodic Keller–Segel dynamics in a translating KPP environment. In the parabolic–elliptic regime, we prove global boundedness, identify the sharp speed \(c_*=2\sqrt{\bar a\,\bar r}\), establish persistence in the interior region \(|s|<c_*\), \(s\cdot e<c\), and derive the complementary sharp upper bounds; for a moving favorable slab, the sign of a principal Floquet exponent determines persistence or extinction under the stated weighted-resolvent condition. For every fixed \(τ\ge0\), under constant population diffusivity and sufficiently weak taxis, we further construct locally unique planar forced time-periodic waves in \(\mathbb R^N\). The main new analytic mechanisms are an exact Yukawa first-contact estimate, which controls the chemotactic drift at the KPP leading-edge scale without losing the sharp propagation threshold, and a two-sided weighted Fredholm–monodromy continuation framework for the temporally nonlocal signal equation, with explicit control of the far-field spectral gaps and the translational neutral mode. To the best of the author's knowledge, planar forced time-periodic waves in \(\mathbb R^N\) for shifting Keller–Segel systems have not been investigated in any previous studies; earlier forced-wave results are confined to substantially more restrictive settings, in particular one-dimensional parabolic–elliptic models. Thus the present construction provides the first multidimensional planar forced-wave theory for a time-periodic shifting Keller–Segel system with temporal signal memory.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36627 ↗

For a $d$-dimensional closed hypersurface in $\mathbb{R}^{d+1}$, we reformulate the (stationary) tangent Stokes, transient tangent Stokes, and transient tangent Navier-Stokes equations by replacing the divergence equation with a pressure Poisson equation, thereby circumventing their saddle-point structure. The pressure Poisson equation hinges on a commutator identity between the divergence and the surface diffusion operator, whose lower-order term depends on the Ricci curvature, and requires no boundary conditions for the pressure. We show that the reformulated operator is the composition of a linear, block-triangular data map with the primal operator. Consequently, each reformulation is well-posed if and only if the primal problem is, and solving it with arbitrary data amounts to solving the primal problem with a modified divergence datum. For hypersurfaces of class $C^2$ and data of the pressure Poisson equation in $H^{-1}$, the tangent Stokes reformulations can be posed in $H^1$-based spaces, which paves the way for conforming discretizations that do not require a discrete inf-sup condition.

comments
25 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36641 ↗

We prove sharp well-posedness and soliton resolution for the half-wave maps equation on the real line $\R$ in \(H^{1/2}_\bullet\). This regularity threshold for global well-posedness is optimal, as we show by constructing examples of ill-posedness in \(H^s_\bullet\) for any \(s<1/2\). Moreover, with the spectral assumption satisfied by a dense set of initial data in the energy space, we prove that solutions in \(H^{1/2}_\bullet\) asymptotically decouple as an infinite superposition of solitons. Our results extend the known global well-posedness and finite soliton resolution for rational data in \(H^\infty_\bullet\) to sharp global well-posedness and soliton resolution for general data in \(H^{1/2}_\bullet\). Different from rational solutions, general data can also involve an infinite discrete spectrum, which makes the reduction analysis framework on the finite rank no longer applicable. For sharp well-posedness, we present a new isometry criterion and stability results for resolvent transform in the explicit formula. For soliton resolution, we develop a spectral method for this equation based on direct resolvent analysis and a novel trace norm estimate that upgrades energy convergence to uniform convergence. Our argument also yields sharp well-posedness and soliton resolution for the generalized half-wave maps equation with targets in the complex Grassmannians, extending the classical sphere-valued case.

comments
Comments are welcome!
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36676 ↗

This paper is concerned with the dissipative weak solutions to the three-dimensional stochastic Euler equations. For any $γ\in \left[ {0,\frac{1}{5}} \right)$, we establish the existence of $γ$-Hölder continuous weak solutions that exhibit a strict energy dissipation in expectation. For any $γ> \frac{1}{3}$, we prove that the three-dimensional stochastic Euler equations do not possess dissipative solutions satisfying the corresponding spatial $γ$-Hölder regularity. It is well known that the stochastic forcing inevitably gives rise to additional stochastic effects to the Reynolds stresses, which cannot be treated by the differential estimates in the classical convex integration scheme introduced by Buckmaster, De Lellis, Isett and Székelyhidi (Ann. Math., 182(1):127–172, 2015). To address this difficulty, we develop a novel material integration technique into the stochastic convex integration scheme, so that those stochastic Reynolds stresses are quantified.

fields
Analysis of PDEs, Probability
pdf ↗ page arXiv:2610.36819 ↗

This paper is concerned with the simultaneous recovery of the trajectory and intensity of an unknown moving point source in an inhomogeneous acoustic medium. We prove that the whole-boundary measurements over a finite time interval uniquely determine the moving source. Furthermore, under suitable propagation and rank assumptions, Lipschitz stability of determining both the trajectory and intensity is derived using waveforms measured at four fixed receivers outside the source region. A novel Volterra–Euler reconstruction scheme is developed for solving the inverse problem, and the error estimate is derived for the proposed scheme. Numerical experiments are presented to demonstrate the effectiveness of the proposed numerical method.

comments
22 pages, 8 figures
fields
Analysis of PDEs, Numerical Analysis
pdf ↗ page arXiv:2610.36830 ↗

We study cooperative nonlocal transport on a circle with species-dependent constant velocities and arbitrary $L^1$ probability kernels. For periodic coefficients, the period map is a compact perturbation of the no-jump map. We construct separate receiving and sending characteristic scans: under a two-tree condition based at one active component, a positive Floquet profile exists exactly when the full period map has strictly larger spectral radius than its no-jump part. The other jump rates may vanish. The scans provide an integral minorant and positive jump return; a separate coefficient inequality verifies the spectral gap. Smooth locking examples distinguish communication from frozen-time mixing, while a resonant example separates positive return from spectral separation. Autonomous nonzero velocities and union-graph connectivity yield a simple principal pair; one active component suffices for exponential mixing. We also prove local perturbation results. For quadratic self-regulation, the same principal projection gives a transverse response whose averaged cubic coefficient is a nonnegative asymptotic variance. It determines both the critical $t^{-2}\log t$ correction and the second-order period-averaged amplitude of the positive branch.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36851 ↗

We study the following degenerate elliptic equation with combined nonlinearities and prescribed mass \begin{equation*} \begin{aligned} \begin{cases} -Δ_γu+ωu+|u|^{p-2}u=|u|^{q-2}u \ \ \mbox{in }\mathbb{R}^{N}, \\ \int_{\mathbb{R}^{N}} |u|^{2} \mathrm{d}z=c, \end{cases} \end{aligned} \end{equation*} where $Δ_γ$ is the Grushin operator and $ω\in\mathbb{R}$ appears as an unknown Lagrange multiplier. We characterize the critical mass threshold and the energy ground state solutions. Firstly, we prove the existence of optimizers for the Gagliardo-Nirenberg inequality involving the Grushin operator together with estimates for the related best constant. Secondly, we characterize the critical mass by the best constant of Gagliardo-Nirenberg inequality and prove the existence of $x$-radial energy ground state solutions. Finally, we describe the relation between the $x$-radial action and energy ground state solutions, and the optimizers of Gagliardo-Nirenberg inequality. The strategy developed here is based on the interplay between the mass-preserving scaling and a sharp Gagliardo-Nirenberg inequality. It therefore suggests possible extensions to other prescribed-mass problems with competing homogeneous terms and threshold-mass phenomena.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36927 ↗

In this paper, we focus on the emergence of large densities of a cell population in finite time in the following parabolic-elliptic Keller-Segel system with signal-dependent motility and a source term of logistic type \begin{equation} \begin{cases} u_{\varepsilon t} = Δ(e^{-v_\varepsilon}u_\varepsilon) + \varepsilon(κu_\varepsilon - μu_\varepsilon^2) ,\\ 0 =Δv_{\varepsilon}-v_\varepsilon+ u_{\varepsilon}. \end{cases} \end{equation} in a smooth bounded domain $Ω\subseteq\mathbb{R}^2$ with homogenuous Neumann boundary conditions and parameters $κ,μ>0$ and $\varepsilon\in(0,1)$. We will show that the solution $u_\varepsilon$ can exceed any certain value $M>0$ before a time $T(M)>0$ under the condition that $\varepsilon>0$ is sufficiently small. To prove the result, we show convergence of classical solutions of the first system to classical solutions of the corresponding limit system as $\varepsilon\searrow 0$ \begin{equation} \begin{cases} u_t=Δ(e^{-v}u),\\ -Δv+v=u. \end{cases} \end{equation} The main challenge is to find suitable bounds for $u_\varepsilon$ and $v_\varepsilon$, that are $\varepsilon$-independent, in Lebesgue and Hölder spaces, whereas for fixed $\varepsilon>0$ global existence and boundedness is known.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.36934 ↗

We study the time evolution of a vertical-slice model for incompressible inhomogeneous rotating fluids perturbed around a stably stratified rest state with linear buoyancy profile. For $\varepsilon$-small localized initial data with vanishing potential vorticity, we prove that solutions remain of size $O(\varepsilon)$ up to times of order $\varepsilon^{-\frac43}$, and up to times of order $\varepsilon^{-2}$ in a regime where the Brunt–Väisälä frequency and the Coriolis parameter are comparable. The first result relies on linear dispersion with amplitude decay rate $\langle t\rangle^{-\frac12}$, while the second exploits the absence of three-wave resonances via a normal form transformation. Our results remain valid for small nonzero initial potential vorticity, whose size is propagated by means of an intrinsic change of coordinates.

comments
21 pages
fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2610.37013 ↗

We prove global well-posedness and scattering for the energy-critical defocusing nonlinear Schrödinger equation \begin{equation*} \begin{cases} i\partial_t u + Δu = |u|^{\frac{4}{d-2}} u, u(0) = u_0, \end{cases} \end{equation*} on hyperbolic space $\mathbb{H}^d(d \geq 4)$. We mainly utilize the concentration-compactness/rigidity method. We establish a linear profile decomposition on high-dimensional hyperbolic spaces and prove that the nonlinear profile on hyperbolic spaces can be approximated by the solutions of the energy-critical defocusing nonlinear Schrödinger equation on Euclidean spaces, then obtaining the global well-posedness and scattering of the energy-critical defocusing nonlinear Schrödinger equation on the high-dimensional hyperbolic space.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37077 ↗

In this paper, we consider the Cauchy problem to the compressible viscous non-resistive magnetohydrodynamic (MHD) system in $\mathbb{R}^3$, in which the velocity is planar while the magnetic field is vertical. For regular initial data with small energy but possibly large oscillations, we establish the global existence and uniqueness of strong solution with vacuum. In addition, the large-time behavior of the solution is investigated. A key ingredient is an application of the Bourgain–Brezis divergence estimate to handle the total pressure $ρ^γ+m^2$ for any adiabatic index $γ>1$. This enables us to control both components of the total pressure, $ρ^γ$ and $m^2$, by a single estimate that remains valid in the presence of vacuum. To the best of our knowledge, our result is the first one on global solvability to the compressible, viscous, non-resistive MHD system in the special 2$\frac12$-dimensional whole space allowing vacuum.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37078 ↗

We consider the Cauchy problem for the two-dimensional compressible micropolar fluid system on the torus $\mathbb T^2$. The initial density is bounded and nonnegative and may contain vacuum, while the initial velocity and microrotation belong to $H^1(\mathbb T^2)$. For sufficiently large bulk viscosity $ν$, we prove the global-in-time existence of solutions without a smallness assumption on the $H^1$ size of the data, provided that the initial compressible component is well prepared in the sense that $\sqrtν\|\operatorname{div}u_0\|_{L^2}$ is uniformly bounded. The proof combines the effective viscous flux with a coupled energy scheme that exploits the vorticity–microrotation dissipation at both the basic and material-derivative levels. For the linear pressure law, we also prove uniqueness in the stated solution class by means of a coupled backward parabolic problem, logarithmic interpolation, and Osgood's lemma. For fixed initial data with $\operatorname{div}u_0=0$, we also justify the limit $ν\to\infty$: every sequence of the corresponding solutions admits a subsequence converging to a solution of the two-dimensional inhomogeneous incompressible micropolar system.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37099 ↗

We study the saturation free boundary of the cell density and its interaction with the nutrient depletion interface in a coupled parabolic model for tumor growth with sublinear reactions. We establish a sharp quadratic–superquadratic growth dichotomy along the saturation interface and quantify the degeneration of quadratic profiles as they approach the joint saturation–depletion boundary. We further show that intrinsic regularity of the saturation interface can occur only at joint free boundary points, where half-space blow-ups of either component determine a unique joint blow-up profile. Finally, we prove a local backward coincidence between the joint and nutrient free boundaries near regular points, which yields spatial and temporal regularity of the joint interface.

comments
28 pages, 3 figures
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37144 ↗

We consider positive solutions of \begin{equation*} \begin{cases} -Δu=pλh(x)u^{p-1}\mathrm{e}^{u^p}&\text{in }Ω,\\ u>0&\text{in }Ω,\\ u=0&\text{on }\partialΩ, \end{cases} \end{equation*} where $p\in[1,2]$, $λ>0$, $Ω\subset\mathbb{R}^2$ is a smooth bounded simply connected domain and $h\in C^2(\overlineΩ)$ is positive on $\overlineΩ$. The cases $p=1$ and $p=2$ correspond to the Liouville and Moser-Trudinger equations, respectively. We prove that the normalized parameter \begin{equation*} β=\frac{λp^2}{2}\left(\int_Ωh(\mathrm{e}^{u^p}-1)\mathrm{d}x\right)^{\frac{2-p}{p}}\left(\int_Ωhu^p\mathrm{e}^{u^p}\mathrm{d}x\right)^{\frac{2(p-1)}{p}} \end{equation*} is bounded by a constant depending only on $Ω$ and $h$, uniformly over all $p\in[1,2]$, $λ>0$ and positive solutions. At $p=2$, this excludes positive constrained critical points of the weighted Moser-Trudinger functional at sufficiently large Dirichlet energies. For $p=2$ and $h\equiv1$, we further prove that the energy converges uniformly to $4π$ as $λ\to0$ when $Ω$ is convex. We also develop a boundary exclusion argument for positive weights that rules out boundary blow-up for $1<p<2$ without a priori energy bound.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37177 ↗

In this work, we study quantitative homogenization for a class of nonlocal equations with periodically oscillating Hölder continuous coefficients. For Dirichlet problems in bounded $C^{1,α}$ domains, we prove $L^2$ convergence rates that are expected to be optimal. In addition, we establish interior Lipschitz estimates and optimal boundary Hölder estimates for weak solutions, with constants independent of the period of oscillation.

comments
74 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37209 ↗

Let $N \geq 2$, $s\in (0,1)$, and $p \in (1, \infty)$ with $N>sp$. For the fractional $p$-Laplacian, the power function $|x|^{-γ}$ with $0<γ<\frac{N}{p-1}$ satisfies $$(-Δ_p)^s \big(|x|^{-γ}\big)=Φ(γ)|x|^{-γ(p-1)-sp} \, \text{ in }\mathbb{R}^N\setminus\{0\}.$$ Starting from the power function formula of Del Pezzo and Quaas (2026), we derive a paired integral representation of $Φ$, and show that $Φ$ is strictly increasing and then strictly decreasing, with a unique maximum at $\frac{N-sp}{p}$. This maximum $\barμ:=Φ(\frac{N-sp}{p})$ equals one half of the sharp fractional Hardy constant. Consequently, for every $0<μ<\barμ$, the equation $Φ(γ)=μ$ has exactly two roots $0<γ_{-}(μ)<\frac{N-sp}{p}<γ_{+}(μ)<\frac{N-sp}{p-1}$. These roots determine the homogeneous solutions $|x|^{-γ_{-}(μ)}$ and $|x|^{-γ_{+}(μ)}$ of the Hardy equation \begin{equation*} (-Δ_p)^s u - μ\frac{u^{p-1}}{|x|^{sp}}=0 \quad \text{in }\mathbb R^N\setminus\{0\}, \; u> 0 \text{ in } \mathbb{R}^N \setminus \{0\}. \end{equation*} For $0<μ<\barμ$, we then study the following critical Hardy-Sobolev equation \begin{align}\tag{$\mathcal{P}$}\label{P} (-Δ_p)^s u-μ\frac{u^{p-1}}{|x|^{sp}} = u^{\frac{Np}{N-sp}-1} \text{ in } \mathbb{R}^N, \; u > 0 \text{ in } \mathbb{R}^N. \end{align} Let $p \geq 2$ and $U$ be a positive, radially symmetric, nonincreasing finite-energy weak solution to \eqref{P}. We prove that there exists $C>0$ for which \begin{align*} U(x) \leq C|x|^{-γ_-(μ)}\;\text{as }|x| \rightarrow 0,\quad \text{ and } \quad U(x)\le C|x|^{-γ_+(μ)} \; \text{as }|x| \rightarrow \infty. \end{align*} Further, we show that $γ_-(μ)$ and $γ_+(μ)$ are the optimal exponents describing the singularity at the origin and the decay at infinity respectively.

comments
52 pages. Comments are welcome!
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37212 ↗

For a class of two-dimensional Euler–Poisson–Korteweg systems of quantum hydrodynamic type, we establish global existence and scattering for small, smooth, spatially localized perturbations of a constant equilibrium with irrotational velocity. This extends the three-dimensional well-posedness result of Song [Math. Ann. 392 (2025), 2157–2224] to two dimensions. Our proof follows the space-time resonance approach developed by Deng, Ionescu, Pausader, and Pusateri for the three-dimensional gravity-capillary water-wave system. The new ingredients are twofold. First, to allow non-neutral initial data, we exploit the absence of space resonances and of time resonances—including those arising after the second and third iterations—and the improved approximate-finite-speed-of-propagation bounds. Second, in the energy estimates, we adapt the almost orthogonality argument used in the works of Ionescu–Pausader and Guo–Ionescu–Pausader to the regime in which the input and output frequencies are all comparable to $1$. In this regime, time resonances may occur and the relevant curvature quantity may vanish.

comments
78 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37220 ↗

We construct smooth, compactly supported stationary Euler flows in two and three dimensions. The three-dimensional flows are axisymmetric with swirl, and their pressure varies along some streamlines. The planar flows are not locally radial. Both constructions start from simple radial profiles, which we perturb by solving scalar elliptic equations on fixed domains. We choose these equations so that their solutions can be exactly constant on open sets and meet these constant regions smoothly. Comparison arguments produce the constant regions, while an a priori regularity theorem ensures that all derivatives vanish where a solution meets them. This gives compact support and smoothness without having to describe the boundary of the support.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37251 ↗

The present paper is devoted to exhibit solutions of the Riemann problem for the Self-Organized Hydrodynamics model. The 2 x 2 system under consideration is strictly hyperbolic. With ambiguous non-conservative products, we adopt a non usual Generalized Rankine-Hugoniot jump condition to suitably define the shock wave solutions. This jump condition is obtained from the information given by the travelling wave solutions of the viscous extension of the system that involves the diffusion terms arising from the physics of the problem. We are then able to determine the simple waves of the non-conservative system studied here and we are able to construct some solutions of the Riemann problem of interest.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37252 ↗

We study the asymptotic behavior as \(d\to0^+\) of the principal eigenvalue \(λ(d)\) of the one-dimensional elliptic problem \[ \begin{cases} -dφ''(x)-2αm'(x)φ'(x)+c(x)φ(x) =λ(d)φ(x), & x\in(a,b),\\ φ'(a)=φ'(b)=0, \end{cases} \] where \(α>0\) is fixed, \(m\in C^{1,1}([a,b])\), and \(c\in C([a,b])\). We assume that \(m'=0\) only at \(x_0\in(a,b)\), where \(m\) attains a strict minimum. Within this class, we construct a counterexample $m\in C^{1,1}$ for which \(λ(d)\) has two distinct subsequential limits as \(d\to0^+\), leading to the non-convergence of $λ(d)$. We then consider another class of $C^{1,1}$ advection functions for which $m'(x)/(x-x_0)$ has positive one-sided limits at $x_0$. For this class, we prove that $λ(d)$ converges as $d\to0^+$. Thus, \(C^{1,1}\) regularity alone does not ensure convergence of the principal eigenvalue, even when \(m\) has exactly one strict interior critical minimal point.

comments
22 pages, 1 figure
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37259 ↗

We establish an infinite-dimensional sewing lemma for discontinuous paths on a scale of Banach spaces. This allows us to construct Young and rough convolutions with jumps. Based on this we analyze the local and global well-posedness of semilinear parabolic rough partial differential equations (PDEs) with jumps in mixed space-time regularity spaces.

fields
Analysis of PDEs, Dynamical Systems, Probability
pdf ↗ page arXiv:2610.37401 ↗

For the lattice circle $P_n=\{z\in\mathbb{Z}^2:|z|^2=n\}$, we prove that the threefold additive energy $J_3(P_n)\le C|P_n|^3$, with an absolute constant $C$. This establishes the optimal bound suggested by Krishnapur–Kurlberg–Wigman and Bombieri–Bourgain. We also prove a uniform weak $L^6$ bound for $L^2$-normalized eigenfunctions on the square torus, and hence uniform $L^p$ bounds for every $1\le p<6$. This extends the classical $L^4$ bounds of Cooke and Zygmund and establishes Bourgain's conjecture in this range. Applications include mean-square convergence for $L^{6/5}$ observables and a logarithm-free remainder in the nodal length variance of arithmetic random waves.

fields
Analysis of PDEs, Combinatorics, Number Theory, Spectral Theory
pdf ↗ page arXiv:2610.37423 ↗

The first eigenvalue of the $p$-bilaplacian over a domain $Ω$ is the inverse of the Poincaré constant for $W_0^{2,p}(Ω)$. In dimension two, we prove that it is minimised by disks under area constraint, thereby obtaining a sharp uniform Poincaré inequality.

comments
9 pages, 1 figure
fields
Analysis of PDEs, Spectral Theory
pdf ↗ page arXiv:2610.37433 ↗

For a convex surface tension $\varphi$ and the wall-energy density $σ(n)=-λ\,\varphi(-n)$, we prove sharp capillary inequalities outside arbitrary convex planar obstacles for $-1<λ\leq0$ and every finite-perimeter competitor of finite positive volume, including non-even and crystalline tensions. Equality holds precisely for translated dilates of Wulff caps in minimum-volume directions, with the entire truncating facet resting on a flat supporting face. A uniform quantitative estimate controls component concentration, shape up to translation, and directional cap-volume excess. The proof combines exact filling, planar completion, and quantitative Wulff stability. For the transport $DΦ$, $Φ=\varphi^2/2$, the zero-adhesion boundary-image estimate characterizes centered ellipsoids among even anisotropies with $Φ\in C^3$ away from the origin and positive definite Hessian. Two-point slab tests with arbitrary boundary data suffice: their projection balances yield a cubic identity forcing $Φ$ to be quadratic. Facets and smooth perturbations provide counterexamples to the transported estimate, including failures at prescribed adhesion levels in $(-1,1)$. Strict finite deficits persist under Hausdorff perturbations of the Wulff body. The classification concerns an unrestricted boundary-data property, stronger than the image estimate needed only for traces of auxiliary Neumann solutions.

comments
54 pages, 6 figures
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37440 ↗

We study degree $1$ minimisers $u: (Σ,g) \to S^2$ for a Baby-Skyrme model with a Sacks-Uhlenbeck regularisation term $E^α(u) = \frac{1}{2} \int_Σ (1+\vert \nabla u \vert^2)^{α+1} -1 dv_g$ and establish convergence of degree $1$ minimisers to a simple bubble tree as $α\to 0$. We also obtain a new no-neck result for the classical Sacks-Uhlenbeck energy $E^α$.

fields
Analysis of PDEs, Mathematical Physics, Differential Geometry
pdf ↗ page arXiv:2610.37457 ↗

We investigate the structure of degree $1$ minimisers of a Baby-Skyrme energy of maps $u:Σ\to S^2$ regularised by a Sacks-Uhlenbeck perturbation, and we establish quantitative convergence rates to a simple bubble tree consisting of a single degree $1$ bubble and a constant base map.

fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2610.37499 ↗

We prove global Hölder solvability for uniformly parabolic divergence-form equations with locally finite measure data under an interior Morrey growth condition and a capacity density condition on the spatial domain. The parabolic distance to the initial and lateral boundary is assumed bounded on the source support. For nonnegative data, we establish necessity results that demonstrate the sharpness of the growth and domain assumptions. We also derive estimates for divergence forcing, prove solvability for equations with lower-order terms, and obtain convergence rates in periodic homogenization.

fields
Analysis of PDEs, Probability
pdf ↗ page arXiv:2610.37528 ↗

The main purpose of this paper is to prove local Hölder continuity for weak solutions to quasilinear parabolic equations of $p$-Laplace type under finite integrability assumptions on the forcing, covering the supercritical singular range and the entire degenerate regime. The key ingredient is an $L^\infty$-valued temporal Hölder estimate of the solution after any fixed positive waiting time, which is achieved by coupling a time-rescaling argument with De Giorgi iteration. We also provide counterexamples involving forcing terms with weaker spatial integrability to show that even bounded continuous solutions with compatible data may possess no positive temporal Hölder exponent. In particular, these examples establish the sharpness of the spatial integrability threshold in the degenerate case. Finally, a separate counterexample is constructed to demonstrate that the temporal Lipschitz endpoint can fail for bounded continuous forcing when it depends on time.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37538 ↗

We give a complete list of the possible local shapes of soap films where they wet the whole wire. In rigorous terms, given a straight line $\ell \subset \mathbb R^3$, we study $2$-dimensional cones in $\mathbb R^3$ which contain $\ell$ and are Almgren minimizing with respect to $\ell$.

fields
Analysis of PDEs, Mathematical Physics, Differential Geometry
pdf ↗ page arXiv:2610.37559 ↗

Given a straight line $\ell \subset \mathbb R^3$, we classify $2$-dimensional cones in $\mathbb R^3$ which contain $\ell$ and are Almgren minimizing with respect to $\ell$. The main idea of this paper is due to an LLM, and the paper was written by an LLM, under the guidance of the authors.

fields
Analysis of PDEs, Mathematical Physics, Differential Geometry
pdf ↗ page arXiv:2610.37560 ↗

We consider the broad class of scalar hyperbolic conservation laws with generic non-convex fluxes, which may thus have an arbitrary finite number of inflection points. We investigate the properties of classical and nonclassical shock waves realized as traveling-wave limits of a linear diffusive-dispersive regularization and provide a phase-plane classification beyond the case of a single inflection point. Within the speed range of the generalized Maxwell construction, our main result classifies the left outgoing profiles by their limiting equilibria. In an elementary bowtie configuration, we prove the existence of nonclassical shock profiles with arbitrarily many oscillations. For nested configurations, we order the saddle connections on each entry interval, taking account of inaccessible saddles. We show that the number of oscillations present in the non-classical shock depends on the diffusion-dispersion ratio, and tends to infinity along the elementary families of saddle connections as this ratio tends to zero. Our construction is based on a generalization of Maxwell's equal area rule, which provides us with an invariant region in the phase space for the study of diffusive-dispersive traveling waves.

comments
38 pages, 7 sections, 12 figures
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37623 ↗

We construct two distinct solutions of the two-dimensional incompressible Euler equation with the same initial vorticity in $\bigcap_{1\le p<\infty}L^p(\mathbb R^2)$. The vorticities are compactly supported in space and have mean zero, so the corresponding velocities have finite energy. The nonuniqueness mechanism is a cascade of transitions in quadrupole dynamics across infinitely many scales.

comments
86 pages, 1 figure, Comments welcome, Lean formalisation coming soon at https://web.math.princeton.edu/~hc9325/
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.37653 ↗

Combinatorics43

The average eccentricity $\avec(D)$ of a strong digraph $D$ is the mean eccentricity of all vertices in $D$. A tournament $T$ is an orientation of a complete graph. In this paper, we prove that the average eccentricity satisfies $2 \le \avec(T) < \frac{n}{2k} + \frac{11}{2} + \frac{k}{8n}$ for a $k$-connected tournament $T$ of order $n$. We further examine the case for strong tournaments where $k = 1$, and show that $\avec(T) \le \frac{n+11}{2} + \frac{1}{8n}$ for any strong tournament $T$. Moreover, we show that the lower bound $2$ is sharp and that the upper bounds are best possible apart from an additive constant.

fields
Combinatorics
pdf ↗ page arXiv:2610.35970 ↗

Let $p>q>5$ be primes. We determine the socles and corresponding actions of quasiprimitive imprimitive groups of degree $2pq$, all of which are almost simple. Using this classification, we prove that a connected graph of order $2pq$ admitting a quasiprimitive automorphism group contains a Hamilton cycle, except possibly for three specified families of actions with socle $\PSL(2,p)$ or $\PSL(2,q^2)$.

fields
Combinatorics, Group Theory
pdf ↗ page arXiv:2610.35976 ↗

Pseudocomplemented lattices satisfy a lot of nice properties and were studied by a number of algebraists. However, even when a poset P with 0 is not pseudocomplemented, we can assign to every element a of P the set $a^p$ of all maximal elements being orthogonal to a. This assignment $^p$ is a non-trivial operator provided P satisfies the so-called Ascending Chain Condition (ACC). In particular, we investigate this operator when the poset P in question is a semilattice or a lattice. We characterize this operator and describe its properties that turn out to be similar to the corresponding properties of pseudocomplementation. Moreover, we introduce an equivalence relation $Θ_p$ on P induced by $^p$ similarly as it was done by V. I. Glivenko for distributive pseudocomplemented lattices. Finally, we investigate for which lattices L with 0 satisfying the ACC this equivalence relation turns out to be a congruence and when the quotient lattice $L/Θ_p$ is complemented respectively modular.

fields
Combinatorics
pdf ↗ page arXiv:2610.35988 ↗

A two-terminal graph is a simple graph equipped with two distinguished vertices, called terminals. Let $T_{n,m}$ be the class consisting of all non-isomorphic two-terminal graphs on $n$ vertices and $m$ edges. Let $G$ be any two-terminal graph in $T_{n,m}$, and let $d$ be any positive integer. For every $ρ\in [0,1]$, the \emph{$d$-constrained two-terminal reliability of $G$ at $ρ$}, denoted $R_G^d(ρ)$, is the probability that $G$ has some path of length at most $d$ joining its terminals after each of its edges is independently deleted with probability $ρ$. We say $G$ is a \emph{$d$-uniformly most reliable two-terminal graph} ($d$-UMRTTG) if for each $H$ in $T_{n,m}$ and every $ρ\in [0,1]$ it holds that $R_{G}^d(ρ)\geq R_{H}^{d}(ρ)$. Previous works studied the existence of $d$-UMRTTG in $T_{n,m}$ when $d$ is greater than or equal to $n-1$, or equivalently, when the distance constraint is dropped. In this work, a characterization of all $1$-UMRTTGs and $2$-UMRTTGs is given. Several structural properties of 3-UMRTTGs are established. For every $d\geq 3$, we prove the existence of a unique $d$-UMRTTG in $T_{n,\binom{n}{2}-1}$. In contrast, for every $d\geq 4$ and every $n\geq 11$, no $d$-UMRTTG exists in $T_{n,m}$ whenever $20\leq m\leq 3n-9$ or $3n-5\leq m\leq \binom{n}{2}-2$. Additionally, for each $d\geq 5$, it is proved that no $d$-UMRTTG exists in $T_{n,m}$ when $n\geq 11$ and $m=3n-8$.

fields
Combinatorics
pdf ↗ page arXiv:2610.36120 ↗

A classical problem in spectral graph theory is to determine, among all simple graphs with a fixed number of vertices and edges, those that maximize the number of spanning trees. Such graphs are called $t$-optimal. Petingi and Rodríguez [Discr. Math. 244 (2002), 351-373] reduced the construction of infinitely many $t$-optimal graphs to the determination of almost-regular graphs with a minimum number of induced $3$-paths, known as almost-regular $ν$-min graphs. They conjectured that the class of $d$-regular $ν$-min graphs is invariant under the addition or deletion of copies of the complete graph $K_{d+1}$, which would provide a method to construct large families of $t$-optimal graphs. This conjecture was recently verified for $d\in \{1,2,3,4,5\}$ by Gong et al. [Discr. Math. 349(4), 2026]. In this work, we establish an asymptotic version of this conjecture. We prove that the class of $d$-regular $ν$-min graphs on at least $n$ vertices is invariant under the addition or deletion of copies of $K_{d+1}$ whenever $d\geq 3$ and $n\geq d^3$. In addition, we introduce a new method to construct $t$-optimal graphs based on the concept of trace-minimal graphs, originally introduced by Ábrego et al. [Linear Algebra Appl. 412 (2006), 161-221]. As a consequence, we obtain infinitely many graphs that are both trace-minimal and $t$-optimal.

fields
Combinatorics
pdf ↗ page arXiv:2610.36127 ↗

We construct a new infinite family of antipodal $2(q-1)$-covers of complete graphs $K_{q^2+1}$ of diameter $3$, where $q=2^{2e+1}\ge 8$, admitting a half-transitive action of the Suzuki simple group $Sz(q)$. We further show that each constructed cover admits an involutory automorphism interchanging the two $Sz(q)$-orbits on arcs, which yields that its full automorphism group is arc-transitive.

fields
Combinatorics, Group Theory
pdf ↗ page arXiv:2610.36165 ↗

We prove that mutation-acyclicity of quivers on four vertices is decidable. If a four-vertex quiver $Q$ is mutation-acyclic, then some sequence of mutations takes it to an acyclic quiver without any arrow multiplicity exceeding $\operatorname{max}\{\operatorname{ht}(Q),m(Q)\}$, where $\operatorname{ht}(Q)$ is the largest arrow multiplicity of $Q$ and $m(Q)$ is the Markov invariant of Fomin and Neville evaluated on the totally proper wiggle class of $Q$. A search within a computable form of this bound decides mutation-acyclicity of four-vertex quivers with $O(h^{24})$ states for input height $h\ge1$. Together with a theorem of Caldero and Keller, it also decides mutation equivalence of two four-vertex quivers up to relabeling whenever one of them is mutation-acyclic. We conjecture that $\operatorname{ht}(Q)$ alone suffices as the bound, which would reduce the search to $O(h^{6})$ states, and we verify this for all four-vertex quivers of height at most three. In the other direction, for every $n\ge4$ we construct an affine family of connected $n$-vertex quivers whose nonnegative even members are mutation-acyclic and whose positive odd members are mutation-cyclic. Consequently, unlike in rank three, no fixed finite Boolean combination of polynomial equations and strict polynomial inequalities in the arrow weights (i.e., a semialgebraic criterion) decides mutation-acyclicity of $n$-vertex quivers, even allowing polynomials that are not mutation-invariant. Finally, we correct the criterion for four-vertex quivers of height at most two stated in our earlier paper, and recover its classification counts with short certificates. The calculations in the proof of the height bound are verified by computer for all values of their parameters.

comments
30 pages
fields
Combinatorics, Representation Theory
pdf ↗ page arXiv:2610.36179 ↗

A Sidon set is a set $\mathcal{A}$ of integers with the property that if $0\neq a-b=c-d$ and $a,b,c,d\in\mathcal{A}$, then $(a,b)=(c,d)$. Our main result is an improved and explicit bound on the diameter of a Sidon set: if $\mathcal{A}$ is a Sidon set with $k\ge 1$ elements, then \[\operatorname{diam}(\mathcal{A}) \ge k^2 \left(\frac{1}{1 + 0.945/\sqrt{k}}-\frac{0.945}{\sqrt k}\right) -1;\] in particular, $\operatorname{diam}(\mathcal{A})\ge k^2-1.89 k^{3/2}$. This implies that if $\mathcal{A}\subseteq[1,N]$ is a Sidon set, then $|\mathcal{A}| \le N^{1/2}+0.945 \, N^{1/4}+0.447$. Our technique also gives an improved bound for the case of large $k$, namely $\operatorname{diam}(\mathcal{A}) \ge k^2-1.88607 k^{3/2}$ for $k\ge 483$. A $γ$-Golomb ruler is a set $\mathcal{A}$ of integers with the property that there are not $γ+1$ distinct pairs $(a_i,b_i)\in\mathcal{A}^2$ with $0\neq a_1-b_1=\cdots=a_{γ+1}-b_{γ+1}$. Our diameter bounds extend in a natural way to the case of Golomb rulers. Our techniques can also be adapted to address a number of questions on the distribution of Golomb rulers: we provide upper bounds on the two quantities $\left|a_\ell-a_1 -\frac{\ell-1}{k-1}(a_k-a_1)\right|$ and $\max_{1\le i<k} (a_{i+1}-a_i)$, where $\mathcal{A}=\{a_1<\dots<a_k\}$ is a $γ$-Golomb ruler with $\operatorname{diam}(\mathcal{A})\le (k^2-bk^{3/2})/γ$ (where $b\in \mathbb{R}$ is a free parameter).

comments
33 pages, 9 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.36207 ↗

Determining the Turán densities $π(F)$, uniform Turán densities $π_{\therefore}(F)$, and codegree Turán densities $γ(F)$ of $3$-graphs $F$ is a notoriously difficult problem in combinatorics. A landmark result by Erdős characterized the $3$-graphs whose Turán density is zero. Half a century later, Reiher, Rödl, and Schacht characterized the $3$-graphs whose uniform Turán density is zero. Recently, Ding, Lamaison, Liu, Wang, and Yang suggested a possible characterization of $3$-graphs whose codegree Turán density is zero, proved one direction of it, and conjectured the other. In this work we confirm this conjecture, completing the characterization of $3$-graphs with codegree Turán density zero. A particular difficulty of this result is that it is known that there are $3$-graphs with arbitrarily small codegree Turán density, contrasting the behavior of the normal and the uniform Turán density. Similar results were independently announced by Chen, Fang, and Yang.

fields
Combinatorics
pdf ↗ page arXiv:2610.36273 ↗

We study edge complexity, the minimum Fourier $\ell^1/\ell^2$ ratio of a graph's adjacency matrix over cyclic vertex labelings. For $K_{m,n}$, with $1\leq m\leq n$, we reduce the problem to minimizing the Fourier $\ell^1$-norm of indicators of $m$-element subsets of $\mathbb Z_{m+n}$. We prove the sharp bound $\operatorname{FR}_{\min}(K_{m,n})\geq\frac{4n^2}{(m+n)\sqrt{2mn}}$, with equality precisely when $m\mid n$, and classify all labelings attaining it. We determine all Fourier ratios and minimizing labelings for $K_{2,n}$, and all minimizing labelings for $K_{3,n}$. Put $N=n+3$. The optimal three-element sets are precisely the cosets of the subgroup of order $3$ in $\mathbb Z_N$ when $3\mid N$, and precisely the arithmetic progressions with invertible common difference otherwise. The triple classification uses polynomial minorants and additive energies. We prove that attainment of the graph-energy bound is invariant under uniform independent set blow-ups. Edge complexity itself can decrease by an arbitrarily large factor under a twofold blow-up, and we give a finite example. For complete multipartite graphs, we give an exact labeling formula and sharp lower bound, and classify all part sizes and labelings attaining that bound. This bound strictly improves the graph-energy bound when the part sizes are unequal. We prove a lower bound for arbitrary simple graphs in terms of their average degree and maximum absolute degree deviation. For Laplacian projectors, we apply the equality criterion for the multiplicity bound to unbalanced complete bipartite graphs, whose top-projector minimum reduces to the same set minimization. Finally, we apply existing Fourier recovery and sampling theorems to approximate a graph from random adjacency samples under a fixed known labeling with bounded Fourier ratio, bounding the number of incorrectly recovered edges.

comments
68 pages; computational verification programs and recorded results included as ancillary material
fields
Combinatorics
pdf ↗ page arXiv:2610.36276 ↗

In 2019, Sivaraman conjectured that every $P_k$-free graph has cop number at most $k-3$ for $k\geq 5$. Chudnovsky, Norin, Seymour, and Turcotte proved $P_5$-free graphs have cop number at most $2$. For $k\geq 6$ Sivaraman's conjecture remains widely open. There is little evidence to believe Sivaraman's conjecture is optimal, making the discovery of substantially better bounds interesting, even for subclasses of $P_k$-free graphs. We consider this problem for graphs with no long holes, which generalizes $P_k$-free graphs, that also have some forbidden shorter holes. Specifically, we show that for $t\geq 13$, $(C_4,\dots, C_{t}, C_k,C_{k+1},\dots)$-free graphs have cop number at most $\lceil \frac{k-2}{\lfloor \frac{t}{4}\rfloor }\rceil+1$. Taking $C_{k-1}$, which has cop number $2$ whenever $k\geq 5$, demonstrates that this bound is tight up to a small constant factor for infinity many pairs $k$ and $t$. The proof relies on showing that in graphs without short holes, the closed neighbourhood of a short induced path can be guarded by $1$ cop, even when these paths are not geodesics. Additionally, for $t\geq 2$ we show that $(C_{t+2},C_{t+3},C_k, C_{k+1},\dots, )$-free graphs have cop number at most $(t-1)\lceil \frac{(k-2)}{t}\rceil$. This show that $(C_4,C_5,C_k,C_{k+1},\dots)$-free graphs have cop number at most $\lceil \frac{k-2}{2}\rceil$, extending work by Clow and Meger on $(P_k,\text{claw}, \text{butterfly}, C_4,C_5)$-free graphs to a substantially more general class.

comments
23 Pages
fields
Combinatorics
pdf ↗ page arXiv:2610.36290 ↗

Given a finite graph $G$, we characterize the automorphisms and automorphism groups of $G$ that can be realized as geometric symmetries and symmetry groups of straight line drawings of $G$ in the plane. In doing so, we correct a previous version of this characterization from the literature. As an application, we show that the graphs recently constructed by the first two authors to disprove the Polycirculant Conjecture, which have orders $16{,}464^k$ for each positive integer $k$, are the first known vertex-transitive graphs with no non-trivial realizable geometric symmetries. We also construct infinite families of graphs that exhibit all possible kinds of geometric symmetry.

comments
20 pages, comments welcome
fields
Combinatorics, Group Theory
pdf ↗ page arXiv:2610.36394 ↗

Fuglede's conjecture in a finite abelian group says that a nonempty subset tiles by translation if and only if it is spectral, meaning that the space of complex-valued functions on the subset has an orthogonal basis consisting of restrictions of group characters. We prove the conjecture in the eight-dimensional vector space over the field with two elements. In fact, every spectral set and every tile in this space has a tiling complement which is a linear subspace. We first prove that the Fourier support of a nonempty proper subset of density different from one half contains a two-dimensional subspace. Consequently, every spectral set of density one quarter has a subspace tiling complement. This gives a simpler proof of Kadowaki's dimension-seven result, replacing his use of the classification of Hadamard matrices of order 32 by the classical five classes of order 16. Our dimension-eight proof uses exact finite verifications involving Hadamard matrices of order 32 and perfect codes of length 15. We include the proofs of the reductions used from Kadowaki's preprint, so that neither positive result depends on his dimension-seven theorem. Finally, we give an elementary spectral example of density one eighth in dimension nine whose Fourier support contains no three-dimensional subspace. Thus, the Fourier-support argument at quarter density does not extend in this form to eighth density. Dimension nine is the only case still unresolved over the field with two elements, and this obstruction shows why the present Fourier-support method cannot resolve it as it stands.

comments
22 pages. Programs, data, certificates, and verification logs are included as ancillary files. Standalone treatment of selected results from the preliminary preprint arXiv:2610.30712, with a new, simpler dimension-seven proof
fields
Combinatorics
pdf ↗ page arXiv:2610.36417 ↗

Let $S_n$ denote the maximum length of a coil in the $n$-dimensional hypercube $Q_n$, i.e., a simple cycle whose induced subgraph has no additional chords. We combine classical square-counting identities and structural properties of maximum coils with new valid combinatorial inequalities into a parameterized linear program whose optimal value provides an upper bound for $S_n$. For $n\ge13$, we construct two analytic dual feasible solutions. Their objective values yield upper bounds on $S_n$, and the smaller value is the exact optimum of the parameterized LP. One dual solution results in the bound \[ S_n\le 2^{n-1} \frac{8n(n-1)(5n-14)} {40n^3-131n^2+4n+136-(5n-14)\lfloor3n/4\rfloor}, \] which improves upon the upper bounds in Lukito, Lukito-van Zanten, and Zémor for $13\le n\le19853$. The second yields the upper bound \[ S_n\le 2^{n-1}\left( 1-\frac1{30\sqrt{3n}}+O(n^{-1}) \right), \] improving upon the asymptotic upper bound of Zémor. Taken together, the two bounds improve upon the upper bounds for $S_n$ in Lukito, Lukito-van Zanten, and Zémor for every $n\ge13$.

comments
25 pages
fields
Combinatorics, Optimization and Control
pdf ↗ page arXiv:2610.36461 ↗

We prove a conjecture proposed by Chen that, for every fixed $n$, the sequence $(a_{n,k})_{k=1}^n$ is log-concave, where $a_{n,k}$ counts permutations of $[n]$ whose longest increasing subsequence has length $k$.

comments
15 pages
fields
Combinatorics, Probability
pdf ↗ page arXiv:2610.36483 ↗

The Ramsey number $R(G_1,G_2)$ is the smallest positive integer $n$ such that for every graph $H$ on $n$ vertices, either $H$ contains a copy of $G_1$ or the complement $\overline H$ contains a copy of $G_2$. Ramsey numbers involving the book graphs $B_n$, which consist of $n$ triangles sharing a common edge, have attracted recent attention, especially in the ``nearly diagonal'' case $R(B_{n-1},B_n)$. Rousseau and Sheehan showed the upper bound $R(B_{n-1},B_n) \le 4n-1$ nearly fifty years ago, but no general matching lower bound construction is known. This paper gives several new infinite families of $n$ for which $R(B_{n-1},B_n) = 4n-1$. Some of the main results in this paper were found with assistance from the Station, a multi-agent AI system. These results also appeared in a larger work on autonomous mathematical discovery by the authors and Du in arxiv:2608.23691. This paper is a follow-up to that work: we give complete proofs of the book Ramsey results and discuss their connections to objects in combinatorial design theory.

comments
24 pages, 1 figure
fields
Combinatorics
pdf ↗ page arXiv:2610.36522 ↗

In the theory of hyperplane arrangements, the study of logarithmic derivation modules is one of the most important subjects. If the module is free over a graded ring, the multiset of degrees of its homogeneous basis is called exponents. We utilize some basic theory of linear algebra to study the exponents of multiarrangements. As an application, we show a new exponent-preserving duality for the Coxeter multiarrangements of type $B_2$, which was conjectured by Torsten Hoge.

comments
11 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.36547 ↗

A basis of a matroid is adapted to a flat if it contains a basis of the flat. We prove that a family of flats in a rank r matroid has a common adapted basis whenever every subfamily of at most r+1 flats does. The ground set and the family may be infinite; only the rank must be finite. This result, in particular, contains Klyachko's vector space Helly theorem for subspaces. As an application, we obtain a splitting criterion for tropical toric vector bundles on projective spaces. As a secondary application, we derive quadratic bounds for the Helly number for distributive generation in the lattice of flats.

comments
16 pages
fields
Combinatorics, Algebraic Geometry
pdf ↗ page arXiv:2610.36566 ↗

A graph $G$ is degree-truncated DP-$k$-colourable (respectively, degree-truncated $k$-choosable) if it is DP-$f$-colourable (respectively, $f$-choosable), where $f(v)=\min\{k,d_G(v)\}$ for every vertex $v$. Hutchinson proved that every $2$-connected maximal outerplanar graph other than the triangle is degree-truncated $5$-choosable. Lo, Wang, Zhou and Zhu strengthened this result by proving that every $2$-connected $K_{2,4}$-minor-free graph other than cycles and complete graphs is degree-truncated DP-$5$-colourable. Outerplanar graphs are precisely the $\{K_4,K_{2,3}\}$-minor-free graphs. We study how the truncation threshold grows when $K_{2,3}$ is replaced by $K_{2,t}$, while $K_4$-minor remains excluded. A bound independent of $t$ is impossible, as shown by the family $\{K_{2,n}| n\ge 2\}$. Let $\mathcal S_t$ be the class of $2$-connected $\{K_4,K_{2,t}\}$-minor-free graphs other than cycles. Writing $χ^{\text{\sout{d}}}_{\mathrm{DP}}(\mathcal S_t)$ for the least truncation threshold valid throughout this class, we prove that, for every integer $t\ge3$, \[ \left\lfloor2\sqrt{2t}\right\rfloor \le χ^{\text{\sout{d}}}_{\mathrm{DP}}(\mathcal S_t) \le \left\lceil4\sqrt{2t}\right\rceil. \]

comments
17 pages 3 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.36586 ↗

We extend the Bakry–Émery curvature framework on graphs, broadening its applications and strengthening the resulting diameter bounds. We study diameter bounds and finiteness for amply regular graphs, distance-regular graphs, $(s,c,a,k)$-graphs, and locally strongly regular graphs. In particular, for an amply regular graph with parameters $(v,k,λ,μ)$, we prove that \[{\rm{diam}}(G)\le4+8\log\frac{k}{λ+2}\] when $μ>2$ and $k\ge3λ^2/μ$, improving the previous linear bound $4k/(λ+4)$ to a logarithmic bound. We also address several related conjectures. In their monograph \emph{Distance-Regular Graphs}, Brouwer, Cohen, and Neumaier conjectured that every amply regular graph with $μ>1$, except the icosahedron, satisfies ${\rm{diam}}(G)\le2k/(λ+2)$. We confirm this bound when $μ>λ+1$ and give counterexamples to the conjecture with $μ=λ$. For amply regular graphs with $μ\geλ$ and $μ>1$, we obtain ${\rm{diam}}(G)\le 2k/μ+1$. This proves the Qiao–Park–Koolen conjecture. In 1983, Terwilliger conjectured that every $(s,c,a,k)$-graph with $c>1$ is finite. We prove this conjecture when $s>2$, $c>2$, and $(s,c)\ne(4,3)$. Moreover, we prove Weetman's finiteness conjecture for locally strongly regular graphs when the local parameter $μ>1$, and disprove his stronger conjecture that every strongly regular graph is an $\mathscr{S}$-graph. We also prove that every graph whose local graphs all have rank 3 is finite, which in particular proves a conjecture of Soicher in 1992.

fields
Combinatorics
pdf ↗ page arXiv:2610.36620 ↗

We describe two-parametric Littlewood identities for inhomogeneous stable spin Hall-Littlewood and inhomogeneous spin $q$-Whittaker symmetric functions. These identities generalize the Littlewood identity found by Warnaar (2006) for the ordinary Hall-Littlewood polynomials. However, instead of Rogers-Szegö coefficients used by Warnaar, coefficients in our identities are given by products of Al-Salam-Chihara polynomials. Our proof relies on integrable lattice models. A key step includes three new solutions to the reflection and boundary equations for the fully fused six-vertex model. To find these solutions we use isomorphisms between tensor products of evaluation $U_q(L\mathfrak{sl}_2)$-modules, as well as covectors stable under a $q$-Onsager coideal subalgebra.

comments
69 pages
fields
Combinatorics, Mathematical Physics, Quantum Algebra
pdf ↗ page arXiv:2610.36644 ↗

We improve the upper bound on the classical Ramsey number $R(3,3,3,3)$ from 62 to 61. The proof is based on a global incidence count, classifications of small neighborhoods and exhaustive SAT computations. We formalize the mathematical reductions in Lean and refute every resulting formula with two SAT solvers.

comments
15 pages, 3 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.36648 ↗

For conjugate shapes $λ$ and $λ'$, Stanley's duality relates the Jack symmetric functions $J_λ(x;α)$ and $J_{λ'}(x;1/α)$. Through Stanley's combinatorial formula, it is an identity between the sums of the weights of the semi-standard Young tableaux giving the same monomial. We introduce the Jack multiple zeta functions, which keep these weights separate, and prove that Stanley's duality lifts to them, including the case of skew shapes. Thus the duality holds before the weights are summed up, and in fact step by step along chains of partitions. Since the duality is an identity among formal symbols, the same duality holds when the multiple zeta functions are replaced by other functions, and replacing them with the monomial quasi-symmetric functions recovers Stanley's duality. Hence our duality is a refinement of Stanley's duality which contains it as a special case. In particular, we obtain a duality for the Schur multiple zeta functions and the classical relation between the multiple zeta and zeta-star functions. We next show that any expansion of a Jack symmetric function in terms of Schur functions can be lifted to Jack multiple zeta functions, yielding an analog of a formula conjectured by Stanley and proved by Koike. We also show that the Jack multiple zeta functions of one-row shape give a one-parameter family interpolating the multiple zeta and zeta-star functions, and we discuss their values at $\{2\}^r$ and at $\{2\}^a,3,\{2\}^b$.

comments
28 pages
fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2610.36688 ↗

We prove strict pairwise negative correlation under both the uniform forest measure and the uniform connected-subgraph measure on the complete bipartite graph $K_{m,n}$ for all $m,n\geq2$. Under the connected-subgraph measure, any two distinct edges of a fixed copy of $K_{2,2}$ remain strictly negatively correlated after conditioning on the subgraph induced by the remaining vertices.

fields
Combinatorics
pdf ↗ page arXiv:2610.36749 ↗

For a given connected graph $G(V,E)$, the geodesic transversal problem finds a set $S$ of vertices of $G$ with minimum cardinality such that every maximal geodesic of $G$ contains at least one vertex of $S$. The cardinality of such a smallest set $S$ is called the gt-number of $G$ and is denoted as $gt(G)$. In this paper, we introduce the \textit{Edge geodesic transversal problem} as the edge-based extension of the geodesic transversal problem. The problem of determining a set $S\subseteq E(G)$ with minimum number of elements such that every maximal geodesic $P$ in $G$ traverses at least one edge of $S$ is called the edge geodesic transversal problem, and the cardinality of such a set is called the edge geodesic transversal number or the $gt_e$-number of $G$, denoted as $gt_e(G)$. Further, we investigate the effect of an edge cut of a graph on the edge geodesic transversal problem and prove a novel result called the edge cut transversal theorem which is used in the computation of edge geodesic transversal number of hypercube, stacked prism and ladder graph.

fields
Combinatorics
pdf ↗ page arXiv:2610.36821 ↗

The energy of a finite graph is the sum of the absolute values of the eigenvalues of its adjacency matrix. Let n be p to the power a times q to the power b, where p and q are distinct primes at least five and a and b, both positive integers, have the same parity. Among all nonempty sets D of proper divisors of n, the energy of the corresponding integral circulant graph is maximised by exactly two divisor sets: the divisors p to the power i times q to the power j with i plus j odd, and the divisors with i plus j even after removing p to the power a times q to the power b. We also determine the exact maximum energy. For the proof, weighted prime-power Ramanujan transforms yield an exact tensor-product representation of the graph energy, reducing the extremal problem to a weighted cut separating two prescribed vertices. After a sign conjugation, the relevant transforms decompose into embedded path blocks; the nontrivial blocks correspond to a weighted path together with chords of length three of weight negative one. We prove a sharp lower bound for cuts separating the two prescribed vertices in Kronecker products of these path blocks. For products of two such nontrivial blocks, equality holds only for cuts that isolate one of the two prescribed vertices. Combined with the energy-to-cut correspondence, these equality cases yield exactly the two maximising divisor sets.

comments
22 pages, 0 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.36883 ↗

We study a new variant of the classical linkage problem for digraphs, which we call {\bf unilateral linkages.} A digraph $D$ is {\bf unilaterally $\bm{k}$-linked} if it satisfies that for every choice of $2k$ distinct vertices $s_1,s_2,\ldots{},s_k,t_1,t_2,\ldots{},t_k$ there are vertex disjoint directed paths $P_1,\ldots{},P_k$ in $D$ such that $P_i$ is either an $(s_i,t_i)$-path or a $(t_i,s_i)$-path for each $i\in [k]$. This is a relaxation of the property of $D$ being $k$-linked, where we insist that each $P_i$ must go from $s_i$ to $t_i$. Clearly every $k$-linked digraph is also unilaterally $k$-linked. It is easy to see that the unilateral linkage problem is NP-complete for general digraphs and that it is trivial for semicomplete digraphs (digraphs with no pair of non-adjacent vertices). We show how Thomassen's construction of highly connected digraphs which are not 2-linked can be modified to show that there is no connectivity which guarantees that a digraph is unilaterally 2-linked. We show that every 4-strong split digraph and every 5-strong semicomplete multipartite digraph is unilaterally 2-linked. Both bounds are best possible. A digraph is {\bf co-bipartite} if its vertex set can be covered by two induced semicomplete digraphs. We prove that every 6-strong co-bipartite digraph is unilaterally 2-linked. Here we do not know whether the correct answer is 5. We also prove that every 16-strong digraph with independence number at most 2 is unilaterally 2-linked.

fields
Combinatorics
pdf ↗ page arXiv:2610.36895 ↗

Let $G$ be a simple graph of order $n\ge2$. Let $e(G)$ and $ω(G)$ denote its number of edges and clique number, respectively. Let $λ_1(G)\ge\cdots\geλ_n(G)$ be its adjacency eigenvalues. In 2007, Bollobás and Nikiforov conjectured that every graph $G\neq K_n$ satisfies $λ_1^2(G)+λ_2^2(G)\le2\left(1-\frac1{ω(G)}\right)e(G).$ This conjecture was recently completely resolved by Coutinho, Liu, Spier, Tang and Zhang. For every edge $e\in E(G)$, let $c_G(e)$ denote the maximum order of a clique containing $e$. Kannan, Kumar and Pragada proposed the edge-localized conjecture $λ_1^2(G)+λ_2^2(G)\le2\sum_{e\in E(G)}\left(1-\frac1{c_G(e)}\right)$ for every graph $G\neq K_n$. This conjecture was subsequently proved by Shaik. In this paper, we provide complete structural characterizations of the equality cases in both the Bollobás-Nikiforov inequality and its edge-localized version.

fields
Combinatorics
pdf ↗ page arXiv:2610.36912 ↗

We classify the Schubert polynomials that are products of key polynomials and explicitly determine their finest key factorization. Their indexing permutations are precisely those avoiding ten classical patterns of length six, which form the minimal pattern basis. More generally, the connected components of the column interaction graph give the unique finest factorization of a dual flagged Weyl character into nonconstant diagram characters. Every other such factorization is obtained by grouping these components. We also prove Makhija's conjecture on products of elementary symmetric polynomials in initial alphabets, allowing repeated alphabets. In this case, the elementary factors are uniquely determined, up to order, by the nonempty Rothe columns.

comments
16 pages, 5 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.36993 ↗

Let $Γ$ be a distance-regular graph with valency $k$, diameter $D\ge3$, and eigenvalues $k=θ_0>θ_1>\cdots>θ_D$. Put $b\coloneqq b_1/(θ_1+1)$, and let $1=u_0>u_1>\cdots>u_D$ be the standard sequence associated with $θ_1$. Lv and Koolen asked whether the negative of the smallest eigenvalue of a distance-regular graph can be bounded above in terms of its diameter $D$ and the parameter $b$ alone. In this paper, we answer this question affirmatively by proving that $-θ_D\le\prod_{i=2}^D(1-u_i)/(u_1-u_i) <(b+1)^{D-1}$. Equality in the product bound holds if and only if $Γ$ is $Q$-polynomial with respect to the primitive idempotent corresponding to $θ_1$. For every integer $D\ge4$, we further establish the explicit bound $-θ_D\le b^{D-1}+b^{D-2} +\frac{27}{2}(D-2)b^{D-3}$ whenever $b\ge\max\{8,\sqrt{12(D-2)}\}$. Both leading coefficients are best possible for each fixed $D\ge4$. The proof uses a spectral product theorem of Micchelli and Willoughby applied to Schur multiplication in the Bose–Mesner algebra.

fields
Combinatorics
pdf ↗ page arXiv:2610.37010 ↗

Canonical Ramsey theory studies structured colour patterns in colourings with arbitrarily many colours. Lefmann's canonical Rado theorem extends the canonical van der Waerden theorem of Erdős and Graham to partition-regular linear systems. Recently, Alvarado, Kohayakawa, Morris, Mota and Ortega proved a sparse canonical van der Waerden theorem for random set $[n]_p$. In this paper, we prove a sparse canonical Rado theorem. For every irredundant partition-regular matrix $A$ of positive rank with $k\ge3$ columns, and every prescribed partition witnessing the columns condition, if $p\ge Cn^{-1/m(A)}$, then with probability at least $1-e^{-cnp}$ every colouring of $[n]_p$ contains $$Ω\big(p^k n^{k-\rank(A)}\big)$$ canonical solutions of $A\mathbf x=0$ with distinct coordinates. A matching 0-statement holds when the prescribed partition has one or at least three blocks; the Schur equation shows that it can fail for two-block patterns. For every prescribed partition, we also establish a robust threshold at the same scale under deletion of a fixed positive proportion of the random set.

fields
Combinatorics
pdf ↗ page arXiv:2610.37053 ↗

Let $C_n$ be a cyclic group of order $n\ge 2$, and let $S=g_1\cdot\ldots\cdot g_{\ell}$ be a sequence over $C_n$, with repetitions allowed. We denote by $\mathsf N_0^n(S)$ the number of subsets $I\subseteq[1,\ell]$ with $|I|=n$ and $\sum_{i\in I}g_i=0$. A conjecture proposed by Bialostocki and Lotspeich in 1992 states that $$\mathsf N_0^n(S)\ge \binom{\lfloor |S|/2\rfloor}{n}+ \binom{\lceil |S|/2\rceil}{n}.$$ Several partial results toward this conjecture are known. In this paper, we prove this conjecture completely.

fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2610.37125 ↗

Let $G$ be a finite abelian group, let $p$ be the smallest prime divisor of $|G|$, and let $c_0(G)$ be the least integer $t$ such that every regular sequence over $G$ of length at least $t$ is an additive basis. Additive bases have been extensively studied in additive combinatorics, and the exact value of $c_0(G)$ has been determined for many groups of rank at least three and some groups of rank two (in which case groups of the form $C_p\oplus C_{pn}$ turn out to be exceptional). In this paper, we continue the study of $c_0(G)$ and obtain the following results. (1) Together with known results, we prove that $c_0(G)=|G|/p+p-2$ for all abelian groups $G$ with $\text{rank}(G)\geq 3$. (2) For distinct odd primes $p<q$, we determine \[ c_0(C_p\oplus C_{pq})=pq+2p-3,\qquad c_0(C_q\oplus C_{pq})=q^2+p-2. \] (3) Let $p$ be an odd prime, let $2\leq a\leq b$, and let $n$ be a positive integer with no prime divisor smaller than $p$. Then \[ c_0(C_{p^a}\oplus C_{p^bn})=p^{a+b-1}n+p-2. \] (4) For every odd integer $n\geq 3$, we prove \[ c_0(C_3\oplus C_{3n})=3n+3. \] Therefore, $c_0(G)$ is determined except possibly for groups of the form $C_p\oplus C_{pn}$ with $p\geq 5$.

comments
21 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.37136 ↗

Let $D$ be a proper subset of $\mathbb{Z}_n$ such that $|D|$ divides $n$ and is relatively prime to $\frac{n}{|D|}$. We first establish a necessary and sufficient condition for $D$ to be an efficient dominating set of some circulant graph $\Cay(\mathbb{Z}_n,S)$. Moreover, for an inverse-closed subset $S$ of $\mathbb{Z}_n\setminus\{0\}$ such that $|S|+1$ divides $n$ and is relatively prime to $\frac{n}{|S|+1}$, we give a necessary and sufficient condition for the existence of efficient dominating sets in $\Cay(\mathbb{Z}_n,S)$. We also obtain some results about the set of all efficient dominating sets in any circulant graph. Similar results for efficient open dominating sets are also obtained. These results extend some theorems in [Discrete Math. 340 (2017) 1503–1507], [Discrete Math. 340 (2017) 1522–1527], and [Des. Codes Cryptogr. 94 (2026), 87].

comments
9 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.37167 ↗

Let $\operatorname{odd}(G)$ be the number of odd-degree vertices of a graph $G$. We prove that, if $p=o(1)$ and $np/\log n\to\infty$, then with probability tending to one the random graph $G(n,p)$ has an edge decomposition into exactly $\operatorname{odd}(G)/2$ paths. This confirms a conjecture of Glock, Kühn and Osthus from 2016. The proof combines a key lemma with cycle decompositions of random graphs.

comments
15 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.37173 ↗

Version Control Systems, like Git and Mercurial, represent the history of a project as a Directed Acyclic Graph, capturing divergences and merges throughout development. A widely adopted workflow, the feature branch workflow, imposes a particularly simple structure: a single main branch and features branches evolving in parallel without interfering with one another. This work presents an enumerative study of these Git graphs. We first establish a bijection between Git graphs and a simpler combinatorial class, which we term cyclariums, and derive their exact count at size n, namely \( \sum\_{k=2}^n \sum\_{f = 1}^{k-1} {k \brack f} { n - k - 1 \choose k - f - 1} \). Using a saddle-point method, we further show that this number is asymptotically equivalent to \[ \frac{e^{1/8}}{2} \left(\frac{n}{2e}\right)^{\frac n 2} \exp\left(\frac 1 2 \ln\frac n 2 \sqrt{\frac n 2} + \sqrt{2 n} + \frac{(\ln\frac n 2)^2}{32} \right) \left(\frac n 2\right)^{-3/8}. \] Additionally, we analyze the limit laws of key parameters. Finally, we address random generation by designing both a uniform sampler and a Boltzmann sampler for these graphs.

fields
Combinatorics
pdf ↗ page arXiv:2610.37208 ↗

Let $m(k)$ be the maximum size of an intersecting $k$-graph with covering number $k$. In the present paper, it is proved that for all $k>e^{32}$, \[ m(k)\ge k^k\exp\bigl(-50\sqrt{k\log k}\bigr). \] Together with Zakharov's upper bound \cite{Z}, this implies $m(k)=k^k\exp\bigl(-k^{1/2+o(1)}\bigr)$.

comments
20 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.37315 ↗

Using results of Leighton and Shor and of Shor and Yukich, we present a connection between $d$-dimensional bandwidth and (sharp) thresholds in $d$-dimensional random geometric graphs: namely, that the sharp threshold for a given graph property in random geometric graphs is determined by its $d$-dimensional bandwidth, and vice versa, whenever the $d$-dimensional bandwidth of this property grows sufficiently fast (at least polylogarithmically) with the order of the graphs. This connection provides a number of corollaries, such as new information about the sharp thresholds for containing different spanning subgraphs in $1$-dimensional random geometric graphs, as well as asymptotically tight bounds for the $d$-dimensional bandwidth of structures such as graphs of bounded treewidth or genus, giving an asymptotic resolution to several problems raised by Chung in 1988.

fields
Combinatorics
pdf ↗ page arXiv:2610.37408 ↗

We show that the Frobenius group $\langle x,y,z\mid x^p=y^p=z^4=1,\ xy=yx,\ zxz^{-1}=y,\ zyz^{-1}=x^{-1}\rangle$ of order $4p^2$ contains no $(4p^2,2p^2-p,p^2-p)$ difference set if $p\equiv 3\pmod 4$ is a prime with $p \geq 7$. Together with results of McFarland, Iiams and Liebler this shows that no group of order $4p^2$ with $p \equiv 3 \pmod 4$ and $p \geq 7$ contains a Hadamard difference set.

fields
Combinatorics
pdf ↗ page arXiv:2610.37417 ↗

A proper coloring of the edges of a graph is one in which every $K_{1,2}$ is totally multicolored. A B-coloring is a coloring of the edges of $G$ in which every $K_{1,2}$ and every $K_{2,2}$ is totally multicolored. In this paper, we improve two results by the authors and Gyárfás. The first shows that an outerplanar graph with maximum degree $Δ$ can be B-colored with $\max\{Δ,9\}$ colors, improving a bound of $Δ+1$ and this is best possible up to the constant 9. The second shows that a planar graph with maximum degree $Δ$ can be B-colored with $\max\{2Δ+4,22\}$, improving a bound of $2Δ+8$.

comments
21 pages, 5 figures
journal
Discrete Math. 350 (2027), no. 2, Article No. 115426
fields
Combinatorics
pdf ↗ page arXiv:2610.37497 ↗

For a graph, its $r$-packing number is the size of the largest $r$-separated set and its $r$-covering number is the size of the smallest $r$-net. We show that the $r$-covering number and $r$-packing number are continuous in the fringe topology on locally finite trees. This recovers convergence results by Meir and Moon for e.g. uniform binary trees, uniform plane trees, uniform Polyá trees and random recursive trees, and proves new results for many other tree models, such as size-conditioned Bienaymé trees, simply generated trees, $m$-ary search trees, the minimum spanning tree of the complete graph and preferential attachment trees. Moreover, our formulation provides an intuitive interpretation of the limiting constant, that also provides an easy route for explicit computations in the case of size-conditioned Bienaymé trees. Our method relies on the definition of a canonical minimum $r$-net, and a relation between covering numbers and packing numbers due to Meir and Moon.

comments
11 pages
fields
Combinatorics, Probability
pdf ↗ page arXiv:2610.37539 ↗

An ordered graph is capped if, for any four vertices $a<b<c<d$, the edges $ac$ and $bd$ imply the edge $ad$. We prove that every nonempty capped graph $G$ satisfies $χ(G)\leq\min\{3^{ω(G)-1},18ω(G)^3\}$, where $χ(G)$ and $ω(G)$ are its chromatic number and clique number. This proves a conjecture of Davies, Krawczyk, McCarty and Walczak.

comments
10 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.37578 ↗

We construct tensor products of uniform matroids which are strictly freer than their generic linear tensor products. In particular, the generic tensor product of two copies of $U_{7,9}$ is not maximal. More generally, for every $a,b\ge2$, this holds for $U_{7,a+7}$ and $U_{7,b+7}$. The construction disproves the uniform tensor-product formulation of Mason's conjecture. Its dual is a relaxation of an explicit circuit-hyperplane of a generic birigidity matroid. By the author's representability theorem, the resulting matroids are non-representable over every field. We also obtain full-rank counterexamples to the Jackson–Tanigawa hyperconnectivity conjecture in every dimension $d\ge2$.

fields
Combinatorics, Algebraic Geometry
pdf ↗ page arXiv:2610.37616 ↗

Optimization and Control37

Wind farm layout optimization becomes increasingly difficult as turbine count grows, especially when high-fidelity aerodynamic models are used. Conventional Kriging-based Bayesian optimization often suffers from low sample efficiency in high-dimensional layout spaces because sparse training data provide limited domain coverage. We develop a Graph-informed Kriging framework that embeds reusable aerodynamic interaction information into the surrogate mean function. The wind farm is represented as a directed aerodynamic graph, where pairwise turbine interactions defined in a low-dimensional relative-coordinate space are assembled into a farm-scale mean. This structured mean captures the dominant variation in annual energy production (AEP), allowing the Gaussian process to model only the residual structure and uncertainty. Numerical experiments show that Graph-informed Kriging consistently improves AEP prediction accuracy over ordinary and generic-trend Kriging under sparse-data conditions, with increasing benefits as layout dimension and wake-interaction strength grow. Embedded in Bayesian optimization, it substantially improves sample efficiency and achieves competitive solutions for the IEA Wind Task 37 benchmark with about 280 AEP evaluations. In a controlled 16-turbine case, it also outperforms ordinary and multi-fidelity Kriging under the same high-fidelity evaluation budget. The framework is further extended to a 48-turbine wind farm, demonstrating applicability to large-scale CFD-based layout optimization. These results show that Graph-informed surrogate modeling can enable more efficient use of high-fidelity aerodynamic models for reliable large-scale wind farm layout design.

fields
Optimization and Control
pdf ↗ page arXiv:2610.35975 ↗

We study full-information online linear optimisation on the unit $\ell_p$ ball in $\mathbb{R}^d$, for fixed $2\le p<\infty$ and losses of dual norm at most one. Let $Δ_T$ denote the cumulative loss of the best fixed action above each round's minimum. A follow-the-regularised-leader (FTRL) algorithm attains, on every loss sequence and without knowing $T$ or $Δ_T$, regret $O_p(1+\min\{Δ_T^{1-1/p},\, d^{1/2-1/p}\sqrt{Δ_T}\})$. Matching lower bounds show that this rate is minimax for each fixed $p$, and its constant $(2+o(1))p/\log p$ is optimal as $p\to\infty$ among guarantees uniform in $d$ and $T$. The algorithm uses a bounded radial regulariser and switches at most once to its quadratic augmentation. On some paths where follow-the-leader (FTL) has regret one, its regret grows with $T$. A switching extension keeps the sharp constant and also guarantees $\mathrm{Reg}_T\le 2\,\mathrm{Reg}_T^{\mathrm{FTL}}$. For i.i.d. losses whose mean has no zero coordinates, we compute the exact $\log T$ coefficient of the expected regret, which separates FTL's variance term from the bias of the regulariser.

comments
49 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.36097 ↗

We study the stability of the two-dimensional discrete dynamical system defined by \[ x_{i,j} = A x_{i-1,j} + B x_{i,j-1},\] where $A, B \in \mathcal{M}_n(\mathbb{C})$. Although this system, commonly known as the Fornasini-Marchesini model, has been examined for decades, several key stability results have lacked rigorous proofs. This paper establishes precise and self-contained criteria for Lyapunov, asymptotic, and neutral stability under general bounded boundary conditions. In particular, we prove that asymptotic stability is equivalent to the spectral condition $ ρ(z_1 A + z_2 B) < 1 $ for all $ |z_1| \leq 1, |z_2| \leq 1 $. We further analyze the commutative case, derive recurrence relations for anti-diagonal sums, and present new examples and counterexamples illustrating the nuances of stability. The results are complemented by sufficient conditions based on matrix norms, Lyapunov functions, and the joint spectral radius.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36173 ↗

As fossil power plants increasingly cycle to balance variable renewable generation, carbon capture plants must be designed for a wide range of part-load operating conditions, to avoid excessive capital costs. However, accounting for the variability in design optimization yields higher computational complexity, and is therefore not a current design methodology. Here, we apply two-stage stochastic optimization to identify optimal carbon capture plant designs accounting for part-load flue gas conditions. We reduce computational complexity through a data-driven approach-to-equilibrium model of the absorption and desorption processes. Operating scenarios are derived from real-world part-load operation data of coal and natural gas combined cycle power plants. Our results demonstrate that designs optimized over multiple operating scenarios substantially reduce equipment sizes, lowering capital costs by 5-9 % at the expense of higher operating costs, resulting in a moderate reduction in total annualized cost by 0.5-1.6 %. The lowest costs are achieved when targeting an average capture target across operating scenarios, which allows CO2 capture to be shifted from full-load to part-load periods to reduce peak equipment sizes without compromising overall CO2 emission reductions. These cost benefits are robust across sensitivity analyses and increase with higher shares of part-load operation. Thus, part-load variability should be considered at the design stage, particularly for plants subject to variable load.

comments
32 pages, 11 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.36232 ↗

Tensor methods offer improved complexity guarantees for nonconvex optimization, but evaluating high-order derivatives and solving the corresponding subproblems can be computationally expensive. In this paper, we develop a proximal point Newton framework that exploits third-derivative smoothness using only gradients and Hessians. Under Hessian Lipschitz continuity, a single-loop proximal point Newton method that alternates between negative-curvature directions and Newton steps finds an $(\varepsilon_g,\varepsilon_H)$-approximate second-order stationary point within $\mathcal{O}(\max\{\varepsilon_g^{-3/2},\varepsilon_H^{-3}\})$ second-order oracle evaluations. When the third derivative is also Lipschitz continuous, we introduce an accelerated proximal point Newton method and obtain a dimension-free complexity of $\widetilde{\mathcal{O}}(\max\{\varepsilon_g^{-4/3},\varepsilon_H^{-2}\})$, matching that of the third-order tensor methods up to logarithmic factors without constructing tensors or solving tensor subproblems. We further develop an adaptive inexact Newton–CG variant that requires no prior knowledge of Lipschitz constants and preserves the improved oracle complexity with high probability. Under uniformly bounded queried Hessians, this variant achieves a Hessian-vector product complexity of $\widetilde{\mathcal{O}}(\max\{\varepsilon_g^{-5/3},\varepsilon_H^{-5/2}\})$, improving upon the previous $\widetilde{\mathcal{O}}(\max\{\varepsilon_g^{-3}\varepsilon_H^{5/2},\varepsilon_H^{-7/2}\})$ result attained by existing second-order methods such as cubic regularized Newton method and Newton–CG method. Preliminary numerical experiments are encouraging.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36256 ↗

This paper introduces a novel unified family of multiblock ADMM schemes by combining Gabay duality with a graph-based splitting methodology. This framework allows us to recover existing ADMM algorithms as particular instances, as well as to generate new ones by varying the underlying network topology. We focus specifically on the case of the complete graph, a structure of significant practical interest due to its dense connectivity, for which we provide a fully explicit and directly implementable scheme. We further derive an equivalent reformulation of the proposed family in which the algorithm is expressed in terms of a reduced set of dual variables, which clarifies its connection to the classical ADMM framework. Numerical experiments are conducted to compare the performance of different graph choices, showing particularly favorable results for the complete graph.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36278 ↗

Large-scale linear programs with millions of variables and constraints strain time and memory budgets. First-order methods such as the primal-dual hybrid gradient scale to this regime because they avoid matrix factorizations, but they require many iterations. Learned accelerators can reduce these iterations, yet fewer iterations do not guarantee a shorter solve. Learning inference also takes time, and running the accelerator longer trades solver time for inference time. Existing unrolled accelerators handle this trade-off poorly: each layer processes every variable and constraint, the predicted iterates still need solver refinement, and every instance runs the same layers. We therefore ask: What neural accelerator design achieves end-to-end speedups on large-scale linear programs across sizes and target accuracies? We introduce AHEAD, an accelerator that predicts step sizes for first-order linear programming solvers while keeping their update rules. Its weight-shared predictor reads compact solver statistics and rescales the primal and dual step sizes, so each warm-start iteration stays inexpensive and a short warm-start phase hands the solver a warm start. A learned handoff policy sets the number of warm-start iterations for each instance and target accuracy, after which the solver finishes under its own termination criterion. On packing linear programs with up to $10^7$ variables, AHEAD achieves $6.5$--$11.6\times$ end-to-end speedups over Gurobi's first-order solver from a cold start, including on instances of unseen sizes and distributions. The gains extend to PageRank linear programs and hold across difficulty levels and target accuracies.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36321 ↗

The curse of dimensionality associated with solving the Hamilton-Jacobi-Bellman equation for the value function remains a key obstacle for real-time optimal feedback control of high-dimensional nonlinear dynamical systems. This is especially severe in aerospace applications such as relaminarizing unsteady turbulent flows and recovering from aerodynamic stall where nonlinear dynamics play a significant role. We introduce Balanced Active Subspaces for Optimal Control (BASOC), a model reduction framework which seeks to find directions in state space along which the value function is most sensitive, and the distribution of states exhibits maximum variance along optimal trajectories. BASOC leverages trajectory optimization and the Pontryagin minimum principle to obtain state and co-state (value function gradient) data along locally optimal trajectories for constructing state and gradient covariance matrices. Balancing these matrices provides an oblique projection for approximating the value function, which can also be used to construct reduced-order models and surrogate feedback control laws. We show that for linear-quadratic optimal control problems, BASOC becomes equivalent to a snapshot method for initial condition balancing of the optimally controlled closed-loop realization. The framework is demonstrated on a nonlinear three-dimensional system where POD and balanced truncation are known to perform poorly, and on the nonlinear complex Ginzburg-Landau equation to demonstrate feasibility for larger-scale systems.

fields
Optimization and Control, Dynamical Systems, Numerical Analysis
pdf ↗ page arXiv:2610.36342 ↗

Several dairy retailers share a refrigerated cold-chain network, each allocating products across heat-sensitivity subgroups to maximize quadratic-fractional efficiency (DEA style). This formulation measures spoilage risk per unit of logistics cost, separating technical from scale efficiency. The single-agent problem is a non-convex fractional QCQP, while the multi-retailer game forms a generalized Nash equilibrium problem (GNEP) with a shared risk-budget constraint, which is generally NP-hard. Governed by the Arrhenius spoilage law, the thermal-risk covariance inherits a rank-one-plus-diagonal structure that unlocks hidden convexity. The authors propose the HSDI (Hidden SOCP Dinkelbach Iteration) algorithm via a joint Cholesky-spectral transform. Key contributions include: 1. Establishing simultaneous diagonalization of without requiring matrix commutativity. 2. Deriving an explicit sufficient condition for exact second-order cone programming (SOCP) relaxation, alongside a closed-form suboptimality bound for inexact regimes. 3. Proving global convergence: the outer Dinkelbach loop converges Q-superlinearly (exact) or R-linearly (inexact), and the coupled variational-equilibrium (VE) iteration converges R-linearly under strong monotonicity, achieving an overall complexity while ensuring Lipschitz stability of the primal-dual pair. Calibrated on Vietnam and US dairy markets, HSDI matches the objective value of multi-start non-linear baselines without random restarts, yielding global optima under exactness. HSDI exhibits an empirical scaling exponent of 0.93 (versus 1.30 for SLSQP), offering a certified polynomial-time framework and a stable shadow price for prioritizing cold-chain investments.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36350 ↗

The power grid has been dominated by large synchronous generators and clearly separated control layers. Today, power-electronics energy conversion interfaces are rapidly being deployed calling for another look at system-level control hierarchies. This article examines how multi-time-scale modeling, feedback control, and feedback optimization can help reinterpret the traditional organically grown hierarchy of primary control, automatic generation control, and economic dispatch, and enable new approaches for control of contemporary power systems. Beginning with an overview of power system models and conventional operational practices, the article develops connections to modern control architectures that can coordinate heterogeneous resources, account for operational constraints, and exploit faster actuation without overlooking first-principles physics.

comments
Submitted for publication
fields
Optimization and Control
pdf ↗ page arXiv:2610.36357 ↗

We describe an asynchronous parallel method to calculate Lagrangian bounds for mixed-integer stochastic programming problems. As in some prior work, we use a fully corrective Frank-Wolfe method on subproblems within a decomposition method, eventually optimizing subproblems for individual scenarios over the convex hulls of their integer-feasible vertices. However, instead of using progressive hedging as our decomposition algorithm, we employ an asynchronous projective hedging method. Like our earlier work on projective hedging, our algorithm need only solve a subset of the scenario subproblems at each iteration. Here, we further exploit the capabilities of projective hedging to create a parallel implementation capable of fully asynchronous operation. As we demonstrate empirically, this capability can lead to significant efficiency improvements when the subproblems are solved by Frank-Wolfe and have highly variable solve times.

comments
26 pages, 4 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.36376 ↗

In this paper we study shape optimization problems for planar convex sets with prescribed minimal width or constant width. We prove that the Reuleaux triangle minimizes torsional rigidity among sets of constant width. We also show that the first Dirichlet-Laplace eigenvalue is maximized by the equilateral triangle under a minimal width constraint and by the Reuleaux triangle under a constant width constraint. The proofs use subsets consisting of a disk and three straight or circular caps. Reflection arguments reduce the study to equilateral configurations, which contain a polygon or a disk polygon depending on one parameter. The same cap constructions give a common proof of the known area, Cheeger and inner parallel area inequalities, and yield sharp bounds for all positive moments of the distance to the boundary. We also prove sharp inequalities for concentric disk intersections and decreasing radial integrals, and compare interaction energies for kernels close to a positive constant. This includes a nonempty interval of positive Riesz exponents. Finally, we investigate whether inclusion monotonicity and Minkowski concavity suffice to obtain these comparisons for other functionals.

fields
Optimization and Control, Metric Geometry
pdf ↗ page arXiv:2610.36382 ↗

We study a finite-state mean-field game with entropic control cost. In the large population limit, we show that its stationary Nash equilibria satisfy the detailed balance condition and their distributions take the Gibbsian form with the value function playing the role of a Hamiltonian. We apply the entropic control cost to a ferromagnetic mean-field Ising game and investigate its stationary and non-stationary Nash equilibria. We find that the stationary Nash equilibria exhibit a second-order phase transition from disordered to ordered equilibria as temperature decreases. A center-manifold reduction yields non-stationary Nash equilibrium near the critical temperature and reveals the critical slowing down in the relaxation toward the stationary Nash equilibrium.

comments
11 pages, 1 figure
fields
Optimization and Control
pdf ↗ page arXiv:2610.36518 ↗

Distribution logistics with capacity constraints and service priorities can be modeled as a variant of the \emph{TSP with Profits} (TSPP), in which not all customers need to be visited and the selection of which customers to serve is part of the problem. This paper proposes an applied variant with two adaptations: (i) a 0/1 Knapsack objective function weighted by client priority squared ($v_i \cdot p_i^2$), and (ii) a priority bias in the Nearest Neighbor heuristic for the TSP, parameterized by $α$. The approach is evaluated on a scenario with 14 customers and a 1,100~kg capacity, complemented by a comparative experimental protocol against the classical greedy heuristic ($v_i/w_i$ ordering) and an integer programming solver. Results on the base scenario show 9 customers served, 96.4\% capacity utilization, and R\$~18,392.60 revenue, with all 5 excluded customers being low-priority. A sensitivity analysis of the priority exponent justifies the quadratic choice. The limitations of the hierarchical decomposition are explicitly discussed.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36526 ↗

Derivative-free optimization (DFO) addresses problems where derivatives are unavailable or costly to obtain. This setting often arises in simulation-based and experimental applications, where evaluating the black-box objective can itself be costly. Thus, it remains a central challenge in DFO to effectively exploit accumulated observations to select informative evaluation points under a limited budget. In this light, we introduce the NextEval solver, an agent-based DFO solver, which integrates the capabilities of a large language model (LLM) to guide the sequential selection of evaluation points. Specifically, we establish a system that combines an LLM agent with a numerical environment. The environment provides tools for analyzing past evaluations and executing numerical methods, while maintaining a shared evaluation history and method-specific state to support continuation across method switches. Using observations and tool feedback, the LLM agent selects tools, configures their parameters, and allocates the evaluation budget, while numerical methods compute and evaluate candidate points. Building on this system, we also introduce NextEval Bench, a benchmark protocol for comparing agent configurations on shared DFO tasks using the same numerical tools under matched evaluation budgets and information constraints. Numerical results on low-dimensional DFO problems demonstrate that the NextEval solver outperforms the state-of-the-art DFO baselines in function-evaluation efficiency.

comments
27 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.36553 ↗

Chunking is a fundamental preprocessing step in retrieval-augmented generation (RAG), yet most existing chunkers rely primarily on heuristics. We introduce the Question–Evidence-Supervised Chunker (QESChunker), to our knowledge the first method for constructing static chunk indices from question–evidence supervision. QESChunker constructs document-specific chunk indices by optimizing over human-annotated or LLM-generated question–evidence pairs, without access to held-out test queries. We formulate non-overlapping and overlapping index construction through a unified question–evidence utility-maximization objective, turning chunking into a supervised combinatorial optimization problem. Under the non-overlap constraint, the objective decomposes over chunks and can be optimized exactly in polynomial time by dynamic programming. When overlap is allowed, the objective becomes monotone submodular; exact optimization is NP-hard, but our greedy achieves a $(1-1/\sqrt{e})$-approximation. Across two self-constructed and three public datasets, both variants outperform existing chunking methods in complete-evidence recall in nearly all evaluated settings, under the same overlap regime and a fixed hybrid retriever. These results recast chunking as supervised, budget-aware index construction for evidence retrieval.

comments
16 pages, 8 figures, 4 tables
fields
Optimization and Control
pdf ↗ page arXiv:2610.36557 ↗

We introduce the Axiomatic Learning Structures (ALS) and Axiomatic Probabilistic Structure (APS) in order to study the interactions of Machine Learning, Optimization, with research frameworks- which, we take the view, can be characterized at the core by the corresponding axioms. The ALS explores the interplays between structures that one is interested in Learning Theory such as Loss functions and the axioms of the objects and transformations involved in these functions. The APS studies the interaction between structures of interest in Probability Theory such as Gibbs distributions, equations with noise, etc. and the axioms of the corresponding objects and transformations. We also introduce in Section 6 Transformations-Axiomatic Based Learning (TABLe) on Mathematical data-sets. One feature of this approach is the ranking of transformations from a table for a certain task based on scores which reflect which transformations work better than others, similarly to when human investigators try various methods on a task. We also study structure recovery from noisy data. In particular, to exemplify the ideas, we consider loss functions which contain derivations and derivations-like transformations. As a first application we study recovery in the context of polynomials with integer coefficients using standard derivatives along with other types of derivations which appear in Arithmetic: the p-derivations which satisfy \textit{different axioms} from the usual derivatives. As a first application of the APS we study a class of noisy (partial/mixed) derivations equations. In the last sections, to provide further potential applications of this framework, we give an overview of an initial simplified Opinion Dynamics model and its (simultaneous) study of rules of updates.

comments
42 pages, 14 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.36601 ↗

We establish first- and second-order Pontryagin (a.k.a. stochastic) maximum principles for rough stochastic control problems with general nonlinear $\mathrm{Lip}^γ$-rough vector fields. This complements the dynamic programming approach of Friz-Lê-Zhang [FLZ26b], and extends the maximum principle of Horst-Zhang [HZ26] beyond affine rough vector fields. The paper concludes with applications to pathwise stochastic optimal control.

fields
Optimization and Control, Probability
pdf ↗ page arXiv:2610.36653 ↗

It is well known that microbiome sequencing data are naturally compositional and often have a complex subcomposition structure, as well as a lot of noise and some outliers. These characteristics make classical log-contrast regression difficult to apply, because it is sensitive to extreme values and cannot fully meet the linear constraints of compositional predictors. To tackle these issues, in this paper, we propose a robust log-contrast regression framework that incorporates the Huber loss function along with a Lasso penalty, which not only boosts resistance to outliers but also facilitates sparse variable selection. It is important to note that the general linear constraints, which are explicitly imposed, serve to preserve the algebraic coherence of compositional and subcompositional covariates. Under some mild assumptions, we theoretically prove that the proposed method achieves non-asymptotic guarantees in finite samples, comparable to those of Huber loss estimators, ensuring robustness. To implement the method efficiently, we develop a symmetric Gauss–Seidel-based alternating direction method of multipliers based on its dual formulation with a convergence guarantee, and it is also easy to implement in the sense that each subproblem admits a closed-form solution. To evaluate the statistical performance of the proposed method, we conduct a series of numerical experiments using simulated data and gut microbiome data, which show that the proposed method achieves better estimation accuracy, competitive prediction performance, and more reliable outlier identification than the benchmark method RobRegCC.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36701 ↗

High-frequency control problems are often studied through time discretization, but refining the timestep $δ$ increases the discrete horizon $H=T/δ$ and can make standard finite-horizon regret guarantees deteriorate even when the physical horizon $T$ is fixed. We study episodic value-based reinforcement learning for Euler discretizations of controlled diffusions. The performance is measured as the cumulative regret over $K$ episodes relative to the optimal policy of the corresponding Euler model. Under stability and regularity conditions on value function gradients, Bellman residual derivatives, and fitted critics, we prove an $\mathcal{O}(\log K)$ regret bound with constants uniform in $δ$. For dynamics with drift linear in unknown parameters and unknown constant diffusion covariance, we verify localized versions of these conditions for a constrained, model-based controlled-diffusion fitted-$Q$ algorithm (CD-FQ) and obtain high-probability $\mathcal{O}(\log K\log\log K)$ regret for fixed and finite $T$, again with constants uniform in $δ$.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36758 ↗

We prove that verifying whether an H-polytope is control invariant for a discrete-time linear system with polytopic input constraints is \textsc{coNP}-complete. This establishes a fundamental computational complexity barrier for set-theoretic control. As an immediate consequence, the exact computation of maximal control invariant subsets of H-polytopes is \textsc{coNP}-hard. These hardness results hold regardless of whether the input constraint set is given in H- or V-representation.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36773 ↗

Stowage planning for Roll-On/Roll-Off (RoRo) vessels can be represented with different mathematical formulations for efficient stowage plans: hybrid-slot, grid-based, and lane-based representations. In this work, we implement and compare three formulations equally across a diverse set of instances. We extend these three representations to handle heterogeneous cargo in terms of size and weight, and to address advanced stability requirements, such as centre of gravity limits, stress force limits, and ballast-water adjustments. We evaluate the models in three configurations: cargo placement, simple stability with centres of gravity limits and ballast adjustments, and full stability, which adds shear-force and bending-moment limits. The tests are conducted on three different vessel sizes and solved with a mixed-integer programming solver. The hybrid-slot and grid-based models both perform well finding solutions for 447 out of 486 instances across all configurations; however, the hybrid-slot model placed 8% more cargo than the grid model. By contrast, the lane model only finds solutions for 142 instances and placed 15% less cargo than the grid-based model. The solutions generated by the hybrid-slot model are validated using a commercial vessel-stability software to verify our stability estimations, showing issues in only 24 of 162 instances, which are easily avoided by tightening the stability constraints. Finally, we address the placement of hazardous and refrigerated cargo in an efficient manner for the hybrid-slot model by employing lazy constraints which is shown to solve realistic sized instances.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36879 ↗

We consider a dynamic blocking problem: a fire, starting on a bounded open set $R_0\subset\mathbb R^2$, propagates in the plane with unit speed in all directions. To block it, a barrier can be constructed in real time, at speed $λ>0$. A long standing conjecture by the first author states that the fire can be entirely surrounded by the barrier, in finite time, if and only if $λ>2$. The present paper improves on earlier lower bounds. Namely, if $λ\leq Λ=1.62621170228\ldots $ we show that the fire cannot be blocked. We also prove that, if $λ\leq 2$, the fire cannot be blocked by constructing a barrier which is a simple closed curve. More precisely, if the barrier does not contain purely delaying arcs, i.e.~arcs which are eventually surrounded by the fire on both sides, then for $λ<2$ the region burned by the fire at time $t>0$ has area $A(t)>(2-λ)t^2/4$.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36919 ↗

Global supply chains are increasingly exposed to trade policy shocks and physical disruptions arising from common geopolitical events. We develop a robust optimization framework with two decision stages for strategic supply chain reconfiguration. Supplier qualification and transportation arc activation decisions are made before uncertainty is realized, while operational flows adapt afterward. Geopolitical uncertainty is represented through country level shocks that can jointly trigger tariff escalation, policy induced restrictions on cross border capacity, facility disruptions, and physical transportation failures. The resulting min max min problem is reformulated by dualizing the operational recourse problem, yielding a nonconvex mixed integer bilinear adversarial problem solved within a column and constraint generation procedure. Comparison with complete enumeration verifies the implementation on small instances, while experiments on 60 synthetic supply chain networks of varying sizes assess computational performance. The results show that omitting geopolitical transmission channels during strategic design can lead to substantial deterioration in cost and service under full geopolitical stress. Broader protection is achieved primarily through changes in transportation network configuration rather than increased supplier qualification. Robust performance also depends on the severity and composition of geopolitical consequences, with restrictions on cross border capacity exerting a stronger effect than comparable proportional changes in tariff severity over the tested ranges. A geographically grounded case analysis further shows that the value of individual uncertainty channels depends on network structure and geographic exposure.

fields
Optimization and Control
pdf ↗ page arXiv:2610.36977 ↗

We investigate observer-based event-triggered stabilization of infinite-dimensional linear systems, such as those governed by partial differential equations. A Luenberger observer and a feedback gain are assumed to stabilize the closed loop in continuous time, and we study what happens when the control input is only updated at instants generated by a dynamic event-triggering mechanism. Unlike approaches that rely on state-dependent triggering rules, our condition only involves the observer state and the measured output, which makes it implementable when the full state is not measured. It compares the deviation of a monitored signal of the observer since the last update with a threshold made of two parts. The first one is a relative part, proportional to the current size of this signal, and the second one is a positive internal variable, which does not depend on the observer state. In particular, the case where the threshold reduces to the internal variable alone is covered. Under suitable assumptions, we establish the well-posedness of the closed-loop system and the absence of Zeno behavior for general choices of these two parts, and we prove its exponential stability under a small-gain condition on the relative part. We further analyse the existence of a dwell-time, namely a positive lower bound on the time elapsed between two consecutive triggering instants. When the sampled dynamics is bounded, as in finite-dimensional systems or systems with a finite-dimensional unstable part, we obtain a dwell-time independent of the initial data, together with exponential stability. For unbounded generators, we exhibit a system satisfying all our assumptions for which no triggering rule can achieve both a dwell-time independent of the initial data and bounded trajectories.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37069 ↗

University course timetabling is a large-scale combinatorial optimization problem involving strongly coupled temporal and resource constraints. This paper proposes a two-stage optimization framework for real-world large-scale university course timetabling, integrating orthogonalized hierarchical modeling with conflict-driven adaptive large neighborhood search (ALNS) repair. The first stage performs global course-level optimization, while the second identifies residual conflicts through a conflict graph and localizes them into connected conflict regions. The Course–Occurrence–Session representation provides the structural basis for multi-granularity repair, which employs Cluster-, Course-, Occurrence-, and Session-level destroy operators within the ALNS framework, followed by fine-grained greedy repair. The framework is evaluated on a real-world university timetabling instance involving 3,073 course groups, 8,015 scheduling events, 998 student classes, and 1,511 instructors. Ablation experiments demonstrate the effectiveness of hierarchical repair granularity and adaptive operator selection. Across ten independent runs, the ALNS + Greedy framework achieves a mean residual conflict count of 18.50, compared with 172 for deterministic Greedy-only repair under the same initial scheduling conditions, representing an 89.24\% reduction. These results support the effectiveness of the proposed conflict-driven hierarchical repair framework for large-scale university course timetabling.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37100 ↗

The Traveling Repairman Problem minimizes cumulative arrival times rather than the total route length, as in the classical Traveling Salesman Problem. The literature contains several legacy metaheuristics for the TRP; however, most are evaluated on small- and medium-scale instances and show limited scalability for large and extreme datasets. GILS-RVND, the current gold standard in metaheuristic algorithms, performs well on small and medium instances but struggles to scale due to the exhaustive neighborhood search in the RVND phase. To address this limitation, we introduce Restricted 2-Opt And Ruin-and-recreate (ROAR), a Large Neighborhood Search-based method that integrates ruin-and-recreate with greedy insertion and a restricted, insertion-anchored 2-opt operator. Computational experiments demonstrate that ROAR produces high-quality solutions for large-scale instances and outperforms existing methods under strict time limits. We further introduce an O(1)-time, O(1)-space algorithm for evaluating 2-opt moves during a contiguous systematic sweep, which, to the best of our knowledge, is unprecedented. Finally, we evaluate ROAR on instances with up to 200,000 nodes, substantially exceeding the largest instances previously reported in the literature (11,849 nodes).

comments
Accepted at RIVF 2026 20th International Conference on Computing and Communication Technologies
fields
Optimization and Control
pdf ↗ page arXiv:2610.37128 ↗

We study a scalar finite-horizon controlled diffusion with a finite action set. An action may affect the drift, the instantaneous variance, or both, and performance is assessed by a uniformly convex terminal deviation cost. The model represents systems in which decisions are made continuously in time but only finitely many operating settings are available. Under uniform nondegeneracy and an explicit shape condition on the terminal cost, we show that the value function is classical at positive times, establish global positive upper and lower bounds for its second state derivative, and prove that the ratio of its first and second state derivatives is strictly increasing. This ordering implies that each action's optimality region is connected, the active actions occur in monotone order, and adjacent regions are separated by unique switching boundaries. It also yields an exact coefficient-based screen that removes actions that cannot be optimal on any open state region. For smooth continuous-action benchmarks, both the value loss from restricting the action set and the additional loss from rounding the continuous optimal feedback to the nearest available action decrease quadratically with the action-set resolution. Numerical experiments illustrate the resulting action screens, threshold policies, and implementation losses.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37344 ↗

We study difference-of-convex optimization problems with several composite inequality constraints and develop a majorization-guided switching proximal gradient (MGSPG) method. Specifically, at each feasible outer point, quadratic majorization and linearization yield a strongly convex composite surrogate problem. To solve this surrogate approximately, we employ a switching proximal gradient (SPG) method, which takes a proximal gradient step for the objective when the current point is nearly feasible and for a most violated constraint otherwise. This construction requires neither projection onto the joint surrogate feasible set nor multiplier computation. By establishing a common invariant ball and using an increasing weight argument, we obtain an \(O(τ^{-1})\) inner complexity bound for SPG. Building on this bound, we show that, under boundedness of the feasible level set and a uniform Slater condition for the surrogates, MGSPG finds an \(\varepsilon\)-KKT point in \(O(\varepsilon^{-2})\) outer iterations and \(O(\varepsilon^{-4})\) selected component proximal gradient updates. Numerical experiments on SCAD-constrained and groupwise-constrained regression problems demonstrate the viability and computational efficiency of our method compared with existing optimization algorithms.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37362 ↗

Finding a Sudoku completion and proving that the completion is unique are different computational tasks. Despite both concerning the same puzzle, their computational effort need not be closely related. In this paper, we investigate how this relationship depends on the mathematical formulation, solver and measure of computational effort. Using 32,000 uniquely solvable puzzles and 64 harder instances, we compare integer programming formulations and CP-SAT. Further, we prove that the raw natural formulation has an extremely weak linear programming relaxation, even with uniqueness exclusions. Computational experiments show that the relationship between completion and certification effort depends on the formulation, puzzle population, solver and effort measure, rather than on the puzzle alone.

comments
43 pages, 8 figures, 32 tables, 3 appendices
fields
Optimization and Control
pdf ↗ page arXiv:2610.37411 ↗

With exact, delay-free gradients, Nesterov's accelerated gradient (NAG) achieves optimal $\sqrtκ$ dependence for $L$-smooth, $μ$-strongly convex objectives, where $κ=L/μ$; heavy-ball (HB) achieves the same scaling on strongly convex quadratics. In asynchronous and distributed settings, gradients can be inexact, with delays varying across workers and iterations. We isolate staleness using a fixed delay $τ\ge1$, replacing $\nabla f(y_k)$ with $\nabla f(y_{k-τ})$ at iteration $k$. Arjevani, Shamir, and Srebro establish a worst-case lower bound of $Ω(τ\sqrtκ\log(1/\varepsilon))$ iterations to reach objective error at most $\varepsilon$ as $\varepsilon\to0$. We show that for any fixed positive delay, neither NAG nor HB retains $\sqrtκ$-acceleration on strongly convex quadratics when updating every iteration, for any stable constant step size and positive momentum. We introduce delay-resistant accelerated momentum (DREAM), which decouples gradient-query extrapolation from momentum and uses explicit delay-dependent parameters. DREAM attains a linear convergence rate $1-Θ(1/(τ\sqrtκ))$ for all smooth strongly convex objectives as $κ\to\infty$, yielding complexity matching the delayed first-order lower bound up to logarithmic factors. Our proof combines characteristic-polynomial stability analysis on quadratics with an analytic Zames–Falb integral quadratic constraint certificate. We extend DREAM to smooth convex objectives, obtaining near-optimal iteration complexity up to logarithmic factors for sufficiently small target accuracy. Experiments on strongly convex quadratics and regularized logistic regression corroborate DREAM's accelerated condition-number dependence under delay, unlike NAG and HB, and suggest this scaling persists when parameters use a conservative bound on the true delay.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37444 ↗

In robust optimization, budgeted uncertainty sets offer a popular trade-off between modeling flexibility and computational tractability. Data that is observed in practice is often discrete in nature, which creates the need for tools to construct budgeted uncertainty sets based on observed data. This paper presents a principled approach for this purpose, which ensures that any solution found through the robust problem with respect to the constructed budgeted uncertainty set offers an approximation guarantee with respect to the optimal value of the problem with discrete data. Our approach is purely based on the geometry of data and can be applied to any nominal optimization problem. We discuss the problem of evaluating budgeted uncertainty sets, and how to construct them using exact and heuristic methods. In computational experiments, we show that it is possible to construct sets with little computation time that offer guarantees that are close to optimality.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37537 ↗

We use differential Galois theory to prove that given two arbitrary points $X$ and $Y$ on the Stiefel manifold, there is no closed-form analytic expression for the Riemannian logarithm $\log_X Y$, the geodesic between $X$ and $Y$, or the geodesic distance between $X$ and $Y$. In other words, these cannot be expressed in terms of elementary functions of the coordinates of $X$ and $Y$; in fact, our result goes further --- even if we allow for repeated integrals of elementary functions, there is still no such expression.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37568 ↗

Many real-world optimization tasks are governed not only by what should be optimized, but also by what is allowed: a solution must first satisfy physical, operational, safety, budgetary, or structural requirements before its objective value becomes meaningful. Yet benchmark design for constrained continuous optimization often treats these restrictions as fixed properties of predefined test instances. Existing generators and benchmark suites provide essential foundations for reproducible comparison, but they either focus primarily on configurable objective landscapes or expose constraint properties within limited, predefined regimes, making it difficult to ask targeted questions about constraint-induced difficulty: whether an optimizer struggles because feasible regions are small, boundaries are restrictive, variables are tightly coupled, or constraint information is poorly scaled. We introduce GNBG-C, a Generalized Numerical Benchmark Generator for Constrained Continuous Optimization. The central idea is to treat the feasible region itself as a configurable object rather than as an inseparable attachment to a fixed test function. By generating controlled constraint regimes around the known GNBG global minimizer, the generator enables systematic studies in which feasibility pressure can be varied while preserving that point as a global minimizer of the constrained problem. This allows researchers to examine how optimizers discover feasibility, handle constraint pressure, and improve objective quality under diverse and controllable feasibility conditions. GNBG-C therefore provides a flexible and auditable framework for probing the strengths and weaknesses of constrained continuous optimizers under targeted forms of feasibility-side difficulty.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37590 ↗

We study a class of linear-quadratic mean field games in which the state of each agent follows a stochastic Volterra equation of convolution type. The resulting dynamics are non-Markovian and account for memory effects through a completely monotone kernel, while the mean field interaction enters both the drift and diffusion coefficients. Exploiting a Markovian lifting of the Volterra dynamics, which restores the Markov property at the price of moving to an infinite-dimensional state space, we characterize the optimal response of a representative agent for a given population mean through a system of Riccati equations in Banach spaces, thereby obtaining a feedback representation of the optimal control. We then study the consistency condition defining the mean field equilibrium, which reduces to a forward-backward fixed point equation for the equilibrium mean: we prove existence and uniqueness of the equilibrium on short horizons and, when the control does not act on the diffusion coefficient, on arbitrary horizons under a monotonicity condition on the interaction. Finally, we consider the corresponding finite-player game and establish a quantitative propagation of chaos result. As a consequence, the feedback control of any mean field equilibrium, when implemented by each agent on its own state, yields an $\varepsilon_N$-Nash equilibrium for the $N$-player game, with an explicit rate as the number of agents tends to infinity.

comments
31 pages
fields
Optimization and Control, Probability
pdf ↗ page arXiv:2610.37622 ↗

We study nonconvex stochastic functions that admit a structured second-order representation, a generalization of lower- and upper-$C^2$ functions to normal integrands. Similarly to the deterministic case, lower-$C^2$ normal integrands are defined as the pointwise maxima over a compact random index set of functions that are twice-continuously differentiable in the decision variable with an integrable bounded Hessian. Our main result is a local value function characterization of this class of functions, namely, that they satisfy a local sufficient ascent property almost surely. Equivalently, they differ from a convex normal integrand by a random quadratic. The upper-$C^2$ theory follows by a change of sign, and delivers the corresponding lower envelopes, the form in which these integrands appear as second-stage value functions of stochastic programs with compact recourse. We also show that for this class integrability of the function values, of the local Lipschitz moduli, and of the minimum-norm Clarke subgradients are equivalent. We then study smooth integrands, finite lower envelopes, difference-of-convex integrands, Moreau and forward–backward envelopes, compositions with a smooth map, monotone convex compositions, and compact-recourse stochastic programs within these classes of nonconvex stochastic functions. As an application, we show that the optimistic loss of partial-label regression and the variational formulation of the conditional value-at-risk fit within this framework.

fields
Optimization and Control
pdf ↗ page arXiv:2610.37629 ↗

Weight decay and explicit weight-norm constraints improve optimizer performance in language model pretraining, yet a unified understanding of why such control helps and how much is needed remains incomplete. We analyze norm control in Muon over the entire training trajectory using the \emph{accumulated effective learning rate}, $C_T=\sum_{t<T}η_t/\|W_t\|_F$, where $η_t$ is the nominal learning rate and $η_t/\|W_t\|_F$ is the effective learning rate relative to the weight norm. For bounded update directions, accumulation over any training interval bounds relative weight displacement. We show that vanilla Muon can accumulate only logarithmically, potentially limiting exploration and helping explain its slower empirical progress, whereas MuonW and MuonH guarantee at least linear accumulation under the stated conditions. Going beyond bounded-norm control, we show that sufficiently slow norm growth guarantees faster accumulation than the logarithmic worst case, opening up less restrictive optimizer designs. Based on this insight, we propose MuonGS, which uses Gram–Schmidt (GS) projection to remove the radial component of each Muon update without fixing the weight norm. With bounded update directions and a constant nominal learning rate, we prove that the weight norm grows at most proportionally to $\sqrt{t}$ after $t$ steps. Consequently, MuonGS guarantees $C_T=Ω(\sqrt{T})$, with accumulation sustained even in later training. Further, for smooth scale-invariant objectives we establish its $O(T^{-1/2})$ stationarity guarantee. In pretraining experiments with 160M- and 400M-parameter dense models as well as a 135M active- and 1B total- parameter MoE model, MuonGS achieves lower validation perplexity than MuonH at long training horizons. In particular, for the dense model it achieves an estimated token-equivalent speedup up to about $15\%$ at matched validation loss to MuonH.

comments
21 pages, 9 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.37647 ↗

Probability36

We study an accelerated Langevin equation in which the classical overdamped Langevin drift is supplemented by a nonlocal gradient velocity determined through a target-weighted Poisson equation. The target $p$ is only assumed to be strictly positive and need not be normalized. Writing the dynamics in terms of $ω=\log(p/q)$ yields a coupled parabolic-elliptic system whose parabolic component contains the usual Langevin diffusion, whereas the elliptic component generates a global transport. The system is invariant under multiplication of $p$ by a positive constant. On a bounded smooth domain with homogeneous Neumann boundary conditions, we establish global existence and uniqueness of classical solutions for strictly positive $C^{2+α}$ target weights and compatible $C^{2+α}$ initial data. For long-time behavior, the oscillation of $ω$ decays at the explicit rate $e^{-t}$ and the solution density exponentially converges to target density. Moreover, an exact entropy-dissipation identity holds relative to the target density, in which the relative Fisher information and the Jeffreys divergence give the two dissipation mechanisms associated with the Langevin and weighted-Poisson components. In particular, KL divergence also converges to zero with rate $e^{-t}$ without a convexity assumption on the target.

fields
Probability, Analysis of PDEs
pdf ↗ page arXiv:2610.35892 ↗

We ask how much of a large random system must be observed before a global average can be inferred with controlled uncertainty. For the random transposition shuffle, a conjugate Bayesian estimator of the average displacement, computed from $m$ uniformly sampled cards, has relative uncertainty $1/\sqrt{2m}$, independent of the deck size $n$. The law of total variance exposes a second error, the floor $R_n$, which arises because the observed configuration is a single draw from equilibrium: no relative tolerance below $κ\sqrt{R_n}$ can be certified, even by a census. For the shuffle we derive $R_n=(2n^2+7)/\bigl(5(n-1)^2(n+1)\bigr)$ and show that the posterior remains calibrated. At the critical point of the two-dimensional Ising model the floor of the energy vanishes, whereas that of the magnetisation stays near $0.072$; its mean-field limit is $\sqrt{2}-1$. For correlated fields read by sensor networks, the usual 95\% interval covers the ensemble mean in 7\% of cases. Bounding the floor from one realisation, as in single-chain Monte Carlo diagnostics, requires an assumption on the decay of correlations.

fields
Probability
pdf ↗ page arXiv:2610.35973 ↗

In a prior paper, we investigated what could be said about the number in the pair that was not observed when we employed the BlackwellCover Guessing Strategy. In this paper, we generalize prior results to arbitrary distributions. We use a variant of the BlackwellCover strategy ro take pairs of coins chosen from a mixture of two different distributions G and H such that one of the coins is flipped, and as a result of that flip the unobserved (unflipped) coin is placed in one of two sets A and B. We show that, regardless of which distributions are used for G and H, there are situations in which this can be done so that the mean of the heads probabilities of the unobserved coins in A is less than the mean of the heads probabilities of the unobserved coins in B. What makes this result surprising, in addition to its generality, is a comparison with a corresponding result for marbles. If one knows that one will always be weighing one of two marbles with different specific weights, such as 10g and 20g, knowing the weight of the marble being weighed automatically determines the weight of the other. However, even if one knows that one will be flipping one of two coins with specific heads probabilities such as 1/3 and 2/3, the outcome of the flip of one of the coins will not immediately enable one to conclude anything about the heads probability of the unflipped coin, which is where the BlackwellCover Guessing Strategy comes into play.

comments
16 pages
fields
Probability
pdf ↗ page arXiv:2610.36016 ↗

Accelerated dynamics methods, such as parallel replica dynamics, require estimates of the decorrelation time, after which a process trapped in one of its metastable states is distributed according to the quasi-stationary distribution (QSD). In a method proposed by Binder, Lelièvre and Simpson, decorrelation times are defined by coupling a particle approximation of the non-linear conditioned probability flow of the process with MCMC-inspired diagnostics. We explore an alternative to MCMC diagnostics, based on coarse-graining of non-linear probability flows and simulation-based inference. We first develop a coarse-graining framework for killed Markov jump processes conditioned on survival, proving uniform-in-time, non-asymptotic estimates approximating the non-linear probability flow of the true process by the non-linear flow of a surrogate process, defined with a collective variable. The framework does not assume reversibility of the true process, and also gives bounds which scale explicitly in an adiabatic regime. Motivated by these results, we then construct a quasi-stationarity diagnostic using simulation-based inference: a recurrent neural network (RNN), trained on particle approximations of non-linear effective flows, estimates posterior probabilities of decorrelation. The learned diagnostic is then transferred to collective-variable traces of high-dimensional particle approximations. We investigate the behavior and robustness of this diagnostic on a set of numerical experiments, comparing it to the MCMC diagnostic of Binder et al.\ on both synthetic examples and a molecular system, in and out of its training prior distribution. When its prior assumptions hold, the RNN diagnostic outperforms the MCMC diagnostic in ROC analysis, is less conservative, fails less often, and has an interpretable decision threshold.

comments
91 pages, 18 figures, 3 tables
fields
Probability
pdf ↗ page arXiv:2610.36078 ↗

This work proves the convergence of stochastic nonlocal conservation laws, with spatially inhomogeneous noise, to stochastic conservation laws when the range of particle interaction vanishes. This provides rigorous justification for a class of stochastic conservation laws that has received considerable attention in the last decade. To handle technical challenges such as gradient-dependent noise, modified versions of the equations with specific cutoff boundaries are utilized and new uniform estimates to ensure stability during the limiting process are established. In addition, the work demonstrates that this convergence depends on the presence of a viscous term, as a counterexample shows that without it, the limiting solution may fail to be an entropy solution.

fields
Probability
pdf ↗ page arXiv:2610.36099 ↗

Regarding the Besov space ${\rm B}_1^{Φ_2}$, defined by S. Geiss and Ylinen (Memoirs AMS 1335, 2021) by a coupling method, we prove for $ξ\in{\rm B}_1^{Φ_2} \cap L_2$ a distributional estimate in terms of the Lorentz-Zygmund space $L_{1,\infty}(\ln L)^{-1}$ for the Malliavin derivative ${\rm D}(T_uξ)$ uniformly in $u>0$, where $(T_u)_{u\ge }$ is the Ornstein-Uhlenbeck semi-group. Here $ξ$ is not necessarily Malliavin differentiable in the ${\rm D}_{1,1}$-sense. This provides for Theorem 4.22 of the aforementioned reference an extension to the borderline case $p=1$.

fields
Probability, Functional Analysis
pdf ↗ page arXiv:2610.36152 ↗

We prove a functional central limit theorem for trace statistics of Grover walks on random $N$-lifts of a fixed finite connected simple $d$-regular graph, $d\geq3$. All observations are computed from the same random lift. The squared centred traces, divided by their observation times, have averages converging in probability to a degree-dependent constant. Their partial sums, centred at this constant and divided by the square root of the window length, converge to Brownian motion. For a window of length $M$ beginning at $t\geq1$, it suffices that $M\to\infty$ and $R^4(d-1)^R/N\to0$, where $R=t+M$. The initial time may remain fixed, with no further restriction on $M/t$. Convergence holds jointly with fixed low-degree polynomial trace statistics; their limiting vector is independent of the Brownian motion. For non-bipartite bases, we also identify Gaussian trace limits at fixed time offsets. Under common initial-time normalisation, subtracting a deterministic growth profile gives an explicit Brownian time change when $M/t$ has a finite limit. Bipartite bases retain the same averaged constant and have twice the Brownian variance rate. The proof combines a growing-dimensional Poisson approximation, a uniform bilinear Chebyshev-energy estimate, a time-decaying Poisson–Gaussian coupling, and a Gaussian quadratic-array theorem. Separate moment estimates give expectation-centred versions under an additional sufficient condition.

comments
51 pages, 1 table
fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2610.36320 ↗

We study spectral limits of random operators whose incident edge weights are coupled by local normalization, and the reflection quantum walks associated with them. At each vertex of a $d$-regular graph, an independent Dirichlet probability vector distributes unit mass among outgoing directions. At fixed degree, the empirical spectral measures of the associated self-adjoint operators converge almost surely on locally tree-like graph sequences. We give a closed-walk formula for every moment of the limiting tree law. Our main result identifies two distinct high-degree scales. After rescaling the deterministic expected root spectral law by $\sqrt d$, a weak limit exists if and only if $dα_d^2$ converges in $[0,\infty]$, where $α_d$ is the Dirichlet parameter. The limits range from a point mass at zero through non-semicircular critical laws to the semicircle law. The critical laws admit a Poisson-tree representation with an explicit Bessel intensity. The extended family, including both endpoint laws, provides a first-Wasserstein approximation uniform over positive Dirichlet parameters. In the regime $dα_d^2\to0$, the rescaled law converges weakly to a point mass although its second moment remains one. Weighting spectral mass by its contribution to this moment and rescaling again reveals a different family of limits, classified by $dα_d$. These limits have explicit Beta-product descriptions. A quantitative comparison transfers the variance-scale classification to the centered and rescaled inherited eigenphases of the quantum walk.

comments
47 pages
fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2610.36326 ↗

Classical Langevin Monte Carlo algorithms, such as the unadjusted Langevin algorithm (ULA), are based on discretization schemes of the overdamped Langevin dynamics and are widely used for sampling in machine learning because they scale to high dimensions and large datasets. However, they require differentiable log-densities, and are less efficient for sampling heavy-tailed distributions. Although various approaches have been proposed in the literature for sampling non-differentiable and heavy-tailed targets, all of them are developed in the overdamped setting only. We propose a novel momentum-based underdamped anchored Langevin dynamics (UALD), and underdamped anchored Langevin Monte Carlo (UALMC) algorithm based on a discretization scheme of UALD. Our paper is the first to provide a kinetic (momentum-based) Langevin algorithm for non-differentiable and heavy-tailed targets that is free of asymptotic bias. The method replaces the original potential with a smooth reference potential and modifies the underdamped Langevin dynamics via multiplicative scaling in the drift term. We establish non-asymptotic guarantees in the 2-Wasserstein distance to the target distribution and show that the resulting iteration complexity has better dependence on dimension and accuracy level compared to overdamped-based anchored Langevin algorithms. We provide numerical experiments to illustrate the theory and practical performance of our proposed approach.

comments
66 pages
fields
Probability
pdf ↗ page arXiv:2610.36470 ↗

This paper introduces a model for infinite-server queues driven by a Cox process with stochastic intensity given by a squared Bessel (BESQ) process. Exploiting the affine structure of the underlying dynamics, we derive a closed-form expression for the joint moment-generating function (MGF) of the arrival intensity, the number of customers in the system, and the integral of the number of customers in the system over a time interval. This MGF can also be obtained at two (or more) dates, thereby allowing an analysis of the time evolution of infinite-server queues driven by a Cox–BESQ process. The availability of these explicit MGFs enables the development of risk measures for these variables, such as the joint moments of the arrival intensity, the number of customers in the system, and the integral of the number of customers in the system over a time interval, conditional on one of these variables exceeding a critical threshold. These risk measures can be determined at one or two dates. They rely essentially on the explicit form of the MGF of the state variables and on Fourier representations of the risk measures. Numerical experiments demonstrate the practical implementation of the proposed methodology and highlight the relevance of the theoretical results.

comments
27 pages, 6 figures
fields
Probability
pdf ↗ page arXiv:2610.36473 ↗

In this paper, we prove central limit theorems for the quadratic additive functionals of the three-dimensional and four-dimensional voter models. Our main results show that, under proper time-space scalings, the centered quadratic additive functional of the four-dimensional voter model converges weakly to a Brownian motion and that in the three-dimensional case converges weakly to a Gaussian process without independent increment. The proof of the main result in the four-dimensional case extends the martingale decomposition method. In the three-dimensional case, we introduce a quadratic fluctuation limit theorem which shows that the quadratic fluctuation density field of the voter model converges weakly to a generalized Ornstein-Uhlenbeck process. Then our main result in the three-dimensional case follows from an argument where the quadratic additive functional is approximated by an integral of a test function with respect to the quadratic fluctuation density field. The duality relationship between the voter model and the coalescing random walk is repeatedly utilized in proofs of the main results. The proof strategies introduced in this paper are applicable in deducing CLTs for general additive functionals of three-dimensional and four dimensional voter models. We only deal with those with the quadratic forms to make our exposition easy to catch.

comments
45 pages
fields
Probability
pdf ↗ page arXiv:2610.36594 ↗

Token systems support repeated service exchange without monetary transfers, but their viability depends on the stability of token balances. We study token systems with compatibility constraints among agents. A requester is randomly sampled in each period, and only a few randomly sampled agents are available to provide service. We establish a necessary and sufficient condition for stability and show that the minimum-token rule, which selects the available provider with the lowest token balance, achieves the optimal stability region. We then apply this stability criterion to a random token system generated from stochastic block models with uniformly sampled requesters. We tightly characterize the stability regime by identifying two obstructions to stability: insufficient service opportunities for a group of communities or for an individual agent.

fields
Probability
pdf ↗ page arXiv:2610.36744 ↗

Brownian motion and the Brownian bridge have covariance kernels that are Green's functions of the negative second derivative on the unit interval, with mixed Dirichlet-Neumann and Dirichlet boundary conditions, respectively. We characterize the Green's functions of all strictly positive real self-adjoint realizations of this operator. The kernels are in one-to-one correspondence with positive-semidefinite $2\times2$ endpoint covariance matrices. We call the associated centered Gaussian processes Brownian processes. Each is a standard Brownian bridge plus the affine interpolation of an independent Gaussian endpoint vector. Singular endpoint covariance matrices give hard endpoint constraints, such as the pinned start $X(0)=0$ of Brownian motion. This parametrization gives explicit operator domains and energy spaces. On every mesh it gives an exact sparse precision matrix, so the likelihood of $n$ observations can be evaluated in $O(n)$ operations. Nonnegative realizations with a null space are treated separately through orthogonally normalized pseudoinverses, which are proper Gaussian covariances.

comments
12 pages, 2 figures
fields
Probability
pdf ↗ page arXiv:2610.36762 ↗

In this paper, we study once-reinforced random walk on $\mathbb{Z}$, and prove sharp Chung-type and Hartman-Wintner laws of the iterated logarithm for the range. Let $R_n$ denote the range at time $n$ and let $c>0$ be the reinforcement parameter. Then, almost surely, \begin{equation*} \liminf_{n\to\infty}\frac{R_n}{\sqrt{n/\log\log n}}=\fracπ{\sqrt{2}}, \qquad \limsup_{n\to\infty}\frac{cR_n}{\sqrt{2n\log\log n}}=1. \end{equation*} The proofs are based on the independent increment decomposition of the inverse range and sharp generating function estimates.

comments
23 pages
fields
Probability
pdf ↗ page arXiv:2610.36790 ↗

We study the behaviour of stationary measures for a class of dispersive PDEs damped by a linear zero-order term and perturbed by a small kick noise. Under some natural assumptions on the law of the noise, we prove that any family of stationary measures satisfies the large deviation principle in the vanishing noise limit. The result highlights the role of nonlinear smoothing and applies to the nonlinear Schrödinger and Korteweg-de Vries equations, and to a variety of random perturbations including Gaussian and exponential variables.

comments
33 pages
fields
Probability, Analysis of PDEs, Dynamical Systems
pdf ↗ page arXiv:2610.36815 ↗

Despite significant advances in computer vision, semantic segmentation models remain vulnerable to backdoor attacks, posing a security risk. These attacks redirect pixels of a chosen victim class to an attacker-specified target class when a trigger is present, while preserving clean predictions. Existing defenses are largely derived from image classification, while dedicated segmentation defenses remain scarce and typically require retraining or model modification. We introduce RePAIR, a post-hoc, label-free, attack-unaware defense based on the observation that backdoor-induced shifts can be stronger in the classifier head than in internal representations, leaving recoverable victim-class semantics. RePAIR infers the victim-target relation from an unlabeled, potentially contaminated image pool, constructs a safety-approved semantic reference, and localizes suspicious pixels by comparing head predictions with internal semantic evidence. It then applies feature-space and output corrections toward the inferred victim semantics, while conservative localization and output constraints limit changes to benign predictions. We model the attack as adversarial probability transport and repair as its approximate reversal, deriving bounds and sufficient conditions for localization, residual attack success, clean-side changes, defense-induced target predictions, and attack-unaware pair identification. Experiments across segmentation architectures and attack settings show that RePAIR substantially reduces attack success while largely preserving clean performance.

comments
26 pages total, 9 pages main paper, 2 figures
fields
Probability
pdf ↗ page arXiv:2610.37015 ↗

We examine separation theorems for the increasing convex (stop-loss) order and weak 3-convex order. We improve prior proofs for the increasing convex order by using integrated survival functions. We extend the separation theorem for weak 3-convex order from non-negative random variables to general real-valued random variables with finite second moments via a constructive approach based on higher-order integrated survival functions. A counterexample verifies that the $3$-convex order and $2$-convex order chain does not hold for the weak $3$-convex order. Our results enrich stochastic order theories for actuarial science and risk management.

comments
14 pages, 4 figures
fields
Probability
pdf ↗ page arXiv:2610.37025 ↗

We characterize Gaussian fluctuations of Fourier-Hermite functionals of stationary Gaussian fields on locally compact Abelian groups via spectral delocalization: conditional on the total spectral sum being near the probed frequency, no partial sum concentrates. Under suitable assumptions, this reduces Gaussian convergence to small-ball estimates for convolution powers of the spectral measure, without model-specific contraction estimates. On $\mathbb T^d$, the criterion yields CLTs at every fixed frequency for band-limited fields and random trigonometric polynomials, and extends the known failure of Gaussian convergence for planar arithmetic random waves from the fourth chaos to all even orders $q\ge4$. On $\mathbb R^d$ and $\mathbb Z^d$, we establish a Breuer-Major theorem for Fourier functionals at every frequency with nonzero limiting variance. Cyclical long-memory models exhibit frequency-dependent Gaussian convergence. We also establish CLTs and variance asymptotics for singular continuous spectral measures, including those of stationary Gaussian solutions of linear PDEs. At zero frequency, weak Ahlfors regularity of the spectral support and an $L^2$ density suffice for second-chaos CLTs, covering fractal sets, the Airy surface of Gaussian sea waves, and smooth compact manifolds. For a class of smooth hypersurfaces, we obtain the same variance rates and CLTs known for classical nonlinear functionals of Berry's random wave model. Finally, for random waves with spherical spectral densities bounded above and below, we classify Gaussian convergence at every frequency $\|z\|\le q$: it fails only for the planar second chaos when $0<\|z\|<2$.

comments
37 pages, 1 figure
fields
Probability, Mathematical Physics, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.37113 ↗

We prove that on $\mathbb Z^d$, with $d>6$, and under the usual assumption that the two-point function and the Green's function are of the same order, the probabilities to hit any finite set for a critical Branching random walk starting from the origin and the critical percolation cluster of the origin are within a positive multiplicative constant one from each other, as long as the set is sufficiently far away from the origin comparatively to its diameter. The proof relies on the introduction of a truncated equilibrium measure in percolation and the derivation of an approximate last passage formula.

comments
22p. No use of AI
fields
Probability
pdf ↗ page arXiv:2610.37133 ↗

This paper addresses the solvability of multidimensional infinite-horizon Markovian backward stochastic differential equations with continuous coefficients. We do not require the Lipschitz property or other quantitative controls of differences, and each component of the generator is allowed to depend simultaneously on all components of the multidimensional backward variables. A local L^q-domination framework adapted to the infinite-horizon setting is developed and combined with our linear flow conjugacy analysis to obtain the required distributional control in the presence of unbounded linear drift. We also derive a common Markov representation valid for the entire family of initial states. The analytical obstacles arising from weak regularity and the multidimensional structure are thereby overcome. Furthermore, we consider a class of infinite-horizon nonzero-sum differential games with a general non linear structure, allowing player controls to be fully coupled and permitting payoff criteria that are not restricted to an exponential discount form, establish an equivalence between the game and the corresponding multidimensional infinite-horizon backward stochastic differential equations, and prove the existence of a Nash equilibrium.

fields
Probability
pdf ↗ page arXiv:2610.37141 ↗

We prove sharp almost-sure asymptotics for eigenvector preferential attachment trees in which each attachment uses eigenvector centrality with probability $α\in(0,1]$ and uniform selection with probability $1-α$. A unique persistent hub eventually has the maximum degree, which is asymptotic to $γ^2t^{2/3}$, where $γ=(9α/10)^{1/3}$. The Perron-Frobenius eigenvalue is asymptotic to $γt^{1/3}$, and the second-largest degree is asymptotic to $2γt^{1/3}$. The hub and its neighbors each carry asymptotically half of the squared Perron mass. We also identify the centrality profile by degree and an explicit mixture of geometric laws for the degree of a uniformly sampled vertex. A limiting distance profile quantifies the dandelion geometry, and vertices beyond the hub's neighbors have maximum degree of order at most $\log t$.

comments
24 pages, 1 figure. Comments are welcome
fields
Probability
pdf ↗ page arXiv:2610.37156 ↗

We prove universal Gaussian and stable fluctuation limits for one-dimensional long-range first-passage percolation with edge times $\|x-y\|^αω_{x,y}$. The positive i.i.d. weights satisfy $\mathbb{P}(ω\le t)=t^{θ+o(1)}$ as $t\downarrow0$ and $\mathbb{E}(ω^ζ)<\infty$ for some $θ,ζ>0$. For $α>1+2/θ$, the passage time $T(0,n)$ has a positive deterministic linear growth rate $μ$. For $α>1+3/θ$, we establish a Gaussian central limit theorem with linear asymptotic variance, a functional version, and limit theorems for geodesic hop and cut point counts. For $1+2/θ<α<1+3/θ$, assuming a regularly varying lower tail, we obtain a totally left-skewed stable limit with index $β=θ(α-1)-1$ and an explicit scale, together with convergence to a spectrally negative stable Lévy process in the Skorokhod $M_1$ topology. The proof constructs i.i.d. blocks between hubs through which every crossing geodesic must pass. A spread-out comparison metric makes this regeneration available for every $ζ>0$, even when the edge law has infinite mean. The key estimate identifies the exact left-tail asymptotic of the centered block reward, generated by a single fast long edge. A monotone within-block approximation yields the functional stable limit.

comments
40 pages, 5 figures. Comments are welcome
fields
Probability
pdf ↗ page arXiv:2610.37180 ↗

For a class of smooth stationary Gaussian fields, we study the fluctuations of the volume, surface area and Euler characteristic of the unbounded excursion component in growing boxes $Λ_n=[-n,n]^d$. We establish nondegenerate central limit theorems for all three functionals throughout the supercritical regime in every dimension $d\geq 2$. We further quantify these central limit theorems by establishing Berry–Esseen bounds of order $O\bigl(|Λ_n|^{-(1-1/s)}\bigr)$ in Kolmogorov distance, where $s\in(1,2]$ depend on the regularity and decay assumptions. Especially for the Bargmann–Fock field, we obtain a Berry–Esseen bound of $O\bigl(|Λ_n|^{-1/2}\bigr)$ for all three functionals. A matching lower bound shows that this rate is sharp for the Euler characteristic. For the volume, we improve the previously known rate of $O\bigl(|Λ_n|^{-1/2}(\log|Λ_n|)^{2d}\bigr)$. We mainly use Stein's method and moment bounds for differences arising from white-noise resampling to finish our proof.

comments
42 pages
fields
Probability
pdf ↗ page arXiv:2610.37240 ↗

We study perfect sampling algorithms for Ising spin glasses based on coupling from the past and bounding chains, on several graph families and coupling distributions. For the single-site bounding chain, we establish coalescence in $O(n\log n)$ updates under Dobrushin conditions. On sparse random graphs, our numerical results suggest that efficient sampling extends throughout the tree-uniqueness regime, with a slowdown beyond this threshold. On Erdős-Rényi random graphs $\mathcal{G}(n,d/n)$ with standard Gaussian couplings, we introduce a block bounding chain and prove its coalescence in $O(n\log n)$ block updates in the tree-uniqueness regime for sufficiently large $d$, each block update requiring $O(n)$ operations, adapting a proof in~\cite{Efthymiou_Zampetakis_2024} for the rapid mixing of block Glauber dynamics. On periodic lattices, our experiments instead suggest that the single-site bounding chain is limited by the looseness of its bounding envelope, whereas on the Sherrington-Kirkpatrick model, slow coalescence is accompanied by damage spreading at every inverse temperature tested. These observations motivate alternative bounding representations and couplings as strategies to improve perfect sampling on spin glasses.

comments
47 pages, 10 figures
fields
Probability
pdf ↗ page arXiv:2610.37300 ↗

We study the frozen percolation model on the complete graph $K_n$, where each edge appears at rate~$1/n$ and connected components are destroyed by lightning strikes hitting vertices at rate $λ_n$ (thus the rate at which a connected component is destroyed is linear in its size). We focus on the slightly subcritical sub-regime of the self-organized critical (SOC) regime, assuming $(\ln n)^{4/3}n^{-1/3} \ll λ_n \ll 1$. Our main object of study is the so-called window process, which allows us to study the fine details of the mechanism that drives the SOC behavior of this dynamic random graph. We prove two scaling limit theorems. First, we show that the fluctuations of the window process around its deterministic fluid limit converge to a stationary Ornstein–Uhlenbeck process, whose mean-reverting property captures the SOC nature of the model. Second, by zooming in on a finer space-time scale, we prove that the scaled increments of the window process converge to a tempered stable Lévy process.

comments
44 pages
fields
Probability
pdf ↗ page arXiv:2610.37320 ↗

We study stationary coexistence in stochastic Gause predator-prey systems with independent multiplicative environmental noise and trophic responses depending on both population densities. A weighted population balance cancels the trophic interaction and yields moments of order greater than one without imposing predator self-regulation. These estimates give a direct occupation-measure proof of a sufficient coexistence criterion based on prey establishment and predator invasion of the prey-only stationary environment. For responses whose rare-predator limit is strictly increasing and whose value at positive predator density never exceeds that limit, we also establish necessity of positive predator invasion whenever the prey-only stationary law exists. If the prey per-capita growth attains its global maximum at zero, interior stationarity itself forces prey establishment, giving a necessary-and-sufficient criterion beyond logistic growth. Logistic and bounded saturating growth laws provide explicit Gamma and beta-prime resident distributions. Applications include Holling, Ivlev, Beddington-DeAngelis, and Crowley-Martin responses. Responses with the same rare-predator limit share the coexistence threshold, although predator interference changes the interior distribution. Reproducible quadrature and simulations illustrate this distinction. The analysis builds on existing stochastic persistence theory and identifies precisely the additional comparison argument needed for necessity, including exclusion of interior invariant measures at zero predator invasion.

comments
17 pages, 3 figures
fields
Probability, Dynamical Systems
pdf ↗ page arXiv:2610.37372 ↗

It is known that randomly initialized neural networks have Gaussian behaviour in the $n\to \infty$ limit. Unfortunately, this approximation can be poor for finite $n$ and gets worse as one goes deeper into the network. In this work, we develop a more accurate approximation that is \emph{slightly} non-Gaussian, which we call the para-Gaussian distribution and is constructed from the Hermite polynomials. We show how this para-Gaussian approximation can lead to a practical recursive method to approximate the deeper layers of the network. Finally, we apply this theory to linear neural networks (where things can be solved exactly) and to ReLU neural networks (where things can be approximated).

comments
34 oages, 9 figures
fields
Probability
pdf ↗ page arXiv:2610.37388 ↗

Brownian stochastic interpolants connect paired endpoint samples, yet their sampling error can depend on displacement moments even when neither endpoint has a first moment. For arbitrary Borel couplings, we prove early-stopped, target-to-output KL guarantees for standard Euler sampling from the true source. A fourth displacement moment gives an explicit $O(N^{-1})$ bound; a second moment suffices for fixed-law convergence as drift error vanishes. For one fixed coupling with no displacement moment above two, original bridge-path KL stays of order one, comparison-path KL is of order $1/\log N$, and terminal KL is at most $C/N^2$ on the same grid. The path decomposition identifies the information about the source discarded by Markov projection. When the original bridge is Markov, current-state Euler sampling instead has a second-moment $O((M_2+d)/N)$ bound; finite-entropy Schroedinger bridges for the probability Brownian reference are included. We also construct a current-state learner with expected output KL consistency from iid endpoint pairs. Numerical diagnostics illustrate the distinct errors. Conditional diffusion identities specify a restoration interface, while image controls find no iterative advantage over cheap probabilistic predictors. Uniform terminal rates for general second-moment couplings remain open.

comments
56 pages, 4 figures; 8-page main text and complete appendices
fields
Probability
pdf ↗ page arXiv:2610.37413 ↗

We establish a quantitative local law for random $N$-lifts of a fixed finite connected simple graph with minimum degree at least three. Away from the singular energies of the universal-cover spectrum, we obtain high-probability entrywise Green's function estimates down to spectral scales $(\log N)^C/N$, uniformly in the bulk and near regular edges and cusps. As a consequence, we establish complete eigenvector delocalization: every normalized eigenvector $u$ whose eigenvalue stays a small but fixed positive distance away from the singular energies satisfies $\|u\|_\infty\leq(\log N)^C/\sqrt{N}$ with high probability.

comments
70 pages; 9 figures
fields
Probability, Mathematical Physics, Combinatorics, Spectral Theory
pdf ↗ page arXiv:2610.37463 ↗

We study the scaling of the expected total queue length in an $n\times n$ input-queued switch. It has been conjectured that the optimal scaling is $O(n/(1-ρ))$, up to logarithmic factors, where $ρ$ is the system load. This conjecture has been established in the traditional heavy-traffic regime and in certain sub-regimes of the large-system regime. But it remains unresolved over a wide range of load scalings in the large-system regime. In this paper, we develop a family of scheduling policies parameterized by a positive integer $k\ge 2$. We show that under uniform traffic, our policy achieves an expected total queue length of $O\bigl({n\log\max\{n,1/(1-ρ)\}}\big/{(1-ρ)^{2^k/(2^k-1)}}\bigr)$, as long as $ρ\to 1$ in the large-system regime. As $k$ increases, the exponent $2^k/(2^k-1)$ approaches $1$, bringing our bound arbitrarily close to the conjectured scaling. The expected amortized scheduling cost of the policy is $O(n^3 k)$ per time slot. Our policy is of a batching type, with a novel recursive design that allows an arbitrary number of batching levels. This design principle may also be useful for other scheduling problems in which combinatorial constraints restrict which jobs can be served simultaneously.

fields
Probability
pdf ↗ page arXiv:2610.37521 ↗

In this note, we give a short probabilistic proof of the two-layer Brownian representation of the stationary measure of the open Kardar–Parisi–Zhang (KPZ) equation on a finite interval with arbitrary Neumann boundary parameters, conjectured by Barraquand and Le Doussal and recently established in full generality by Dunlap, Gu and Rosati. Our starting point is Bryc's two-layer Gibbs representation of the stationary measure of the open asymmetric simple exclusion process (ASEP) under Liggett's condition, together with our previously established probabilistic interpretation of the associated Bryc polynomials in terms of inhomogeneous birth–death Markov chains. Martingale estimates yield uniform asymptotics for these polynomials under weakly asymmetric scaling. Combined with known convergence results from open ASEP to open KPZ, these estimates recover the stationary height profile of the open KPZ equation as the first marginal of a reweighted pair of Brownian motions.

comments
17 pages, 1 figure
fields
Probability
pdf ↗ page arXiv:2610.37618 ↗

We study the stationary measure of the open asymmetric simple exclusion process (ASEP) in the fan region with general boundary rates. The stationary height fluctuations in the maximal current phase and on its boundaries within the fan region are known to have limits involving Brownian excursions or meanders. We establish discrete counterparts of these limits: on intervals away from the boundaries, the joint law of the occupation variables is approximated in total variation distance by particle configurations constructed from lazy simple random walk excursions and meanders. In the high- and low-density phases of the fan region, we establish Bernoulli product approximations with errors that decay exponentially in the distance from the relevant boundary. Our methods combine refined combinatorial estimates for Motzkin paths and the matrix product ansatz in the totally asymmetric case with a Fourier-type decomposition (based on Askey–Wilson polynomials) of the change of measure for general asymmetry. In addition, in the maximal current phase and on its boundaries within the fan region, we prove moderate deviation estimates for particle counts on arbitrary sets of sites. These estimates rely on a covariance identity for an auxiliary path measure, through which total nonnegativity (of order two) of the matrices yields negative correlations.

comments
53 pages
fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2610.37620 ↗

For diffusion-limited aggregation on $\mathbb{Z}^d$ with $d \geq 2$, we rule out convergence in volume to bounded regions with boundary measure zero. Moreover, the number of boundary edges grows linearly. For Euclidean DLA in dimensions $2$, $3$, $8$, and $24$, we prove an analogous shape exclusion.

comments
24 pages, 1 figure
fields
Probability
pdf ↗ page arXiv:2610.37625 ↗

We give algorithms for approximating the partition function $Z_N$ of the Sherrington–Kirkpatrick (SK) model and finite mixed Ising $p$-spin models with $p\ge 3$. The algorithms apply at a fixed margin below the second-moment threshold $β_{\mathrm f}$ with homogeneous weak external fields $|h|\leqρ_0N^{-1/4}$, for fixed $ρ_0>0$. For any accuracy $\varepsilon\in(0,1)$ and failure parameter $δ\in(0,1)$, they approximate $\log Z_N$ to additive error $\varepsilon$. The number of exact operations is polynomial jointly in $N$, $\varepsilon^{-1}$, and $δ^{-1}$. The disorder variables have i.i.d symmetric distribution with unit variance. The failure probability is at most $δ$ plus an explicit disorder tail bound. This tail bound tends to zero under a finite fourth moment condition. Our approach builds on the cluster expansions of Aizenman–Lebowitz–Ruelle~\cite{ALR87} and Dey–Wu~\cites{DW23,DW26}. For higher-order mixtures, we retain hypergraphs using few vertices and evaluate their total weight deterministically through local spin averages and inclusion–exclusion. For SK, we impose cutoffs on the number of vertices of degree at least three and on the total number of edges. We estimate the retained sum using randomized color coding~\cite{AYZ95}. A lower tail control based on reverse hypercontractivity~\cite{MOS13} converts the mean-square approximation of the normalized partition function into logarithmic accuracy.

code
github.com/qiangwu2/approximate-counting-spin-glasses ∎ Lean project (lakefile.toml, lean-toolchain) builds, no sorry at 9ba7d11, checked 2026-10-28 build details
comments
38 pages, 3 figures. AI-assisted and a companion lean formalization in https://github.com/qiangwu2/approximate-counting-spin-glasses
fields
Probability
pdf ↗ page arXiv:2610.37630 ↗

We study the stationary measure of the open asymmetric simple exclusion process (ASEP) on the coexistence line. Our starting point is Bryc's two-layer Gibbs representation, together with the probabilistic interpretation, established in the first paper in this series, of the associated Bryc polynomial in terms of an inhomogeneous birth–death chain. Our main result gives a sharp approximation of the stationary measure. Specifically, we identify the shock location as an explicit observable of the particle configuration and obtain optimal bounds on the total variation distance between its law and the uniform measure on the lattice. Conditional on the shock location, we approximate, in total variation distance, the joint law of the occupation variables away from the shock by Bernoulli product measures with different densities to its left and right. These approximations lead to further applications, including fluctuation limits, moderate deviation bounds for the particle density, and limits for correlation functions under the stationary measure.

comments
53 pages
fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2610.37636 ↗

We investigate the role of asymmetry in the stationary measures of integrable probabilistic models with boundary interactions. This paper is the first in a series of three focusing on the open asymmetric simple exclusion process (ASEP). The stationary measure of open ASEP admits a two-layer Gibbs representation in which a correction term, which we call the Bryc polynomial, encodes the dependence on the asymmetry parameter. We give a probabilistic interpretation of the Bryc polynomial in terms of a birth-death-type Markov chain. As an application, we obtain sharp total variation approximations of the stationary measure by Bernoulli product measures in the high- and low-density phases of the shock region. These approximations yield further applications, including moderate deviation estimates for the stationary measure of open ASEP, as well as asymptotic results for the related model of open ASEP with light particles.

comments
41 pages
fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2610.37639 ↗

Algebraic Geometry34

This paper builds on the work of Goncharov by studying the correspondence between topological spectral covers constructed from generic $A_{m-1}$-webs with rank $m\ge2$, Hitchin spectral curves over good Hitchin loci, and the Grothendieck groups $K_0(\mathcal{D}_Γ)$ associated with the quivers with potential $(Q_Γ,W_Γ)$ assigned to corresponding bipartite $A_{m-1}$-graphs. We show that the automorphisms on the non-frozen quiver lattices $Λ_Γ^{\circ}$ induced by compositions of elementary moves on ideal webs/bipartite graphs are the same as the Gauss–Manin monodromy actions on the $SL_m$ Hitchin fiber lattices $H_1(\operatorname{Prym}(Σ);\mathbb{Z})$ locally around discriminant loci. And the Gauss–Manin monodromy actions agree with the automorphisms on the Grothendieck groups $K_0(\mathcal{D}_Γ^{\circ})$ induced by compositions of mutations on the quivers with potential $(Q_Γ,W_Γ)$.

comments
86 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35898 ↗

We classify K3 surfaces with finite automorphism group in positive characteristic. The classification is the same as over the complex numbers, except when the base field is $\overline{\mathbb{F}}_p$.

comments
24 pages, comments welcome
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35899 ↗

In this paper, we give a new proof of the Hopkins-Morel-Hoyois theorem for fields $k$ of characteristic unequal to $p$ containing all primitive $p$-power roots of unity for a fixed prime $p$. This new proof follows Quillen's proof for the structure of the homotopy groups of $\mathrm{MU}$ which utilizes topological power operations. To this end we define a motivic analog of the Steenrod power operations, using the theory of normed motivic spectra as developed by Bachmann–Hoyois and their application for the definition of motivic colimits as developed by Bachmann-Elmanto-Heller.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35903 ↗

We prove a homological obstruction to smoothing projective Calabi-Yau threefolds with nodes and anticanonical cones over smooth del Pezzo surfaces of degrees six and seven. On a crepant resolution that is small over the nodes and blows up the cone vertices, any smoothing forces a relation among link classes bounding in the local Milnor fibers. The relation is nonzero at every node and degree-seven point, and at each degree-six point smoothed through the one-dimensional component. The proof uses vanishing periods and applies to smoothing arcs of arbitrary contact order, extending the necessity direction of Friedman's criterion. We apply the obstruction to a Batyrev hypersurface whose thirty singularities are individually smoothable but which admits no smoothing, even formally. On its mirror partner, two nodes cannot be smoothed by any analytic deformation, independently of a locally nonsmoothable cone point.

comments
34 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35911 ↗

We prove that the nonvanishing conjecture implies the abundance conjecture.

comments
48 pages. Still under revision for writing but forced to submit here due to an announcement of the proof of the abundance conjecture by OpenAI
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35913 ↗

Let $\mathcal{M}$ be a positroid. We prove that the positroid variety $Π_{\mathcal{M}}$ is toric with respect to some algebraic torus if and only if $\mathcal{M}$ is regular. In that case, $Π_{\mathcal{M}}$ is toric with respect to the standard diagonal torus. Thus, there are no unexpected toric positroid varieties.

fields
Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.35914 ↗

We give a necessary and sufficient homological criterion for smoothing projective Calabi-Yau threefolds with nodes and anticanonical cones over smooth del Pezzo surfaces of degrees five, six and seven, or over a smooth quadric surface. For some choice of local smoothing components, the criterion requires a relation on a crepant resolution among link classes whose associated local parameters avoid every discriminant. The complexified relation space describes first-order deformations modulo locally trivial ones. Necessity follows from local period and vanishing-cycle arguments; sufficiency applies Friedman's unobstructedness theorem to a crepant partial resolution. Whenever the criterion holds, the deformation space on the chosen components is smooth. A nonsmoothable example, whose deformation space is reduced with four smooth components, shows why nonvanishing at every singular point is insufficient. We also construct a smoothable Batyrev hypersurface admitting no smoothing induced by proper ambient families obtained from Minkowski decompositions at a single lattice degree.

comments
62 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35915 ↗

We give an alternative proof of Mongardi's result that every finite group of symplectic automorphisms of a hyperkähler manifold of $K3^{[n]}$-type acts trivially on the discriminant groups of its second cohomology lattice and its coinvariant lattice. The proof combines sphere-packing bounds, Vinberg's reflection chambers and structure results on even $2$-elementary lattices.

comments
17 pages, comments welcome!
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35916 ↗

We formulate a parametrized Jacobian Problem over $\mathbb{C}^{n}$ by incorporating independent scale parameters into the linear part of a polynomial map via polar decomposition. We prove that requiring the Jacobian determinant to remain constant uniformly across the parameter space decouples the Jacobian equation into a system enforcing the vanishing of all principal minors of the perturbation matrix. This system strictly forces lower triangularization up to variable permutations, establishing a necessary and sufficient condition for polynomial invertibility and providing an explicit, canonical scheme for constructing the inverse map in every dimension.

comments
11 pages
fields
Algebraic Geometry, Mathematical Physics
pdf ↗ page arXiv:2610.35918 ↗

We investigate how geometric priors shape neural approximations of the complex Monge–Amp`ere equation on the quintic Calabi–Yau threefold. This fixed-$ψ$ diagnostic compares four single-patch ans"atze: a coordinate SIREN potential model, a direct-metric Cholesky baseline, an algebraic section ansatz, and a section-plus-neural hybrid. In a 96-run learning-rate and frequency archive, section-based models reach validation $σ=4.14$--$4.16%$, whereas SIREN and Cholesky remain at $41.29%$ and $44.27%$. Multi-seed controls use a shared affine-$η$ evaluator and five fresh validation clouds. Stable section settings give $3.84$--$4.77%$, compared with $28.46$--$30.59%$ for Cholesky and $34.73$--$38.13%$ for SIREN. Hybrid projective features do not improve best-case accuracy, but they stabilise the stressed regime. They limit $σ$ to $4.26$--$4.41%$, versus $4.45$--$205.45%$ for raw coordinates. Within this protocol, section-based geometric structure is the main driver of accuracy; coordinate-only models plateau at substantially higher error, and raw hybrid flexibility can amplify instability. We provide a projective-scaling derivation, document protocol asymmetries, and identify projectively consistent hybrid corrections as the next step.

journal
2026 Mach. Learn.: Sci. Technol. 7 035049
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35920 ↗

In this research article, we make an in-depth investigation of the computations for Thom Polynomials for $\mathcal{A}$-singularities of parametric curves germs in the plane and space cases using the framework established by Ohmoto and the Restriction Method introduced by Rimányi, and we also propose a geometric way for such computations using degeneracy loci Chern classes formulas proven by Fulton, which seems a promising way to generalize the Thom Polynomial expression for the stabilization of such singularities in $\mathbb{C}^N$, depending exclusively on $N$ as a parameter.

comments
23 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35922 ↗

Let $C(v)$ denote a Schubert cell in the flag variety $\operatorname{GL}_n/B$, and let $X(w)$ be a Schubert variety containing $C(v)$. Mellit has conjectured that, if $X(w)$ is smooth, the existence of an open $B$-orbit in the conormal bundle to $C(v)$ in $X(w)$ is equivalent to $C(w_0w)$ being contained in the smooth locus of $X(w_0v)$. We prove a stronger version of this conjecture, requiring only that $X(w)$ is factorial. We prove our result by constructing a sequence of graphs associated to the pair $v,w$ such that the existence of a cycle in one of these graphs is equivalent to the non-existence of such an open orbit.

comments
23 pages, 10 figures
fields
Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.35923 ↗

In this article we show that a biregular involution $i : X \to X$ defined on an unnodal Coble surface $X$ with irreducible boundary curve $\{C\} = |- 2 K_X|$ is necessarily a lift of a Bertini involution. Then we show with counterexamples that the hypothesis of irreducibility of $C$ and unnodality of $X$ are both necessary.

comments
28 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36072 ↗

We give an obstruction to the existence of bundles $\mathcal E$ on a smooth $n$-dimensional polarized variety such that $\mathcal E$ is $a$-regular and ${\mathcal E}^*(K_X+(n+1)H)$ is $b$-regular. In particular, this gives an obstruction to the existence of weakly Ulrich bundles. Using the latter and some recent examples of [A1] in dimension $2$, we prove that for any $n \ge 2$ there exist many smooth $n$-dimensional varieties $X$ with a very ample line bundle $H$, such that there is no weakly Ulrich bundle for $(X,H)$.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36231 ↗

Let $X$ be a smooth connected complex projective variety. We give a structure theorem for irreducible components of the reduced rank-two cohomology jump loci on $X$. More precisely, for such a component $Z$, if the general point of $Z$ is projectively virtually abelian or projectively rigid, then either $Z$ is a symmetric square of an algebraic torus, or a quotient of a torus by an involution, or its normalization $Z^ν$ is a torus. If the general point is neither, then either $Z^ν$ is a torus, or an étale quotient of $Q^ν \times T$ where $Q$ is a fixed-determinant, fixed-inertia component of the character variety of some DM-curve.

comments
46 pages. Comments welcome!
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36251 ↗

We show that several Dold-Kan type equivalences can be understood as a passage from constructible sheaves on symmetric products (i.e. spaces of polynomials) to polynomial sheaves in the sense of Goodwillie-Weiss calculus, by means of a natural analytic expansion functor. This geometric perspective provides access to standard tools from sheaf theory such as the six functor formalism and $t$-structures. For example, the equivalence between filtered objects and coherent chain complexes in a stable $\infty$-category receives a geometric incarnation as Verdier duality between $!$-sheaves and $*$-cosheaves on an $\infty$-dimensional simplex. As a main application, we obtain an intrinsic geometric explanation for the appearance of vanishing cycle data for perverse sheaves in a paracyclic variant of the Dold–Kan correspondence pioneered by Dwyer and Kan, at the same time lifting their equivalence to stable $\infty$-categories.

comments
53 pages, 3 figures
fields
Algebraic Geometry, Algebraic Topology, Category Theory
pdf ↗ page arXiv:2610.36314 ↗

We introduce a tropical framework for the study of spin Hurwitz numbers and their completed cycles analogues. We use degeneration techniques to obtain an interpretation of spin Hurwitz numbers via tropical covers. We employ the tropical methods to obtain new closed formulae for spin Hurwitz numbers in special cases. Moreover, we employ Fock space techniques to tropicalise completed cycles spin Hurwitz numbers. Using this, we prove that spin Hurwitz numbers admit a piecewise polynomial structure akin to their classical counterparts. We also derive wall–crossing formulae.

comments
56 pages, 19 figures
fields
Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.36390 ↗

We study invertibility and tameness for polynomial maps of affine three-space of total degree at most four. We prove that every complex Keller map in this degree range is a polynomial automorphism and, independently, that every complex polynomial automorphism of total degree at most four is tame. The invertibility argument combines homogeneous factorization of the leading part, invariant rings of conic pencils, and residue obstructions on cubic fibers. The tameness argument uses a criterion based on a constant second directional derivative, low-degree coordinate classifications, and local tameness of plane polynomial automorphisms over a polynomial parameter ring. The invertibility conclusion extends to every field of characteristic zero.

comments
31
fields
Algebraic Geometry, Commutative Algebra
pdf ↗ page arXiv:2610.36495 ↗

We prove that most rational smooth Fano threefolds are compactifications of $\mathbb{G}_m^3$. More precisely, we study which smooth Fano threefolds are cluster type, i.e., for which smooth Fano threefolds $X$ one can find a boundary $B\in |-K_X|$ and an embedding $\mathbb{G}_m^3\hookrightarrow X\setminus B$. We prove that many important classes of smooth Fano threefolds are cluster type. For instance, a smooth Fano threefold which is the blow-up of either $\mathbb{P}^3$ or $\mathbb{Q}^3$ and is general in its deformation space is of cluster type. As a consequence, we show that among the $105$ families parametrizing all smooth Fano threefolds, the general element of precisely $87$ families is of cluster type. We prove that a general prime smooth Fano threefold of genus $7$ is not of cluster type. This is the only family whose general element is rational but not of cluster type. Finally, we prove that cluster type projective klt varieties are cylindrical and we give a criterion for such a cylinder to be anticanonically polar. Using this, we reduce the cylindricity problem for smooth Fano threefolds to the families $\mathcal{N}^{\mathrm{o}}\,1.3$, $\mathcal{N}^{\mathrm{o}}\,1.5$, and $\mathcal{N}^{\mathrm{o}}\,1.6$.

comments
70 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36656 ↗

Let $h\in\C[X_1,\dots,X_n]$ have constant nonzero Hessian determinant, let $h^{*}$ be its formal Legendre transform and $g=\vv{X,\grad h}-h$. The Nambu derivations of $\grad h$ satisfy $δ_i(g)=X_i$ and $δ^α(g)=(\partial^αh^{*})\circ\grad h$. Consequently, in the gradient case the differential Galois criterion of \cite{ACH19}, consisting of $n$ membership conditions, reduces to a single membership condition. Equivalently, $\grad h$ is a polynomial automorphism if and only if $g\in\C[\partial_1h,\dots,\partial_nh]$. Moreover, $\grad h$ is a polynomial automorphism if and only if $δ^α(g)=0$ for $|α|=(°h-1)^{n-1}+2$, and then the coefficients $δ^α(g)(x_0)/α!$ of $h^{*}$ at $\grad h(x_0)$ determine the inverse. If the Hessian determinant of $h$ is merely a nonzero polynomial, then $g$ is still a primitive element of $\C(X)$ over $\C(\partial_1h,\dots,\partial_nh)$. For the Nambu-tower criterion, doubling a Keller map yields no independent conditions. After an affine change of coordinates putting $\grad h$ in the normal form $X+H$, the formal Legendre transform is the sum of the formal Pascal difference series of $g$, and the termination of this series implies polynomial invertibility. In at most three variables the normal form of every gradient Keller map has strongly nilpotent Jacobian part and is therefore Pascal finite, by a theorem of de Bondt, and an explicit gradient automorphism in four variables has a normal form which is not Pascal finite. The five-variable counterexample of Meng and Yang shows that the properties used here do not imply the Hessian conjecture. The results do not settle the open cases $\JC_2$ and $\HC_4$.

fields
Algebraic Geometry, Commutative Algebra
pdf ↗ page arXiv:2610.36736 ↗

Let $f:X\to Y$ be a birational morphism between smooth projective threefolds, Kollár asked that if $f$ could be decomposed into smooth blow-downs and smooth flops. In this paper we show that the toric version is true.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36802 ↗

Given two nonzero polynomials $f, g \in \mathbb{R}[x_1,\dots,x_n]$ vanishing at the origin $0 \in \mathbb{R}^n$, we study the numerical range of the limit $$ \lim_{x \to 0} \frac{f(x)}{g(x)}, $$ denoted by $\mathcal{R}_{f,g}$, which is defined as the set of all possible limits of the rational function $\displaystyle\frac{f}{g}$ as $x$ tends to $0$. When $n = 2$ and $g$ has a non-isolated zero at $0$, we describe all possible forms of $\mathcal{R}_{f,g}$ and provide an effective algorithm to compute this set based on Puiseux expansions. This is a follow-up to our recent results~\cite{Dinh2025}, in which we only considered the case when $0$ is an isolated zero of $g$.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36842 ↗

In the article, we generalize to an arbitrary dimension results obtained by the author and S.Yu. Nemirovski that related to Chisini's conjecture claimed that if $B$ is the discriminant of two generic covers $f_i:X_i\to \mathbb P^2$, $i=1,2$, and $\max (°f_1,°f_2) \geq 5$, then $f_1$ and $f_2$ are equivalent covers.

comments
15 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36908 ↗

Rational equivalence need not be closed in the Chow topology, and its closure can have infinitely generated homotopy groups. For a connected reduced complex projective variety, we prove that this closure is locally contractible and that its quotient in the algebraically trivial cycles is a finite or countable weak product of circles. The quotient map admits a global continuous section. Consequently, the higher homotopy of the cycle space is carried by the closure and is canonically identified with Lawson homology. We describe the remaining fundamental group quotient and its integral Walker comparison. Applications include smooth projective examples with infinite rank closure homotopy throughout prescribed finite intervals and a family of abelian threefolds with four simultaneously dense, nonconstructible quotient rank strata.

comments
32 pages
fields
Algebraic Geometry, Algebraic Topology
pdf ↗ page arXiv:2610.36913 ↗

We study Frobenius action on the trigonometric Ruijsenaars–Schneider system associated with cotangent bundles to complete flag varieties. The construction continues the program initiated by Etingof and Varchenko and developed in quantum $K$-theory by the author and Smirnov. We explain how the monodromy eigenvectors recently constructed by Tarasov and Varchenko fit the quantum/classical duality between Bethe equations and tRS level sets. At a root of unity, all partial shifts cancel in the Newton–Adams combinations of the tRS Hamiltonians, leaving precisely the operators obtained by Frobenius action. This proves compatibility of the joint eigenvalue problem with Frobenius action in every rank and at every root order.

comments
35 pages
fields
Algebraic Geometry, Mathematical Physics, Representation Theory
pdf ↗ page arXiv:2610.36975 ↗

The aim of this article is to study klt projective varieties such that the anticanonical bundle is semipositive from the differential-geometric point of view. The study of the holonomy of such varieties yields generalizations of several results. We show that the dimension of the space of symmetric differentials is bounded by a constant depending on the augmented irregularity, generalizing a result of Kobayashi. If we endow such a variety $X$ with a metric $ω$ whose Ricci curvature is non-negative, then we show that the holonomy of the regular locus is compact. If we assume that $X$ has canonical singularities, then we show that the general fiber of the Albanese map can be decomposed into a product of irreducible Calabi-Yau varieties, irreducible holomorphic symplectic varieties and hereditarily rationally connected varieties by the holonomy generalizing a structure theorem of Campana, Demailly and Peternell. To conclude, we show that the fundamental group of the regular locus of such varieties is virtually abelian.

comments
33 pages. Comments are welcome !
fields
Algebraic Geometry, Differential Geometry
pdf ↗ page arXiv:2610.37002 ↗

We study the large volume limit of Bridgeland stability conditions on smooth projective varieties. For each fixed ray in a suitable angular sector, we prove that an object of $\mathrm{D}^b(X)$ is Bridgeland (semi)stable in the corresponding hearts for all sufficiently large radii if and only if it is a pure $H$-Gieseker (semi)stable sheaf. In this sector, we identify the category of coherent sheaves as the limiting heart in the sense of eventual membership. We also identify a perverse coherent limiting heart in the opposite angular sector. The key ingredients are uniform HN-phase estimates for bounded families of pure sheaves, coefficient bounds for the coherent images of potential destabilizers, and a criterion forcing narrow exact sequences in the varying hearts to be bounded saturated coherent sequences.

comments
35 pages. Any comments are welcome!
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.37032 ↗

We prove that every strict Calabi–Yau threefold of Artin–Mazur height one over an algebraically closed field k of characteristic p>0 admits a projective lift to W(k). No assumption is made on the torsion in its crystalline cohomology, and p is arbitrary. Using the crystalline period map of Brantner and Taelman, we show that the obstruction to extending a lift over W_n(k) is the Bockstein of a torsion coordinate of its period point, and that this coordinate can always be made to vanish by varying the lift. The mixed-characteristic deformation functor need not be smooth: for a Calabi–Yau threefold of height one constructed by Addington and Bragg in characteristic 3, some lifts over W_2(k) do not extend over W_3(k). This disproves a conjecture of Achinger and Zdanowicz.

comments
24 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.37075 ↗

We introduce the notion of maximal geometric quotient, namely a geometric quotient of a polarized pair under a torus action which has a birational contraction to any other GIT quotient. For the action of the maximal torus on a complete flag variety, we show that a maximal geometric quotient exists and it is a Mori dream space, and its Cox ring is isomorphic to the Cox ring of the flag variety. We discuss birational properties of the maximal geometric quotients, their relations with the Chow quotients and explain how to construct them for maximal torus actions on flag varieties.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.37163 ↗

We report a major update to our original software GKMtools and record multiple improvements in the computation of Gromov–Witten invariants of GKM spaces. This includes a refined divisor equation that reduces the complexity from exponential to linear in the number of marked points, three tricks to accelerate the computation of curve class relations in the GKM graph, and an implementation of Mihalcea's algorithm for homogeneous varieties using the equivariant quantum Chevalley formula. As an application, we prove that the small quantum cohomology of the type-$E$ generalized Grassmannians $E_7/P_2,E_7/P_4,E_7/P_5,E_8/P_6$ is generically semisimple. These are precisely the remaining exceptional Grassmannians for which generic semisimplicity of small quantum cohomology was still unknown.

comments
19 pages
fields
Algebraic Geometry, Symplectic Geometry
pdf ↗ page arXiv:2610.37273 ↗

The linear relations among fixed weight multiple zeta values (MZVs) are a well-studied subject in number theory. Elements of putative linear bases for this fixed-weight space are, however, not algebraically independent; they satisfy non-linear relations. We study these relations by approximating them through motivic MZVs, formal MZVs and numerical computations. We show that motivic MZVs in fixed weight span a projective toric variety and explore its geometry. We define an outer approximation to the variety of formal MZVs in fixed weight by introducing toric varieties coming from shuffle and stuffle relations and taking their intersection. We conjecture that this intersection in fact agrees with the variety of formal MZVs and hence conjecturally also with the variety of motivic MZVs. This leads to a novel conjecture on a generating set for the extended double shuffle relations, which we checked computationally up to weight $23$.

comments
40+8 pages, 4 figures, 5 tables
fields
Algebraic Geometry, Combinatorics, Number Theory
pdf ↗ page arXiv:2610.37351 ↗

We find necessary conditions for the projectivization of a rank two vector bundle $\mathcal{E}$ on $\mathbb{P}^3_{\mathbb{C}}$ to be related to a smooth Fano 4-fold by a small $\mathbb{Q}$-factorial modification (SQM). Furthermore, we fully describe the birational geometry of $\mathbb{P}_{\mathbb{P}^3}(\mathcal{E})$ when $\mathcal{E}$ is stable, with $c_1=-1$, $c_2=2$. In this case, we prove that there exists a flip from $X^+$ to $\mathbb{P}_{\mathbb{P}^3}(\mathcal{E})$, where $X^+$ is smooth and Fano. We also determine the birational geometry and numerical invariants of $X^+$.

comments
31 pages, comments are welcome
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.37608 ↗

We use the method of equivariant localization to obtain a recursion for the cobordism-valued one-point psi-class intersections on $\overline{\mathcal{M}}_{0,n}$, and hence the cobordism classes $[\overline{\mathcal{M}}_{0,n}]$. Restricting to the hyperbolic formal group law, we recover the Poincaré polynomials of $\overline{\mathcal{M}}_{0,n}$ as a special case, and obtain new formulas for Hirzebruch $χ_y$-genera of $\overline{\mathcal{M}}_{0,n}$ twisted by tautological line bundles.

comments
25 pages, comments welcome!
fields
Algebraic Geometry, Algebraic Topology, K-Theory and Homology
pdf ↗ page arXiv:2610.37628 ↗

We introduce the $(m,p,q)$-Lefschetz property for $(1,1)$-type endomorphisms of Hodge–Lefschetz modules, providing a framework for studying nef $(1,1)$-classes and characteristic classes of nef vector bundles on compact Kähler manifolds. Under a Hall–Rado type condition on nef twisted vector bundles, we prove Hodge–Riemann relations for products of top Chern classes and, in suitable bidegrees, for products of derived Schur classes. For products of nef $(1,1)$-classes, we show furthermore that the Hall–Rado type condition is equivalent to simultaneous Hodge–Riemann relations. Our proofs build on the purity theorem and descent lemma of Cattani, Kaplan, and Schmid. We extend the descent lemma to Schur classes of nef vector bundles. We also establish formal analogs for projective bundle modules over Hodge–Lefschetz modules, with applications to tautological classes and descent on matroid Chow rings.

comments
52 pages
fields
Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.37650 ↗

Mathematical Physics34

We prove that Borodin-Gorin presentation for Heckman-Opdam $\mathfrak{gl}_n$ hypergeometric function by interlacing integrals is equivalent to two other integral presentations studied in arXiv:2508.18864. We also consider related Baxter operators and the corresponding limiting procedure from the Ruijsenaars hyperbolic model.

fields
Mathematical Physics, Classical Analysis and ODEs, Representation Theory
pdf ↗ page arXiv:2610.30304 ↗

Hermitian polynomials in the quadrature operators $x$ and $p$ are the basic ingredients of continuous-variable quantum information, and increasingly exotic ones are now routinely engineered in superconducting circuits and trapped ions. However, such a polynomial need not define unique quantum dynamics: it may define exactly one, none, or an infinite family of physically distinct ones. Such obstructions are invisible to algebraic arguments and can compromise finite-dimensional truncations, which may fail to converge or silently converge to the wrong dynamics. We develop a method, implemented as an algorithm, which decides which case occurs, and how large the family is, from the coefficients alone, with no functional-analytic input from the user. For Gaussian operators the answer is always one; for non-Gaussian ones, a regime continuous-variable quantum computing cannot do without, the other outcomes occur already for simple and physically natural Hamiltonians. We apply the algorithm to operators drawn from the recent literature.

comments
7+15 pages, 3 figures, 1 table. Code: https://doi.org/10.5281/zenodo.23565539
fields
Mathematical Physics
pdf ↗ page arXiv:2610.33545 ↗

We study the Lane–Emden equation of index $n=5$, the critical polytropic case separating finite-radius from infinite-radius self-gravitating equilibria. The scaling family $θ_λ(ξ)=λ^{1/2}θ(λξ)$ generates a neutral mode $φ_0=\tfrac12θ+ξθ'$ about the Schuster profile. Schuster's transformation and a logarithmic variable reduce the equation to the autonomous Hamiltonian system $\ddotν=\tfrac14(ν-ν^5)$ with a first integral. Besides the Schuster (separatrix) and Srivastava solutions, every orbit crossing $ν=0$ is given in closed form by a Weierstrass $\wp$-function, and the closed forms agree with direct numerical integration to about $10^{-12}$. Up to scaling and sign, the Schuster solution is the only solution regular at the center, and it separates positive solutions from solutions with infinitely many zeros. In homology-invariant variables the singular solution $θ=(2ξ)^{-1/2}$ is a center exactly at $n=5$, where its stability changes. For a distance-varying Emden–Fowler system with a shifted equilibrium we prove that at $n=5$ the unique equilibrium is linearly and nonlinearly unstable. Finally, the linearization about the Schuster profile is the Schrödinger operator $L=-Δ_r-5θ_S^4$ on $L^2((0,\infty),ξ^2dξ)$. Its Friedrichs extension has essential spectrum $[0,\infty)$, a threshold resonance at $0$ carried by the non-normalizable scaling mode, and exactly one negative eigenvalue, $λ_0\approx-1.21$, so the Schuster profile is a saddle point of the associated energy functional.

comments
41 pages, 8 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.35932 ↗

We analytically evaluate in terms of finite sums, the following definite improper integral, which appears in the quantum field theoretical calculation involving elementary charged particles having charge $\pm e$ in presence of a constant external magnetic field $\mathbf{B}=B\hat{\mathbf{z}}$ $$ I_{abln}^{μ_1\dotsμ_r}(\mathbf{q}_\perp,|eB|) = \int\!\!\dfrac{d^2k_\perp}{(2π)^2}k_\perp^{μ_1} \dots k_\perp^{μ_r} \exp \bigg(-\frac{\mathbf{k}_\perp^2+\mathbf{p}_\perp^2}{|eB|}\bigg) L^a_l \bigg(\frac{2\mathbf{k}_\perp^2}{|eB|}\bigg) L^b_n \bigg(\frac{2\mathbf{p}_\perp^2}{|eB|}\bigg)\bigg|_{\mathbf{p}_\perp=(\mathbf{q}_\perp+\mathbf{k}_\perp)} $$ where $μ_i \in \{1,2\}$, $l\ge 0$, $n\ge 0$ and $L_l^a(x)$ are the generalized Laguerre polynomials. We also provide explicit expressions of the result up to the fourth moment i.e. $r=4$, which are sufficient for most of the field theoretical calculations.

comments
Comments are welcome
fields
Mathematical Physics
pdf ↗ page arXiv:2610.35995 ↗

We solve the Schrödinger equation for a particle in the circle with a quantum mechanical bootstrap method. The main idea is that in physical situations one needs solutions of the recursion relations for the Fourier coefficients of the measure that vanish exponentially fast. We show that this results in a (generalized) continued fraction representation of the solution. The solution is an analytical function of the couplings and the energy. We also show how one this method directly computes the density of states and checking that the answer is physical is reduced to checking an inequality for a single variable. We also show how to reconstruct the wave function and how to extract the quasimomentum of the particle as a function of the energy. We illustrate our solution in the simplest case of the Mathieu equation, where we show that the method can easily resolve exponentially small bands in the band spectrum.

comments
6 pages including appendix; 2-column format; Numerical procedure implemented in Python in the ancillary Jupyter Notebook "Analytic_Bootstrap.ipynb"
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36031 ↗

We study the set of $n$-qubit pure quantum states. Specifically, we consider their incarnation as density matrices in Bloch representation. Building on prior work, we define a family of quadratic functions in Pauli expectation values, and fully characterise all instances that result in state-independent purity constraints. The resulting family of pure-state invariants is surprisingly rich in structure, with strong ties to the geometry of symplectic polar spaces. After charting the general landscape, we define a sub-family of pure-state invariants that satisfies the dynamical constraint of being stable under state update. We prove that it corresponds with (different embeddings of) symplectic Lie algebras $\mathfrak{sp}(2^{n-1})\subset\mathfrak{su}(2^n)$, which admit a generating set of Pauli operators. We carefully distinguish the symplectic case from its orthogonal counterpart, using the Arf invariant of quadratic refinements of the symplectic form on the projective Pauli group, which we further identify with quaternionic structures on $\mathbb{C}^{2^n}$ that are both generated by and restricted to Pauli operators. As a consequence of this analysis, we reveal previously largely overlooked structure in the $n$-qubit Pauli group. We highlight this in a number of examples. We characterise all state-independent uncertainty relations in variance for Pauli observables, as well as all Pauli channels for which the output fidelity and output purity are state-independent. We further show that the orbits of the projective stabiliser groups of pure-state invariants, which for $n\geq2$ are proper subgroups of the Clifford group, form complex projective state $2$-designs, while their unitary lifts form symplectic $2$-designs.

comments
44+6 pages; comments welcome
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36076 ↗

For every translation-invariant ferromagnetic Ising chain with $n^2J_n\to B>0$, we prove $2β_cBm_c^2=1$, without restrictions on the finite core or the tail convergence rate. The proof bounds the energy of macroscopic interfaces and rules them out if the critical product exceeds one. Despite the nonzero jump, connected susceptibility is infinite both at criticality and in a neighboring ordered interval. At the critical coupling, the excess magnetization decays more slowly than any positive power of the field as the field tends to zero. We also prove the jump relation in continuous time, for a positive-Stieltjes class of Ohmic spin-boson models.

comments
24 pages, 1 figure
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36131 ↗

We present a self-contained formulation of longitudinal motion in one-dimensional second-gradient continua. Starting from finite-strain kinematics, we then apply the Principle of Virtual Power to identify the governing field equation and contact actions, derive the balance laws, and establish the mechanical-energy identity for hyperelastic materials. Motivated by pantographic microstructures, we study a class of finite-strain energies formulated in terms of normalized strain measures and deformation-dependent strain-gradient moduli. For the associated equation of motion, a traveling-wave first integral yields a turning-point criterion for the existence of solitary waves with positive stretch, exponential decay, and finite mechanical energy, as well as uniqueness at each fixed speed within a prescribed strain interval. For suitable spring energies, a change of strain variable reduces this integral to a polynomial equation, yielding explicit tensile and compressive profiles for Green–Lagrange and Hencky strains under appropriate parameter and kinematic conditions. Under the positivity condition on the quartic potential, the constructed Hencky energy has several physically natural properties: it is nonnegative, admits the natural state as its unique zero and global minimizer, has a positive linearized modulus, penalizes stretch gradients, and diverges under both material collapse and unlimited extension. Furthermore, the stored energy is coercive with respect to the $H^1$ norm of logarithmic strain. Together with mechanical-energy conservation, this yields uniform upper and positive lower bounds on stretch throughout the lifespan of smooth solutions. The theory's accessible mechanical origin, relatively simple mathematical foundation, and diverse solitary waves offer a tractable entry point to studying the rich nonlinear dispersive equations governing higher-gradient elastic continua.

comments
28 pages, 7 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36198 ↗

The modern approach to modeling equilibria in stellarators features two blind spots: lack of foundational agreement between theory and computation, and inability to reliably explore economically-attractive high-$β$ concepts. These blind spots originate at a technical level from lack of a useful existence theory for the partial differential equations believed to describe equilibria, and from the encumbrance of crude, inflexible models for the topology of the confining magnetic field. This Article proposes the first stellarator equilibrium model that rigorously guarantees existence of an equilibrium with any chosen magnetic topology. It completely separates determination of the magnetic topology from the equilibrium existence question, and thereby enables importing realistic tearing physics into equilibrium computations. The new model replaces the standard plasma potential energy functional with its ensemble average over rapid kinetic-scale fluctuations. Averaging ``elasticizes" the potential by energetically penalizing separation of field lines. The elasticized potential's super-linear dependence on plasma strain embodies a phenomenological model for profile stiffness, and plays an essential role in the existence theory.

comments
17 pages, 3 figures, v1
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36241 ↗

These lectures provide an introduction to modern techniques for studying the analytic properties of loop-level scattering amplitudes and correlation functions in perturbative quantum field theory. We begin by developing the ideas behind Landau analysis, which identifies the location and nature of kinematic singularities in Feynman integrals (the universal building blocks that contribute to perturbative amplitudes and correlators). We then show how fundamental principles such as unitarity and causality constrain the analytic structure of Feynman integrals, and develop methods for working out some of these constraints in practice. Finally, we connect these observations to the properties of iterated integrals and their symbol calculus, to show how bootstrap methods can be used to determine the functional form of Feynman integrals from just knowledge of their analytic structure. These notes are based on a series of lectures held during the La Ricotta Summer School "The Disordered Universe 2025" from 21~July to 1~August~2025.

comments
64 Pages, 24 Figures, 2 Tables. A chapter to be published by Springer Lecture Notes in Physics
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36271 ↗

We develop a quantum hydrodynamic theory based on Nambu's unpublished note \emph{A New Look at Fluid Dynamics}. The kinetic action defines a metric on configuration space and a mechanical connection coupling an observable cyclic deformation to unobserved fluid modes. Tracing out these modes yields a quantum channel for the observed deformation. We establish a criterion for the exact preservation of coherence between opposite momenta and show how nonzero connection curvature prevents this coupling from being removed. The leading quadratic decay of quantum coherence is determined by the prepared quantum state. For two shear flows, we identify an exact noiseless sign subsystem. Introducing additional generating flows cancels the initial residual of the Euler equations while producing nonzero connection curvature, governed by the same kinetic integrals that determine the classical correction. At fixed mean energy, choosing amplitudes adapted to the dominant coupling flow substantially reduces the initial decoherence, even along the direction that cancels the Euler residual. This advantage gradually diminishes under conditional evolution, eventually leading to a crossing of the decoherence curves. Our example shows that a preparation with a smaller initial coherence-decay coefficient can have a larger maximum sampled loss over a finite readout interval.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.36275 ↗

We study the many-body boson star equation in the mean-field regime. Under a decay assumption on the solution to the effective one-body equation, we prove a uniform-in-time trace-norm estimate comparing the one-particle reduced density matrix of the many-body state with the projection onto the one-body solution. This establishes uniform-in-time mean-field convergence as the particle number goes to infinity. Furthermore, for the many-body dynamics, we establish both a maximal velocity bound, which can be seen as a Lieb–Robinson-type bound, and a minimal velocity bound. As an application, we recover maximal and minimal velocity bounds for the one-body boson star equation by passing to the mean-field limit.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.36388 ↗

We perform an asymptotic analysis of the biased van der Pol equation $\ddot{x} - ν\,(1 - x^{2})\,\dot{x} + x = a$, where the dissipation parameter is large $(ν\gg 1)$ and the bias parameter $a$ is a real number between $-1$ and $1$, corresponding to large-amplitude relaxation oscillations along a stable limit cycle. Using an abridged singular perturbation analysis, the period of relaxation oscillations $T(a,ν)$ is obtained that is accurate within $0.1\%$ of the numerical results for $ν\gg 1$.

comments
9 pages, 8 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36398 ↗

The Buice-Cowan field theory [PRE 75, 051919 (2007)] describes fluctuations and correlations of neural activity through a path integral whose mean-field limit is the Wilson-Cowan equation and whose critical behavior belongs to the directed percolation class. In its continuum limit the neurons are the points of a Euclidean space and activity spreads by ordinary diffusion, which encodes neither the hierarchical organization of cortical networks nor their small-world property. We formulate a version of the theory in which the neurons are organized in an infinite, self-similar, tree-like structure, described by the p-adic numbers. The path integral carries over unchanged, and finite networks are exact restrictions of the continuum theory. The main change is in the linear response: in the Euclidean theory every short-range kernel reduces to the Laplacian, while in the p-adic theory each connectivity kernel gives its own nonlocal p-adic diffusion operator. For couplings that decay by a fixed factor per level of the hierarchy this is the Taibleson-Vladimirov operator, and the critical theory is an ultrametric Reggeon field theory whose upper critical dimension is twice the decay exponent of the coupling. Hence the decay of the coupling up the hierarchy, and not the dimension of space, decides whether mean-field exponents apply. We obtain a Ginzburg-type criterion, loop-corrected hierarchical Wilson-Cowan equations, log-periodic corrections to scaling and power-law correlations. Exact simulations on a tree support the predictions: the avalanche-size exponent takes the mean-field value 3/2 when the coupling decays slowly and decreases toward the directed-percolation value when it decays fast. The hierarchical network is also small-world when the decay exponent is smaller than the dimension, a property that the Euclidean theory obtains only by assuming long-range coupling.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.36400 ↗

Two new Bohr-Sommerfeld (B-S) type quantization conditions for $d$-dimensional radial potentials are proposed for non-singular and singular potentials, respectively. The power-like potentials \hbox{$V=\operatorname{sign}(m)\,r^m$} for $m=-3/2,-1,-1/2, 1,3,4,6$ and $V=\log r$ (that corresponds to $m=0$) are studied in detail for radial quantum numbers $n_r \in [0,100]$ and for angular momenta $\ell=1,2,3,4,5$ in dimensions $d=2,3$. It is shown that for large $n_r$ the B-S energies $E_{BS}$ always tend to the exact ones. For $m=2$ (the radial harmonic oscillator) and $m=-1$ (the $d$-dimensional analogue of the $3D$ Coulomb potential) the B-S spectrum coincides with the exact one for the first and second quantization conditions, respectively. In general, up to a small parametric shift of the radial quantum number, the analytic formulas for the energy spectrum can be written for arbitrary power-like potential. A comparison of $E_{BS}$ with the energies $E^{(L)}_{BS}$, obtained for the radial potentials modified by the Langer correction for $d=3$, are in good agreement in general: in detail, it was checked for the quartic oscillator $V=r^4$. It is demonstrated that introducing a dimensionless {\it WKB correction} $γ$ into the r.h.s. of the proposed B-S type quantization condition allows us to reproduce the exact energies with $γ$ being very small for positive degrees $m \geq 0$ and sufficiently small for non-positive ones $m < 0$ for all $n_r, \ell$ studied.

comments
39 pages, 6 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36421 ↗

Computationally expensive simulations and real-world observations often provide data at a lower temporal resolution than downstream tasks require. Averaging observations at fixed grid points can introduce errors, and motion estimates alone do not determine how the physical field values change. In this work, we study whether learning to correct a simple numerical interpolation improves the reconstruction of intermediate scientific fields. Our model, interpolative FLow EXpert (iFLEX), first estimates the missing field by taking a weighted average of the two observations at each grid point, giving more weight to the observation closer to the requested time. During training, subtracting this estimate from an observed intermediate field gives the residual that a conditional diffusion model learns to generate. Both observations condition the model, together with the target time, the interval between observations, and available physical parameters. To reconstruct an unobserved intermediate field, the generated residual is added to the weighted average. We evaluate iFLEX on Navier-Stokes Kraichnan turbulence, Shanghai radar precipitation, and sea surface temperature. Across these datasets, iFLEX achieves lower reconstruction error than the evaluated baselines, including optical-flow and diffusion models developed for video frame interpolation in computer vision. In a controlled turbulence comparison at a Reynolds number of 36,000, learning the residual reduces relative Frobenius norm error from about 13.7 for direct field generation to 10.5, with the diffusion backbone and training setup held fixed. Code is available at https://github.com/paulpuren/iflex.

comments
Accepted by Transactions on Machine Learning Research
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36751 ↗

In a recent paper~\cite{BGW}, Bizoń, Glogić, and Wasserman (BGW) reduced the classification of smooth self-similar Yang–Mills profiles in odd spatial dimensions $d\ge11$ to an algebraic problem of counting roots of an explicit polynomial in a prescribed interval. Two roots correspond to explicit profiles. BGW conjectured that there is exactly one additional root and verified this in a range of odd dimensions. Here, we prove the existence of an additional root, and hence of a third smooth self-similar profile, in every odd dimension $d\geq11$. The proof combines an endpoint sign calculation for the resonance polynomial with an algebraic argument showing that the third root is distinct from the two explicit ones. Uniqueness remains open in general.

comments
10 pages
fields
Mathematical Physics, Analysis of PDEs
pdf ↗ page arXiv:2610.36784 ↗

We consider nonequilibrium chaotic states of classical interacting particles in a family of scaling limits motivated by kinetic theory. The particle activity and the strength of the pair interaction are rescaled simultaneously, allowing for both dilute and non-dilute regimes. Using cluster expansion, we derive pointwise bounds for truncated correlation functions (cumulants) of arbitrary order. We cover stable finite-range interactions and repulsive Riesz potentials, including non-integrable tails in bounded macroscopic domains.

comments
Funded by the European Union (ERC CoG KiLiM, project number 101125162). 32 pages, 7 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36788 ↗

By using suitable ansatz which reduces the problem of obtaining exact solutions of the sine-Gordon equation with nontrivial time dependence to obtaining exact solutions of two first order symmetric $φ^4$-like coupled equations in $x$ and $t$, we obtain a large number of hyperbolic and periodic solutions of the sine-Gordon equation. In particular, we not only obtain the well-known two-soliton, soliton-antisoliton and the breather solutions but also obtain periodic two-soliton, periodic soliton-antisoliton as well as periodic breather solutions. Besides, we also obtain superposed real and complex (parity-time reversal or) $\mathcal{PT}$-invariant solutions related to the two-soliton, soliton-antisoliton and the breather solutions of the sine-Gordon equation.

comments
9 Figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.36891 ↗

We study statistics of transport properties in a system of two quantum chaotic cavities coupled in series through a tunable junction and connected to reservoirs with asymmetric leads. Within the random matrix theory framework and in the absence of time-reversal symmetry, the scattering matrix S undergoes a crossover from two independent circular unitary ensembles to a single CUE as the inter-cavity coupling increases. Modeling this crossover using Dyson's Brownian motion approach, with the junction width acting as a time-like parameter, we derive closed-form expressions for the time-dependent averages and variances of conductance and shot noise power for arbitrary lead asymmetry. These analytical results reduce to known expressions in the symmetric and fully coupled limits. The mapping between the Brownian time parameter and the physical junction width is found to be observable-dependent, reflecting the different eigenfunction correlations probed by conductance and shot noise power, and this dependence is examined and separately calibrated for each observable. Validating against random-matrix simulations and tight-binding transport calculations using Kwant, we find quantitative agreement for ensemble-averaged observables throughout the crossover, while variances, being more sensitive to microscopic fluctuations, show deviations from the Brownian-motion predictions.

journal
J. Phys. A: Math. Theor. 59 (2026) 415201
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37090 ↗

We compare two three-particle lattice models on the square lattice at total quasimomentum zero: three identical bosons with attractive contact interactions, and a two-fermion plus one distinguishable particle system with attractive interspecies contact. Using the Birman–Schwinger principle combined with symmetry reduction and strong-coupling asymptotics, we distinguish shared methodology from model-dependent spectral conclusions. For bosons, the ground-state energy behaves as minus three times the coupling plus six minus three over twice the coupling plus higher-order corrections, and exactly two bound states exist below the threshold for large coupling; the second lies at a finite distance 0.03542 below the threshold, a genuine eigenvalue rather than a threshold resonance. For fermions, the critical mass ratio is pi divided by pi minus two, approximately 2.75194; above it, a doubly degenerate level appears at strong coupling. Near onset, the binding energy follows a linear-over-logarithmic law. The results are compared with numerical finite-grid Faddeev computations and with recent three-dimensional classifications of threshold states.

comments
16 pages, 1 figure
fields
Mathematical Physics, Spectral Theory
pdf ↗ page arXiv:2610.37142 ↗

We construct Lax pairs with a spectral parameter for a family of models of interacting tops associated with the $\rm{BC}_n$ root system. The construction uses the parametric $\rm{BC}_n$ associative Yang–Baxter equations, which involve both quantum $R$-matrices and boundary $K$-matrices satisfying the reflection equation. Special cases yield spin generalizations of the elliptic Calogero–Inozemtsev model. We also show how quantum Hamiltonians of interacting tops, combined with the Polychronakos freezing prescription, reproduce recently introduced anisotropic quantum long-range spin chains of Haldane–Shastry–Inozemtsev type.

comments
77 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37165 ↗

Vom Ende, Chruściński, Kimura and Muratore-Ginanneschi proved that every 2-positive trace-preserving map $Φ$ on $M_d=M_d(\mathbb{C})$ satisfies $\operatorname{Tr}Φ\le d\,m(Φ) + d^2 - d$, where $m(Φ)$ is the smallest real part of an eigenvalue of $Φ$, and deduced the universal relaxation-rate bound $\operatorname{Tr} L \le d\,m(L)$ for generators of 2-positive trace-preserving semigroups. They asked whether trace preservation is needed, and conjectured that every conditionally 2-positive map $L$, i.e. every $L$ for which $e^{tL}$ is 2-positive for all $t \ge 0$, satisfies $\operatorname{Tr} L \le d\,m(L) + (d^2 - d)\,s(L)$, where $s(L)$ is the largest real part of an eigenvalue of $L$. We prove this conjecture. The argument normalizes a 2-positive map by a positive definite eigenmatrix of its Hilbert-Schmidt adjoint, which turns it into a 2-positive trace-preserving map in the manner of a quantum Doob transform. The existence of such an eigenmatrix is guaranteed after adding an arbitrarily small multiple of the unnormalized depolarizing map, and the inequality then passes to the limit by continuity of the spectrum. No irreducibility, diagonalizability or faithful-invariant-state assumption is required. We also show that the constant $d$ is optimal within the class of non-trace-preserving completely positive semigroups, and we recast the result as a universal constraint on decay rates measured relative to the dominant spectral growth rate.

comments
9 pages, 1 table. Preprint archived at doi:10.5281/zenodo.23225994; supporting verification code archived at doi:10.5281/zenodo.23226133
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37187 ↗

We analyze the on-shell content of the local Lorentz covariants of the five-dimensional spin-zero Duffin-Kemmer-Petiau algebra in elastic spin-zero-spin-$1/2$ scattering. The scalar, vector, and tensor structures are reduced explicitly to the two independent amplitudes allowed by the physical parity-conserving process. The two scalar structures are proportional on shell, the genuine nonminimal vector $i[P,β^μ]$ is longitudinal and vanishes after contraction with the conserved fermion vector current, and the projected vectors $Pβ^μ$ and $β^μP$ each reduce to one half of the minimal-vector contribution. The tensor covariants introduce no additional on-shell directions. We discuss the consequences for the four invariant choices used in meson-nucleus optical-potential calculations and show that different nuclear equations can arise only after on-shell-equivalent representatives are promoted to optical operators and iterated beyond the order at which the elementary impulse approximation is fixed. These results provide a systematic basis for distinguishing genuinely independent elementary amplitudes from representation-dependent effects generated by the optical embedding.

comments
8 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37263 ↗

We derive explicit overlap functions for integrable matrix product states associated with arbitrary finite-dimensional irreducible evaluation representations of the orthogonal and symplectic twisted Yangians. The central step is to relate the commuting quasi-determinants entering the universal overlap formula to Sklyanin determinants of nested subalgebras. For orthogonal evaluation representations, this relation determines their spectra directly in the Gelfand–Tsetlin basis. In the symplectic case, we combine Sklyanin determinants and minors with a factorization of the nested rank-two boundary matrices to obtain the corresponding spectra. The resulting overlap functions are explicit products parametrized by Gelfand–Tsetlin patterns and require only the highest-weight data of the underlying orthogonal or symplectic representation. Together with the universal ratio of Gaudin determinants and the model-dependent pseudovacuum overlap coefficients, they determine the normalized MPS overlaps. We also express the overlap functions in the orthogonal case beyond evaluation representations in terms of the highest weights arising along the nested twisted-Yangian subalgebras.

comments
27 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37285 ↗

Resistive evolution of a three-dimensional magnetostatic equilibrium requires compatibility among induction, momentum balance and thermal transport. We construct exact thermoviscous evolutions of a non-axisymmetric magnetic torus with nonzero magnetic shear using a path-dependent entropic Lagrangian (PDEL) formulation. A symmetric positive-definite resistivity tensor is constructed so that the resistive differential expression preserves the spatial form of a known sheared toroidal equilibrium. Homogeneous expansion is coupled to an ideal gas and positive bulk viscosity in the absence of body forces and volumetric heat supply. For a one-temperature gas with heat-capacity ratio $5/3$, a $T^{-2/3}$ resistivity prefactor gives elementary time dependence for all fields. A two-temperature extension admits a $T_e^{-3/2}$ prefactor and positive interspecies heat exchange on a finite state interval. Material magnetic flux decays while the rotational-transform profile is preserved. Both branches satisfy local total-energy conservation, moving-boundary power balance and material transport of dissipation histories. A seed-specific obstruction excludes positive scalar resistivity along the same homologous trajectory; the two-temperature exchange law has a nonremovable endpoint pole. Symbolic identities, independent Cartesian differentiation and surface–volume quadrature verify these results. The solutions provide coupled MHD benchmarks with passive, inverse-designe

fields
Mathematical Physics
pdf ↗ page arXiv:2610.37305 ↗

We formalize in Lean 4 the gauge closure of the Feynman functional integral for the free field without Wick rotation. First, the restriction to box divisions is removed via a direction bound and a uniform Riemann-sum estimate for all $δ$-fine cylindrical divisions. Second, the coordinate measure on $\mathbb{R}^k$ is identified with the standard Gaussian one. Third, the Saks-Henstock lemma and additivity are proved. Fourth, the Fresnel formula $\int_{\mathbb{R}} e^{i a x^2/2} dx = \sqrt{2π/a} e^{iπ/4}$ is proved as an improper (gauge) integral. This yields the projective consistency of the oscillating Gaussian premeasure and the existence of the free-field integral. The free field is completely closed: we prove the Fresnel formula for an arbitrary symmetric positive-definite action $A$, compute the generating functional and the two-point function $\langle y_p y_q \rangle = i(A^{-1})_{pq}$, and prove Wick's theorem. The interacting field $λ\varphi^4$, unbounded generators, and renormalisation remain open. All results are machine-verified with no sorry or new axioms.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.37346 ↗

In this letter, we give necessary and sufficient conditions for the real symmetric matrices $A_1 , \ldots , A_m$ to ensure they are simultaneously diagonalizable via congruence.

comments
5 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37355 ↗

Exact normalizable solutions of quantum constraints form a symplectic submanifold of the space of rays, whereas classical first-class constraints define a surface on which the symplectic form is degenerate. We ask how the degenerate directions can be recovered when quantum theory is regarded as fundamental and the constraints are imposed through a single positive master operator. For a nonnegative master function whose zero set is a Morse–Bott critical manifold, the image of the linearized Hamiltonian vector field is the symplectic orthogonal of the zero set, and coisotropy is equivalent to this linearization having square zero. The image is fixed by the zero set, but the theorem identifies it with the response of the master dynamics, which makes it accessible to quantum states. Three realizations follow. The exact quantum physical space is the opposite, rotational case of the same dichotomy. For quadratic masters, the bottom of the spectrum is $(\hbar/2)\Tr_+L$, so first-class zero sets carry no offset of order $\hbar$; within the harmonic approximation this explains the $\hbar$ versus $\hbar^2$ offsets observed in tests of the master-constraint programme. Finally, the second time derivative of the limiting covariance of a semiclassical packet, which can be determined by bounded probes, has the gauge distribution as its image. Two oscillators at fixed total energy give exact master evolution, an exact realization of this covariance shear, and the reduced symplectic limit. For the non-Abelian SL(2,$\R$) model we compute the coherent and Weyl master symbols, prove essential self-adjointness and a finite-time propagation estimate, and derive the spectral bottom $\hbar^2/2$.

comments
29 pages, 2 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37493 ↗

We construct a classical reflected phase space for open Carroll–Weyl null strings with Dirichlet conditions at the target origin or linear mixed target gluings. The reflection-compatible constraints are differentiable, their Hamiltonian flows preserve the boundary domain, and the constraint system is weakly first class on its regular part. The induced boundary algebra extends the boundary Carrollian conformal algebra by an even weight-one Carroll–Weyl current, with normalization fixed by the reflected symplectic form. The regular local embedding count agrees with that of the closed theory. We exhibit a four-dimensional mixed sector with one canonical field pair and a separate solution family carrying a nonzero conserved target-space boost charge and a nondegenerate symplectic pullback. Smooth fixed-point reflection selects characteristic endpoints. For timelike defects with natural tangential variations and no additional boundary interactions, non-characteristic endpoints have vanishing Carroll flow. For displaced Dirichlet data, we derive the canonical affine lift and its required shift of the edge momentum by integrated bulk momentum.

comments
30 pages, 3 appendices
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37546 ↗

Quantum field theory unifies special relativity and quantum mechanics, yet relies on externally imposed prescriptions for quantization, renormalization, and probabilistic interpretation that lack a common algebraic foundation. This work derives these features from a single principle: the finite-resolution realization of the additive group (R, +) under four consistency constraints: finite operational resolution (FR1), multiplicative linearity (FR2), phase coherence (FR3), and statistical typicality (FR4). The Hilbert-space formalism of quantum theory, the Born probability rule, and the Kolmogorov structure of statistical mechanics emerge as limiting projections of a common additive algebra, without additional postulates. Extending this framework to spacetime translations and their automorphisms, we recover the kinematic and dynamical structure of quantum field theory and address ultraviolet divergences through intrinsic finiteness rather than external renormalization. Quantization and unitarity emerge from additive symmetry, while the gravitational field is interpreted as the algebraic kernel transporting boundary information into bulk observables, without requiring an independent quantization of the metric. This framework aims to provide a unified, finite, and algebraic foundation for physical law.nite, and algebraic foundation for physical law.

journal
int. J. Geom. Methods Modern Physics, 2026
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37556 ↗

We consider three-dimensional Bose gases with attractive two-body interactions and repulsive three-body interactions. We prove the stability of the second kind of the system for a large class of interaction potentials and scaling regimes, which includes the Gross–Pitaevskii regime for the three-body potential. In addition, we even prove the classical stability of the system for a more restrictive class of interactions and scaling regimes. The main ingredient of the proof is a bound that was derived recently in a study by the authors on the dynamics of Bose gases with three-body interactions in the Gross–Pitaevskii regime (2026).

comments
17 pages, 4 figures. Comments welcome!
fields
Mathematical Physics
pdf ↗ page arXiv:2610.37565 ↗

We develop a continuum-thermodynamic model for chemically reacting, multi-velocity and multi-temperature fluid mixtures. Partial balances of mass, momentum and total energy are closed using a single entropy density depending jointly on the constituent mass and nonkinetic internal-energy densities. Its partial derivatives determine the constituent temperatures and chemical potentials. For the chosen entropy flux and separately treated volume deformation, the strengthened entropy principle yields a weighted Euler relation constraining the partial thermodynamic pressures. We construct thermodynamically consistent constitutive relations for viscosity, heat conduction, Maxwell–Stefan momentum exchange, thermal relaxation and nonlinear reaction kinetics. The baseline model includes thermo-diffusion and its reciprocal coupling; a fully coupled reciprocal completion is given within the stated isotropic co-factor class. An alternative formulation couples chemical conversion and energy redistribution through specified elementary processes. We identify the material functions required for closure and admissible shifts of constituent energy zeros. An ionization network with resolved electronic levels yields Saha-type ionization relations and Boltzmann excitation ratios under the stated detailed-balance assumptions, recovering the Saha–Eggert relation at common temperature. Competing electron- and heavy-particle processes can maintain stationary populations while transferring energy between groups at different prescribed temperatures.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.37637 ↗

Central to non-Hermitian sensing is the question of whether exceptional points (EPs) can make weak perturbations easier to detect. We show that despite enhanced spectral response, in finite-dimensional Lindblad systems with a unique stationary state and a nonzero relaxation gap, EPs offer no fundamental advantage in how the observation time required for reliable detection of a weak perturbation scales with the perturbation strength. We nevertheless demonstrate that by selectively choosing perturbations that enhance the quantum Fisher information rate, EPs can shorten the observation time needed to reach a predefined detection accuracy relative to a comparable non-EP reference at the same decay rate. As an illustration, we show using an optomechanical model at an EP that we can detect a weak force through the cavity output at the same reliability in less time than a non-EP reference. Our results establish fundamental limits on perturbation detection at EPs and can guide the design of faster non-Hermitian sensors based on optomechanical resonators, superconducting qubits, and trapped ions.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.37660 ↗

Number Theory33

A fusion ring is said to be vanishing if one of its multiplication matrices has determinant zero. We classify all vanishing fusion rings of rank four. We then provide a more explicit classification in the integral case, which leads to new consequences: there are infinitely many nonisomorphic simple integral vanishing fusion rings of rank four and no simple integral fusion ring of rank four is 1-Frobenius. We also classify all Burnside fusion rings of rank four.

comments
17 pages, Theorem 1.8 has been strengthened (we now classify all nonpointed Burnside fusion rings of rank four), and streamlined Section 3. Comments are welcome!
fields
Number Theory, Mathematical Physics, Quantum Algebra, Rings and Algebras, Representation Theory
pdf ↗ page arXiv:2610.22000 ↗

Hofstadter's G function is recursively defined via G(0) = 0 and then G(n) = n-G(G(n -1)). Following Hofstadter, we vary the number k of nested recursive calls in this equation and obtain a family of functions (F\_k). In a previous article, we proved that this family is ordered pointwise: for all k and n, we have $F_k(n) $le$ F_{k+1}(n)$. Here we propose a much shorter proof of this theorem, without infinite morphic words. The key ingredient for this direct proof is a new formula expressing iterates of the preimages of $F_k$ . Thanks to this formula, we also prove several conjectures mentioned in our previous article. In particular we obtain that $F_k(n) < F_{k+1}(n)$ when n > (k + 1)(k + 6)/2. We provide a certified formalization of all these results in the Rocq proof assistant.

fields
Number Theory
pdf ↗ page arXiv:2610.35994 ↗

We give an asymptotic formula for the number of Markov triples with largest entry below $x$. We show that the error term is $o(\log x)$, which improves upon the previous best known results due to McShane and Rivin. We give an explicit error based on classical zero-free regions of the Riemann zeta function, and a sharper estimate under the quasi-Riemann hypothesis. The proof uses the correspondence between Markov triples and simple closed geodesics on the modular punctured torus and a careful analysis of the McShane-Rivin unit ball.

comments
21 pages, 5 figures. All comments welcome
fields
Number Theory, Geometric Topology
pdf ↗ page arXiv:2610.36069 ↗

We construct uniform analytic models of quasi-unipotent differential modules over Robba rings. For modules of rank at most $r$ and highest ramification break at most $b$, the inner radius of the model annulus depends only on $r,b$ and the residue characteristic $p$. At each radius, the model admits a basis, possibly depending on the radius, in which the connection matrix is bounded by a constant depending only on $r,b,p$. Both bounds are independent of the coefficient field, and these analytic models require no coefficient extension. A prescribed absolute Frobenius structure extends to such a bounded model with the same radius and connection bound. After a finite coefficient extension, we also realize quasi-constant modules, which carry the finite local monodromy, as differential direct summands of sums of pushforwards of rank-one modules, with uniformly bounded cover degrees and highest breaks of the inducing modules.

fields
Number Theory, Algebraic Geometry
pdf ↗ page arXiv:2610.36130 ↗

Let $Φ_n(q)$ denote the $q$-deformation (in the sense of Morier-Genoud & Ovsienko) of the metallic number $φ_n=\frac{n+\sqrt{n^2+4}}{2}$. This is a power series with integer coefficients. We prove that the Hankel continued fraction of the $(n+2)$-shifted series $Φ_n^{(n+2)}$ is eventually periodic, and that the corresponding Hankel determinant sequence $Δ^{(n+2)}$ is $2n(n+1)$-periodic if $n$ is even, $2n(n+1)$-antiperiodic if $n$ is odd, and that it takes only the values $-2,-1,0,1,2$. This confirms part of a former conjecture of ours and complements similar results we obtained for all $\ell$-shifts of $Φ_n$ with $0\le\ell\le n+1$. In addition, we formulate a new conjecture of quasi-polynomiality for higher shifts $\ell\ge n+3$. The determinant values arise as telescoping products of the partial numerators of the Hankel continued fraction. Most of the technically involved and lengthy verifications are carried out by a symbolic computation program, whose code is publicly available on GitHub for independent verification.

fields
Number Theory, Combinatorics
pdf ↗ page arXiv:2610.36132 ↗

We computationally investigate a weight 3 analogue of the relationship between higher Bloch groups, algebraic K-theory, higher regulators, and special values of Dedekind zeta functions. In particular, for F a number field F, we define the pruned Bloch group B _3(F) and compare the covolume of its regulator lattice with zeta_F(3). We define a quantity kappa_3(F), which is conjecturally an integer, that is expected to be closely related to the order of K_4(O_F).

fields
Number Theory, K-Theory and Homology
pdf ↗ page arXiv:2610.36136 ↗

Ranjan and Singh proved that for every composite $n$ which is not twice an odd prime, $\varphi(n)$ is at least one plus the sum of $\varphi$ over the prime divisors of $n$, with equality exactly for $n=4,12,30$. \new{We study the weighted equation $\varphi(n)=c+α\sum_{p\mid n}\varphi(p)$. We determine the parameters $(c,α)$ admitting infinite families of solutions, prove that all remaining solutions are finite in number and arise from an explicit linear mechanism, and compute them exactly in a range of parameters; the finiteness of $\{4,12,30\}$ becomes a special case. We also settle the analogous equations for the divisor functions $τ$ and $σ$ and for the Jordan totients, and, from known results on the sum of the prime divisors, derive the full asymptotic expansion of the average deficiency.} \K{In the range $|c|\le10$, $1\leα\le6$ the composite solution set is determined completely\V{, and the number and the size of the sporadic solutions are shown to be bounded by polynomials in $|c|+α$}.} \B{Finally, replacing the prime values $\varphi(p)$ by their iterates $\varphi^{(k)}(p)$, we determine all solutions of $\varphi(n)=1+\sum_{p\mid n}\varphi^{(k)}(p)$: the solution sets are $\{1,4,12,30\}$, $\{1,3,4,10\}$ and, for every $k\ge3$, $\{1,3,4\}$; for the weighted iterated equation we exhibit an infinite family governed by the safe primes.

comments
18 pages , 6 tables
fields
Number Theory
pdf ↗ page arXiv:2610.36260 ↗

A positive integer $a$ is called \emph{retriangular} if there is a positive integer $b$ such that $aT+b$ is a triangular number whenever $T$ is. We show that the retriangular numbers are exactly the odd squares greater than $1$, and that for each of them $b$ is unique. The proof rests on a rigidity principle: a non-constant affine map carrying the values of one quadratic sequence into the values of another is induced by an affine map of the plane sending the vertex of one parabola to the vertex of the other. We use it to classify all affine maps between sets of polygonal numbers; the $s$-gonal numbers map affinely into the $t$-gonal numbers if and only if $M_t\mid M_s$, where $M_s=2(s-2)/\gcd(s,4)$ defines a bijection from $\{3,4,5,\dots\}$ onto the positive integers. We also treat sparse data, polynomial maps, figurate numbers of higher degree and sums of two triangular numbers, and we study the monoid of affine self-maps of an arbitrary set of integers. For the $s$-gonal numbers this monoid is an arithmetical congruence monoid, and the factorization of self-maps into irreducible ones is unique only for triangular and square numbers.

comments
19 pages, 2 figures
fields
Number Theory
pdf ↗ page arXiv:2610.36284 ↗

Given a smooth proper family of varieties over some $p$-adic field, if (sufficiently many of) the closed fibers have good reduction, does the family as a whole have good reduction? This question is asked and positively answered in recent work of Cadoret-Tamagawa for curves in general, and abelian varieties if the base field has absolute ramification at most $p-1$. In this paper we define a general framework for studying this question, yielding new positive answers for families of K3 surfaces and cubic fourfoulds, again with ramification at most $p-1$. Our strategy is to study several extension properties of moduli spaces and how they transfer under various morphisms. Of particular importance are integral canonical models of Shimura varieties, and whether or not their extension properties are preserved under ramified scalar extension. Additionally, building on earlier counterexamples by Raynaud-Ogus-Gabber and Vasiu-Zink, we show that the ramification assumption is optimal for the result of Cadoret-Tamagawa on families of abelian varieties. This reflects the fact that the extension property of Siegel Shimura varieties no longer holds over a field of absolute ramification at least $p$.

comments
44 pages. Comments are very much welcome!
fields
Number Theory, Algebraic Geometry
pdf ↗ page arXiv:2610.36291 ↗

We prove part of the global Gross–Prasad conjecture for $G = \mathrm{SO}_{n} \times \mathrm{SO}_{n-1}$ and $H = \mathrm{SO}_{n-1}$ in the unramified de Rham geometric Langlands setting proposed by Ben-Zvi–Sakellaridis–Venkatesh.

comments
22 pages
fields
Number Theory, Algebraic Geometry, Representation Theory
pdf ↗ page arXiv:2610.36377 ↗

Let $c>1$ be irrational. We prove that, for almost every $x>0$ and every $\varepsilon>0$, there are infinitely many triples $(p,r,s)$, with $p$ and $r$ prime and $s$ an integer, for which \[ 0<px-r<p^{-1/3+\varepsilon},\qquad 0<pcx-s<p^{-1/3+\varepsilon}. \] This improves the exponent $1/5$ in the homogeneous-line case of a theorem of Baier and Ghosh. The arithmetic input is the quantitative prime-counting lower bound in Matomäki's proof concerning primes in shrinking neighborhoods of an irrational rotation. We obtain a local first-moment lower bound by fixing the prime numerator and applying the prime number theorem to the admissible prime denominators. Two Selberg sieve squares, together with explicit trigonometric interval approximations, give a pointwise upper bound whose error tends to zero in $L^1$ after normalization. A truncation and density-point argument then yields the almost-everywhere conclusion.

fields
Number Theory
pdf ↗ page arXiv:2610.36380 ↗

Given an integer $n\ge 2$ and points $\mathbf{x}, \mathbf{y}$ lying in the $n$--dimensional torus $\mathbb{T}^n$ equipped with its usual metric denoted by $\left\langle\;\cdot\;\right\rangle$, let \[ τ_n(\mathbf x,\mathbf y) = \sup\left\{ τ\ge0: \begin{array}{l} \text{there exist infinitely many } H\ge1\text{ for which}\\[1mm] \|A\|\le H \text{ and } \langle A\mathbf x-\mathbf y\rangle\le H^{-τ} \text{ for some } A\in\mathrm{SL}_n(\mathbb Z) \end{array} \right\} \] be the Diophantine exponent of asymptotic approximation of $\mathbf{y}$ by the orbit of $\mathbf{x}$ under the action of the integral special linear group $\mathrm{SL}_n(\mathbb Z)$. Let also \[ \widehatτ_n\!\left(\mathbf{x}, \mathbf{y}\right) = \sup\left\{ τ\ge0: \begin{array}{l} \text{for all but finitely many } H\ge1, \text{ there is } A\in\mathrm{SL}_n(\mathbb Z)\\[1mm] \text{such that } \|A\|\le H \text{ and } \langle A\mathbf x-\mathbf y\rangle\le H^{-τ} \end{array} \right\} \] denote the corresponding exponent of uniform approximation. The main result of this paper proves that for almost all $\mathbf{x}, \mathbf{y}\in\mathbb{T}^n$ (with respect to the induced Lebesgue measure), $$\widehatτ_n\!\left(\mathbf{x}, \mathbf{y}\right)\;=\; τ_n\!\left(\mathbf{x}, \mathbf{y}\right)\;=\; n-1.$$ This is the first result to establish the exact generic approximation exponents for an action of a lattice subgroup of $\mathrm{SL}_n(\mathbb R)$ induced by the standard Euclidean action.

fields
Number Theory, Dynamical Systems
pdf ↗ page arXiv:2610.36397 ↗

We represent normalized generalized Lehmer polynomials as moments of a random variable constructed from two independent negative binomial variables and an independent uniform variable. An integral form of their known exponential generating function yields mixtures of uniform distributions and convergent double series. We recover Eulerian polynomial identities and give a probabilistic interpretation of anti-diagonal sums of poly-Bernoulli numbers. We also prove moment determinacy, characterize the Stieltjes parameter range, derive Hankel total non-negativity and log-convexity, and identify the boundary distributions.

fields
Number Theory, Probability
pdf ↗ page arXiv:2610.36409 ↗

We establish a uniform effective Chebotarev-Sato-Tate estimate for abelian varieties, conditional on the Mumford-Tate conjecture and the standard analytic conjectures for L-functions attached to the irreducible representations of the Chebotarev-Sato-Tate group. The proof uses spectral majorants and minorants of indicator functions. For regions whose boundaries have finite upper (d-1)-dimensional Minkowski content, the resulting exponent improves that in Bucur et al. (Theorem 1.1, 2024) whenever the Sato-Tate group has rank greater than one. We establish uniform estimates for regions and finite Galois extensions that vary with the counting parameter, and for prime-dependent regions satisfying an explicit variation condition. As an application, we obtain a conditional upper bound for the number of primes at which a weight-two newform with a quadratic coefficient field has a rational Fourier coefficient. We also formulate a discrepancy conjecture, motivated by a probabilistic model, that implies the asymptotic predicted by Murty (Conjecture 3.4, 1999).

comments
Comments are welcome
fields
Number Theory
pdf ↗ page arXiv:2610.36464 ↗

We evaluate the archimedean Rankin-Selberg zeta integrals associated with spherical principal series representations of ${\rm SO}_{2n}(\mathbf R) \times {\rm GL}_n(\mathbf R) $ and $ {\rm Sp}_{2n}(\mathbf R) \times {\rm GL}_n(\mathbf R) $, where $ {\rm SO}_{2n} $ is split and $n\ge2$. We obtain explicit formulas expressing these integrals as quotients of products of gamma functions. Our approach uses Mellin-Barnes representations of spherical Whittaker functions, together with new integral identities for their Mellin transforms and transformations of Barnes integrals. In the even orthogonal case, we establish a recurrence in the rank and evaluate the integral by induction. The symplectic case then follows from an explicit identity relating the two families of zeta integrals.

fields
Number Theory
pdf ↗ page arXiv:2610.36489 ↗

We study completely normal elements of $\mathbb{F}_{q^n}$ over $\mathbb{F}_q$, that is, elements that are normal over every intermediate field, through partial Lagrange resolvents. In these coordinates, complete normality splits into independent components, in a way closely related to Hachenberger's decomposition theory. Our main tool is a structure theorem for the components; together with a finite computation, it shows that every component has weight (the sum of the probabilities that its conditions fail) at most $3/4$, uniformly in $n$. Combined with the classical character-sum method, this gives an explicit sufficient criterion for the existence of a primitive completely normal element. We verify the criterion for all $n>q$ except $70$ pairs $(q,n)$, which we settle with explicit elements found by computer. Since the case $n\le q$ is due to Garefalakis and Kapetanakis, this confirms the conjecture of Morgan and Mullen (1996) that every $\mathbb{F}_{q^n}$ contains a primitive element that is completely normal over $\mathbb{F}_q$. The same estimates give two further results. First, Hachenberger's conjectured lower bound $(q-1)^{n'}q^{(p^a-1)n'}$ for the number of completely normal elements, where $p$ is the characteristic and $n=p^an'$ with $p\nmid n'$, holds for all pairs $(q,n)$; it was known for regular pairs. Second, primitivity and complete normality are asymptotically independent: as $q^n\to\infty$, the number of primitive completely normal elements is $(1+o(1))\,\varphi(q^n-1)/(q^n-1)$ times the number of completely normal elements (for fixed $n$ this was known). Our arguments for $n>q$ and for both further results, including every computer-assisted step, are formalised in Lean 4 with Mathlib; the formalisation takes the case $n\le q$ and an explicit bound of Rosser and Schoenfeld as axioms and relies on Lean's compiler and interpreter for the large computations.

comments
49 pages. Lean 4 formalisation and verification code: https://github.com/florian-hellwig/morgan-mullen
fields
Number Theory
pdf ↗ page arXiv:2610.36534 ↗

Assume that the Riemann zeta function has no zeros in $\Re(s)>1-δ$, $s\neq1$, for some fixed $0<δ\leq1/2$. We prove that, for every sufficiently large integer $k$, a positive multiple $m_k k!$ has arithmetic complexity $τ(m_k k!)\leq\exp(C\sqrt{\log k\log\log k})$, where $C$ depends only on $δ$. We first establish a smooth-number asymptotic at every sufficiently large center $x$ in intervals of length at least $x^{3/4}$. Combining this estimate with an established genus-two construction yields the arithmetic bound without Cheng's conjecture on smooth numbers in elliptic Hasse intervals. The bound includes the cost of constructing all integer parameters. This work was carried out with assistance from the UniMath model, developed by UniDT (Shanghai) Co., Ltd., and OpenAI's GPT-6.1 Sol and GPT-6 Astra.

fields
Number Theory
pdf ↗ page arXiv:2610.36588 ↗

Generalizing both [LZ14] and [ScV26, SaV24] we prove the interpolation property at Artin characters of Kupferer's regulator map [Ku20] over a fixed local field $L$ over $\mathbb{Q}_p$ for $p$ any prime. Since our methods cover the case $p=2,$ we in particular close a gap in the literature even for the (cyclotomic) 2-variable regulator map from [LZ14]. Moreover, we study big (dual) Bloch-Kato exponential maps for unramified towers and we use them to improve Colmez' reciprocity law in the Lubin-Tate setting from [SaV24]. Finally, we compare Kupferer's regulator map with the 2-variable regulator map of Loeffler-Zerbes when $L=\mathbb{Q}_p$.

comments
27 pages
fields
Number Theory
pdf ↗ page arXiv:2610.36600 ↗

We formulate Poisson summation with periodic weights in forms adapted to arithmetic sums (arithmetic Poisson summation formulas, APSF), in scalar, polynomial-congruence and vector-divisibility versions, extending Berndt's arithmetic Poisson formula. We use APSF as a common framework to rederive Ramanujan's formula for $ζ(2m+1)$ and its character analogues due to Katayama and Berndt-Straub, the functional equations of $ζ$, of Dirichlet's beta function and of Estermann's function, and the exponents of Tolev and Chen-Wang for power-free values of $x^2+y^2+1$ and $x^2+y^2+z^2+k$. It also gives reciprocity laws with Kelvin-function kernels for Lambert series twisted by a Dirichlet character over imaginary quadratic fields of arbitrary class number; asymptotic formulas with error terms for $r$-free values of $x_1^2+\cdots+x_n^2+k$, with Gaussian-weighted versions and the meromorphic continuation of the associated Dirichlet series; an argument of Terras, relating two-dimensional Bessel series for all odd zeta values and even Dirichlet beta values; and a Gaussian representation of a Mersenne-Fermat prime detector.

comments
42 pages, 2 figures
fields
Number Theory, Complex Variables
pdf ↗ page arXiv:2610.36606 ↗

We describe extended interpolation properties of the Katz $p$-adic $L$-function over an imaginary quadratic field. We have three applications. Firstly, we give an elliptic analogue of the Coleman-Ihara formula. Secondly, we extend Kato's explicit reciprocity law for elliptic units and give a proof of the Tamagawa number conjecture for critical Hecke motives. Finally, we prove a large part of the $p$-adic Beilinson conjecture for the Katz $p$-adic $L$-function.

comments
25 pages
fields
Number Theory
pdf ↗ page arXiv:2610.36607 ↗

This monograph develops finite field theory through root sets, Frobenius orbits, trace duality, and cyclic modules, with applications to polynomial constructions and algebraic coding theory. Root-set descent and base extension are expressed through finite reduced algebras, while linearized polynomials connect normal bases, invariant subspaces, and linear recurrences. Coefficient pairings and derivative formulas recover information lost by trace pairings in quotient algebras with repeated factors. Cyclotomic cosets are studied through additive stabilizers, power substitution, and compatible factor lifting, including extensions to Galois rings. A central theme connects low-degree representations of subfields with conjugate root blocks, polynomial composition, and recovery from relative trace data. Return maps on root fibers determine factor degrees and organize constructions of irreducible polynomials, affine symmetries, and permutation polynomials. Applications include cyclic and constacyclic codes, BCH and Goppa distance bounds, rank-metric codes, trace codes, and Fourier support. Character orthogonality and Gauss and Jacobi sums yield counting formulas, a proof of the primitive normal basis theorem, and recurrences for extension-field character sums. Gaussian periods connect these methods with short linear relations in multiplicative subgroups. Proofs of Weil bounds for polynomial character sums combine auxiliary polynomials with generating polynomials and estimates for their reciprocal roots. The emphasis throughout is on explicit arguments connecting field structure, orbit structure, and linear dependence.

fields
Number Theory, Information Theory
pdf ↗ page arXiv:2610.36727 ↗

Schoof asked how many reduced Arakelov divisors cannot arise from the reduction algorithm. Tian and Tran subsequently characterized the image of this algorithm in real quadratic fields as precisely the $1$-reduced ideals. We give a quantitative answer to Schoof's question in the real-quadratic setting by determining the asymptotic proportion of reduced ideals missed by the algorithm. A key arithmetic input is the criterion that a reduced ideal with associated reduced form $(A,b,-r)$ fails to be $1$-reduced if and only if $A-r<b<2(A-r)$. Hence the missed ideals occupy an explicit region of the natural extension of the Gauss map whose relative measure is $λ_{\mathrm{ex}}=\frac{\log(4/3)}{\log4}$. In particular, the missed region occupies exactly half of the digit-one cylinder with respect to the invariant measure. We establish this limiting proportion in two settings. A lattice-point count gives the average asymptotic over fundamental discriminants $Δ\le X$, with error $O(X^{-1/4+\varepsilon})$. For individual real quadratic fields, Duke's equidistribution theorem yields $\operatorname{ECal}(Δ)/κ(Δ) \toλ_{\mathrm{ex}}$ as $Δ\to\infty$. We also determine the limiting distribution of the canonical defect $k/\sqrtΔ$ of the missed ideals. This distribution is supported on $[0,1/\sqrt3-1/2]$, and its lower tail is asymptotic to $\frac{2}{\log(4/3)}t\log(1/t)$ as $t\to0^+$. An exact relation between the integer defect and an endpoint coordinate of the natural extension allows us to transfer equidistribution to the defect law.

fields
Number Theory
pdf ↗ page arXiv:2610.36800 ↗

Let $Z_F(t):=F(1/2+it) H(1/2+it)^{-1/2}$ denote Hardy's function, where the function $F(s)$ belongs to the Selberg class with polynomial Euler product and satisfies the functional equation $F(s)=H(s)\overline{F}(1-\bar{s})$. We prove that for every function of degree $1$ and for certain functions of degree $2$ belonging to the Selberg class, one has \begin{align*} \max_{T\leq t\leq 2T} \pm Z_F(t) &\gg \exp\left(C_F \sqrt{ \frac{\log T}{\log \log T} } \right), \end{align*} where the constant $C_F$ depends on $F(s)$ and are explicitly determined.

comments
36 pages
fields
Number Theory
pdf ↗ page arXiv:2610.36885 ↗

Let $K$ be a fixed number field. We prove a quantitative expansion theorem for mixed sumsets $A+B$, assuming only that $A$ is covered by relatively few cosets of a multiplicative group of small finite valuation rank. The constants are independent of the generators, their heights, and their prime ideal supports. As a consequence, if $A\subset K^\times$, $|A|=N\ge 3$, and each element has nonzero valuation at at most $k$ finite places, then every $B\subset K$ with $|B|=N$ satisfies $$ \max\{|A+B|,|AA|\} \ge N^2\exp\left[-C_K\left((k+1)\log(\log N+2)+\sqrt{(k+1)\log N}\right)\right]. $$ This is $N^{2-o(1)}$ for $k=o(\log N/\log\log N)$. We also obtain a trade-off between the additive and multiplicative deficits, estimates for iterated linear sumsets, and a transfer theorem for product constructions $A=PG$ in terms of the multiplicative deficit of $P$. The proof recovers local scales from an exact sum and bounds the remaining conditional entropy by the overlap of entropy measures on fixed trees.

comments
24 pages
fields
Number Theory
pdf ↗ page arXiv:2610.36930 ↗

In this paper, we give a structural factorization of elliptic divisibility sequences in terms of finitely many modular functions on the modular curve $X_1(N)$. More precisely, we express the functions obtained by pulling back division polynomials along the universal torsion section as products of fixed modular factors. Then we show that this factorization is compatible with the natural action of $(\mathbb{Z}/N\mathbb{Z})^\times$ on the modular curve $X_1(N)$ and so descends to the modular curve $X_0(N)$ via the norm map. As applications, we obtain a perfect power criterion, local and global growth estimates for the associated elliptic divisibility sequences.

comments
18 pages
fields
Number Theory
pdf ↗ page arXiv:2610.37191 ↗

Let $p$ be an odd prime. Let $F$ be a totally imaginary CM field and $F^+$ be the maximal totally real subfield such that $F/F^+$ is unramified at all finite places, and every place $v | p$ of $F^+$ is split in $F$. We allow arbitrary ramification of $p$ in $F^+$. Let $G$ be a rank two definite unitary group over $F^+$ as defined in [2]. Fix a place $\mathfrak{p} | p$ in $F$. In this article, we determine explicit local Serre weights at $\mathfrak{p}$ attached to a modular mod $p$ Galois representation $\barρ_h$ of the absolute Galois group of $F$, associated to an eigenform $h$ of $G$. The eigenform $h$ has arbitrary weights and small slope at $\mathfrak{p}$. We also show analogous results for certain Hilbert modular eigenforms with arbitrary ramification of $p$ in the totally real field. The eigenforms come from totally definite and indefinite quaternion algebras over totally real fields, split at places above $p$. In proving our results, we identify the non-singular submodules in quotients of certain symmetric power $\mathbb{F}_q[\text{M}_2(\mathbb{F}_q)]$-modules by Dickson polynomials and their twisted versions.

comments
32 pages
fields
Number Theory
pdf ↗ page arXiv:2610.37261 ↗

Let $k$ be a field of characteristic $\neq 2$. Under suitable assumptions on the dimensions of anisotropic quadratic forms over $k$ and $k(x)$, we show that a definite integral quadratic form over the polynomial ring $k[x]$ is universal if the form represents all scalars from $k$ and all scalar multiples of $x$. The coefficient field $k$ in our main result can be a finite field, a non-dyadic higher local field, a global function field, or a one-variable function field over $\mathbb{C}$ or $\mathbb{C}(t)$. We have a finite universality criterion set over $k[x]$ for some coefficient fields $k$, including some cases where $k$ has infinitely many square classes (e.g., $k=\mathbb{C}(t)$).

comments
12 pages
fields
Number Theory
pdf ↗ page arXiv:2610.37339 ↗

We study the $p$-adic valuation trees associated with the sequence \[ e_n=\int_0^\infty e^{-t}F_n(t)\,dt, \] obtained by evaluating the Laplace transforms of the Fibonacci polynomials at $1$. We answer an open question of Givens and Moll in the negative by constructing an explicit counterexample showing that even internal branches need not be simple. We then reformulate the Givens–Moll lifting criteria for the even and odd subsequences in terms of the roots and derivatives of two associated polynomials. In particular, simple roots modulo $p$ give rise to simple internal branches. Finally, we develop a recursive lifting procedure that determines the unique surviving residue class at each level of a branch associated with a simple root and computes the $p$-adic valuations of the corresponding terms.

fields
Number Theory
pdf ↗ page arXiv:2610.37356 ↗

In this paper, we prove a zero-density estimate for the symmetric square $L$-functions associated with Hecke eigencuspforms in the weight aspect. The proofs are based on the mollified second moment of these $L$-functions.

comments
20 pages
fields
Number Theory
pdf ↗ page arXiv:2610.37371 ↗

Let $F$ be an algebraic number field and $\mathcal{O}_{F}$ the ring of integers of $F$. Let $L$ be an integral $\mathcal{O}_{F}$-lattice of rank $n\ge1$ on an anisotropic quadratic space $V$. We prove that if $L$ is norm Euclidean, then $L$ is $\mathcal{O}_{F}$-maximal of class number one, and the class number of $F$ equals the class number of $L\perp\mathbf{H}$, where $\mathbf{H}$ is the hyperbolic lattice with quadratic form $xy$. Under the same assumption, $F$ has class number one if $n\ge3$ and at most two if $n=1,2$. We also prove finiteness up to isometry for totally definite norm Euclidean lattices over totally real number fields in ranks at least three as the field varies, in rank two for fixed field degree, and in rank one over a fixed field. This confirms Clark's finiteness conjecture in ranks at least three.

comments
19 pages
fields
Number Theory
pdf ↗ page arXiv:2610.37377 ↗

We give $L^2$-bounds with respect to the height of the spectral parameter for Epstein zeta functions $\mathrm{Z}_Q(s)$ and non-holomorphic Eisenstein series $E_{\mathfrak{c}}(s,z)$ attached to the Hecke congruence subgroup $Γ_0 (p)$ of prime level $p$. Our bounds are completely explicit in the level and, additionally, constitute an almost sure logarithmic improvement over previously established bounds for the level-one Eisenstein series. Following the approach of Kelmer–Kontorovich–Lutsko, we combine the approximate functional equation for $\mathrm{Z}_Q(s)$ with estimates for the number of pairs of lattice vectors at which $Q$ takes nearly equal values. Under a Diophantine hypothesis on $Q$, we sharpen these counting estimates using results of Eskin–Margulis–Mozes on the distribution of values of quadratic forms of signature $(2,2)$.

fields
Number Theory
pdf ↗ page arXiv:2610.37420 ↗

If $F$ is a number field and $F = F_0 \subset F_1 \subset F_2\subset \cdots$ is a tower of Galois extensions with $\mathrm{Gal}(F_n/F)$ cyclic of order $p^n$, K. Iwasawa gave a closed formula for the growth of the $p$-part of the class number of $F_n$ whenever $n$ is large enough. We present a new proof of this formula, obtained by reinterpreting its summands in terms of Sylvester matrix rank functions on the Iwasawa algebra $Λ$ without resorting to the classification of finitely-generated $Λ$-modules up to pseudo-isomorphism.

comments
21 pages, comments are welcome!
fields
Number Theory, Group Theory, Rings and Algebras
pdf ↗ page arXiv:2610.37490 ↗

For a positive integer $n$, let $D(n)$ be its set of positive divisors and put $N:=τ(n)=|D(n)|$. We characterize the arithmetic progressions $\mathcal P$ of length $N$ that admit a bijection with $D(n)$ pairing relatively prime integers. Such a coprime matching exists if and only if, for every prime $p\mid n$, at most $N/(ν_p(n)+1)$ terms of $\mathcal P$ are divisible by $p$, where $ν_p(n)$ is the exponent of $p$ in $n$. Call $n$ an $M$-number if $p^p\nmid n$ for every prime $p$. We prove that $D(n)$ can be matched with every primitive arithmetic progression of length $N$, meaning one whose first term and common difference are coprime, if and only if $n$ is an $M$-number. This proves a conjecture of McNew and Pomerance. We call $n$ \emph{matchable} if $D(n)$ admits a coprime matching with the primitive arithmetic progression $\{1,2,\ldots,N\}$. The general criterion implies that $n$ is matchable if and only if $\lfloor N/p\rfloor\le N/(ν_p(n)+1)$ for all $p\mid n$. We explicitly classify the matchable non-$M$-numbers, settle a conjectured divisor criterion, and prove that every divisor of a matchable integer is matchable. A Vandermonde determinant gives the required matchings under bounds on the prime exponents, while a partition into shorter progressions handles the remaining cases.

fields
Number Theory, Combinatorics
pdf ↗ page arXiv:2610.37512 ↗

Information Theory24

We study deterministic causal quantization with pathwise error at most $\varepsilon$, using $ω$ shared representatives updated from transmitted addresses alone. For a midpoint realization with first-in, first-out replacement, we compare address entropy rates with process $\varepsilon$-entropy, the least entropy rate of any block quantizer meeting the same constraint. From $\varepsilon$-separated initialization, the gap is at most two bits per sample for every stationary source with finite scalar $\varepsilon$-entropy, without density, independence or mixing assumptions. A causal address-tree argument extends this comparison to any synchronized quantizer with cell overlap $m$, giving $\log_2 m$ bits. Lattice initialization with consistent boundary ownership realizes uniform grids with sharp one-bit scalar and process comparisons. Under an entropy condition and finite Fisher information rate $J$, the gap is at most $2J\varepsilon^2/(π^2\ln 2)$; general initializations add a term of order $ω\varepsilon\sqrt{J}$. The scalar grid estimate gives a nonasymptotic form of Rényi's theorem. Since a grid is causal, the price of causality is at most $\min\{1,2J\varepsilon^2/(π^2\ln 2)\}$ bits per sample, versus the high-resolution squared-error loss $\tfrac12\log_2(πe/6)$. These bounds certify $\varepsilon$-entropy brackets for autoregressive Laplace and Gaussian sources. For memoryless sources, the optimal causal rate equals the scalar $\varepsilon$-entropy. For a stationary process on finitely many $2\varepsilon$-separated centers, adding independent uniform noise of width $2\varepsilon$ gives exact Shannon lower-bound equality at every block length, attained by storing those centers. This includes histograms with finitely many bins. Storing two non-lattice centers beats every uniform grid of the same cell width by half a bit per sample.

fields
Information Theory, Probability
pdf ↗ page arXiv:2610.35933 ↗

Accurate CSI forecasting is essential for mitigating channel aging and supporting reliable downlink transmission in high-mobility wireless systems. In FDD systems, forecasting future DL CSI from historical UL observations requires joint temporal extrapolation and cross-band inference. However, defining historical UL as the input and future DL as the target does not specify how elapsed time and UL-to-DL band separation should be represented in a pretrained predictor's positional space. FAR-CSI establishes an FDD-aware representation of channel forecasting within a joint temporal and operating-band positional field. Within this field, historical UL observations and future DL targets occupy segregated regions according to their temporal and operating-band roles, preserving the distinction between observed UL context and unobserved DL prediction targets. This representation is instantiated through a unified structured routing rule that governs encoder positional assignment, decoder-field reconstruction, and future-DL target selection, ensuring the same FDD geometry is preserved from contextual encoding to prediction readout. Under matched backbone, initialization, and training conditions, FDD-aware routing reduces NMSE by 21.44\% relative to generic routing. Across the evaluated FDD baselines, FAR-CSI achieves higher CSI prediction accuracy and MRT-based spectral efficiency, while retaining its predictive advantage across mobility conditions, prediction horizons, reduced training-data regimes, and noisy historical observations.

fields
Information Theory
pdf ↗ page arXiv:2610.35956 ↗

Distributed speculative decoding has emerged as a promising approach for device–server collaborative large language model (LLM) inference, but its decoding speedup is often limited by the vocabulary-scale communication overhead. In existing distributed speculative decoding methods, exact rejection sampling requires transmitting full token distributions over the LLM vocabulary between the device and the server, which becomes costly in bandwidth-constrained environments. To address this issue, we propose DISCO, a communication-efficient framework for distributed speculative decoding, that eliminates vocabulary-scale distribution transfer while preserving the exact target distribution. DISCO achieves this by introducing a server-side replica of the draft small language model (SLM), which enables the server to locally obtain the draft-side distribution required for rejection sampling. As a result, DISCO exchanges only drafted token indices and scalar probability values between the device and the server. Experimental results on multiple datasets show that DISCO consistently achieves higher decoding speedup than existing distributed speculative decoding methods, particularly under stringent bandwidth constraints.

fields
Information Theory
pdf ↗ page arXiv:2610.35957 ↗

For any desired rate $R \in (0,1)$, and gap to list-decoding capacity $\varepsilon > 0$, we give an explicit $n^{O_{\varepsilon}(1)}$ time construction of Reed-Solomon codes over extension fields of size polynomial in the block length $n$ that can be list-decoded from a fraction $1-R-\varepsilon$ of errors with the optimal list-size of $O(\frac{1-R}{\varepsilon})$. We also give a simpler, strongly explicit construction of evaluation points for which the associated Reed-Solomon code can be list-decoded from $1-R-\varepsilon$ errors with list size $(1/\varepsilon)^{O(1/\varepsilon)}$. We provide a deterministic list-decoding algorithm for both constructions running in time $O_ε(1)\mathsf{poly}(n)$. In fact, our Reed-Solomon codes are near-optimal subspace-design codes over a subfield. Via known results, this directly implies the combinatorial list-decoding guarantee, as well as near-optimal list-recovery bounds. We also present a new reduction showing that the subspace-design guarantee also implies line decoding over the extension field itself, which then yields proximity gaps and mutual correlated agreement up to proximity parameter approaching the optimal $1-R$ radius. Compared to the recent algorithm of Brakensiek, Chen, Putterman, Zhang, and Zheng (2026), which list decodes every Reed-Solomon code over prime fields up to capacity, our approach achieves constant (in fact, near-optimal) list size rather than a polynomially growing one, at the cost of carefully choosing the code's evaluation points. In a companion paper, we show that these respective restrictions are unavoidable.

fields
Information Theory
pdf ↗ page arXiv:2610.36215 ↗

The information bottleneck (IB) trades compression of an input $X$ against prediction of a target $Y$ through a multiplier $β=1/λ$, yet each operating point is usually served by training, or fully conditioning, a new encoder. For jointly Gaussian data, we show that the IB solution factorizes as $A(β)=D(β)W$: the information-bearing directions $W$ do not depend on $β$, and $β$ acts only through scalar gains. We ask whether this separation of \emph{representation} from \emph{rate control} survives beyond Gaussian models, treating the budget as a given operating parameter rather than a quantity to be optimized. For a budget-free representation \(F(X)\), the \emph{separation gap} is controlled by the \emph{sufficiency deficit} $ε_F=I(X;Y\mid F(X))$: We show there is at least $ε_F$ in the worst case for finite inputs, and it is bounded uniformly in $λ$ by a function of $ε_F$ for finite label alphabets. The same certificate extends to a new target distribution and to pooled multi-domain data, where a conditional-alignment term appears. On MNIST, SVHN, CIFAR-10 and STL-10 (10 seeds, eight prespecified budgets), a frozen CNN representation with one compressor–decoder pair per budget reduces estimated training multiply–accumulate operations (MACs) by $7.7\times$ and wall-clock time by about $2\times$ relative to end-to-end VIBs, and attains higher budget-averaged accuracy ($65.6\%$ versus $57.3\%$ per domain; $67.1\%$ versus $57.1\%$ pooled), with the advantage concentrated at strong compression, where the VIBs collapse. At weak compression ($λ\le0.036$), VIB has higher cross-entropy and ECE; its surrogate Lagrangian is higher per domain, indistinguishable from the pooled VIB in the macro average, and lower on the pooled test set.

fields
Information Theory
pdf ↗ page arXiv:2610.36255 ↗

We study the thermodynamic efficiency of message-passing decoders using the mismatch cost framework. This framework quantifies the excess entropy generated when a system operates away from the thermodynamically optimal initial distribution. It provides a computable lower bound on irrecoverable energy loss. The decoder is modeled as a main decoding program with local message updates implemented as subroutines. Memory is treated as an external copying subsystem whose thermodynamic cost is excluded. Studying the $[7,4]$ Hamming code transmitted over a binary erasure channel, we show how the channel-induced initialization statistics and the step-wise evolution of the algorithmic state distribution during decoding determine the total mismatch cost. We further show that different thermodynamically optimal state distributions, representing different underlying physical implementations of the decoder, can produce different thermodynamic costs for identical decoding dynamics. Conditioning the input distribution on correctable and non-correctable erasure patterns further reveals distinct thermodynamic signatures. These results demonstrate that thermodynamic cost is jointly governed by decoding dynamics, input statistics, and hardware characteristics. This points to opportunities for thermodynamically-informed decoder design to realize energy-efficient physical implementations.

comments
Published in IEEE Journal on Selected Areas in Information Theory (JSAIT), August 2026. 36 pages, 10 figures
journal
IEEE J Sel. Areas in Inf Theory, vol. 7, pp. 396-410, 2026
fields
Information Theory
pdf ↗ page arXiv:2610.36348 ↗

This paper studies the fundamental limits of bistatic integrated sensing and communications (ISAC) systems over vector channels. We first consider a class of parallel channels in which, on each component channel, the channel from the transmitter (Tx) to the communication receiver (Rx) is degraded with respect to the channel from the Tx to the sensing Rx, or vice versa, where the direction of degradation may vary across component channels. For this class of channels, we establish matching achievability and converse results, which yield a single-letter characterization of the capacity-distortion function $C(D)$. For parallel Gaussian channels with additive Gaussian states, the evaluation of the capacity-distortion-power function $C(D,P)$ under the squared error distortion measure reduces to a joint power and distortion allocation problem across the component channels. We then extend the analysis to general vector channels and derive single-letter lower and upper bounds on $C(D)$. In particular, for vector Gaussian channels with additive Gaussian states, we prove the optimality of Gaussian superposition coding without requiring a global degradation order between the communication and sensing channels, by combining the channel enhancement technique with a matrix extremal inequality based on Fisher information. This leads to an exact characterization of $C(D,P)$.

fields
Information Theory
pdf ↗ page arXiv:2610.36351 ↗

Temporal Predictive Coding (tPC) networks offer a layer-local alternative to backpropagation. For linear tPC, we show, by applying tools from cellular sheaf theory, that temporal credit assignment is a tug-of-war between spatial displacement and temporal errors, whose ratios are modulated by persistence and exposure operators $P$ and $Q$. Spatial inference therefore gives us the effective recurrent transition $F=PA$, whilst future credit is filtered by $P^\ast$. This creates a persistence–error tradeoff in spatial modes: under non-amplifying invariant recurrence, modes that preserve influence over long delays simultaneously fail to expose that influence strongly as temporal error. Recurrence can circumvent this via routing between persistent and exposed modes.

comments
Accepted to the NeurIPS 2026 Workshop on Symmetry and Geometry in Neural Representations (NeurReps)
fields
Information Theory
pdf ↗ page arXiv:2610.36412 ↗

In a membership inference attack, an attacker is given the data of a particular individual, and tries to determine if that individual was included in a particular dataset, given some kind of statistics derived from the dataset or a model trained on the dataset. Roughly, a membership inference attack is robust if it succeeds even when the statistics are highly noisy, regardless of how that noise is introduced. We give a simple presentation of robust membership inference attacks for mean estimation, covering both multivariate Gaussian distributions and binary product distributions. We also give a summary of the literature on robust membership inference, covering history, extensions, and connections to other research areas.

comments
34 pages, 2 figures
fields
Information Theory, Statistics Theory
pdf ↗ page arXiv:2610.36458 ↗

When one time series informs another, the delay at which that information arrives determines if models can utilize this relationship effectively. A source may be strongly predictive at one delay but uninformative at neighboring lags. Therefore, fixed input windows may obscure this structure by bundling informative and irrelevant history together. Transfer entropy (TE) can recover these delays by measuring how much a source's past tells us about a target beyond the target's own history. However, existing TE estimators require separate fits across source and conditioning depths, with further estimation on local windows to track delays over time. We introduce ATEMPO, a diffusion-based transfer entropy estimator amortized over lags and time. ATEMPO samples source and conditioning depths during training and conditions explicitly on temporal position, allowing a single score network to estimate the full TE surface without refitting. Differencing this surface recovers when information arrives. Across synthetic datasets with ground-truth TE, ATEMPO achieves the lowest mean relative error among baselines, and on datasets with known lags, ATEMPO achieves $0.98\pm0.04$ lag-detection accuracy. ATEMPO requires a single fit across source depths $K=6$ and target depths $L=3$, whereas existing estimators require 18 separate fits to evaluate the same $K \times L$ grid. On three systems with time-varying lag, ATEMPO is the only estimator to maintain perfect lag-detection accuracy across all evaluated window lengths. These recovered delays improve downstream forecasting: ATEMPO-selected inputs reduce MSE on WEATHER-5K by 7.25%, while 120 selected inputs outperform the benchmark's 240-feature default by 4.6%. Across three brain-computer interface tasks, using ATEMPO-selected lags achieves the highest or equal-best $R^2$, improving mean $R^2$ by 4.1% relative to the strongest competing baseline.

fields
Information Theory
pdf ↗ page arXiv:2610.36536 ↗

We present general upper and lower bounds on the optimal error exponent for the Active Simple Hypothesis Testing (ASHT) problem. ASHT has been extensively studied across diverse fields including Information Theory, Bandits, and Operations Research, a precise characterization of this optimal exponent has remained elusive. This work combines a recent Partial Differential Equation (PDE) characterization of the optimal exponent by Vijayan (2025) with the comparison principle of viscosity solutions of Hamilton-Jacobi PDEs. This leads to bounds that are parametrized by convex functions, sets and lower Hamiltonian relaxations, a flexibility that generalizes many known results in the literature.

fields
Information Theory
pdf ↗ page arXiv:2610.36657 ↗

Spatial degrees of freedom (DoF) are fundamental to the performance of multiple-input multiple-output (MIMO) systems. However, the use of conventional fixed-location antennas introduces disparities in the users' distances to a common base station, which makes the multi-user channel matrix more susceptible to rank deficiency and, consequently, a loss of spatial DoF. This paper shows that generalized pinching antennas can be used to design a new form of MIMO robust to the loss of spatial DoF. In particular, by deploying pinching antennas in close proximity to the users, the resulting multi-user channel Gram matrix can have a well-conditioned, almost diagonal structure, which reduces its sensitivity to the users' spatial deployment as well as the risk of rank deficiency. Two performance metrics, namely the rank-deficiency probability and the outage probability, are used to evaluate the performance of pinching-antenna-assisted MIMO transmission. The presented analytical and simulation results demonstrate the capability of pinching antennas to preserve spatial DoF and provide ubiquitous connectivity by effectively supporting users at spatially separated locations.

fields
Information Theory
pdf ↗ page arXiv:2610.36670 ↗

Boolean functions are widely used in cryptography and coding theory. A key challenge lies in finding the weight spectrum for all $n$-variable Boolean functions of degree at most $r$, also known as the Reed–Muller code $\RM(r,n)$. In 2023, Carlet and Solé determined the weight spectra of $\RM(n-c,n)$ for $c=3,4$, and conjectured the weight spectrum of $\RM(n-c, n$) for $n\ge2c$. The weight spectra of $\RM(n-5, n)$ and $\RM(n-6,n)$ were determined by Carlet in 2024 and by Lou and Wang in 2025, respectively, confirming the Carlet-Solé conjecture for the cases $c=5,6$. Very recently, Leuenberger et al. and Wei et al. investigated the weight spectrum of $\RM(7,14)$ independently, and there are eight potential weights which have not been determined. Up until now, the weight spectra of the Reed–Muller codes $RM(r,n)$ are still unknown for $r=3,...,n-7$. In this paper, we determine the exact weight spectrum of $\RM(m-7, m)$ for every $m\ge 14$. By combining affine geometry and Fourier analysis, we exclude the conjectured eight weights for $\RM(7,14)$. We then extend the exclusions to all higher dimensions using a geometric transfer. Consequently, we provide a counterexample to the Carlet-Solé conjecture.

fields
Information Theory
pdf ↗ page arXiv:2610.36687 ↗

Additive codes over $\Ffour$ that are self-orthogonal under the trace inner product correspond to binary quantum stabilizer codes. We study 1-generator $(1,2)$-quasi-cyclic (QC) additive codes and their exact quantum distances. For composite lengths $n=\ell m$, an ancestor-lifting theorem transports dual words of the folded ancestor to the length-$n$ code. A two-layer decomposition for $\ell=2$ and a CRT splitting for odd $\ell$ yield dimension laws and explain equality, escape, and mixed-branch behavior in the certified examples. A palindromic criterion ensures symplectic self-orthogonality, while reversal pairing and a symplectic Gram rank formula describe broader construction families. Exact distances are established by information-set decoding screening, Brouwer–Zimmermann computation on the additive dual, and independent witness verification. We certify twenty-one record-breaking binary quantum codes across twenty entries of Grassl's code table, with twenty-eight certified members in total. Propagation improves thirty-four further entries, and explicit constructions are supplied for three table-tying GV entries marked Code Missing Construction. Six additional candidates are reported only with rigorous upper bounds and heuristic screening evidence. The results connect divisor-lattice structure in additive quasi-cyclic codes with reproducible certification of exact quantum distances.

fields
Information Theory
pdf ↗ page arXiv:2610.36700 ↗

Root mean square (RMS) delay spread is a key parameter characterizing temporal dispersion in radio propagation channels. Its estimation from measured power delay profiles (PDPs), however, depends strongly on the adopted power threshold. This paper investigates the sensitivity of RMS delay spread estimates to relative PDP thresholding in a time-varying millimeter-wave channel at 33 GHz. The analysis uses wideband channel sounding measurements conducted in a dynamic road environment with moving vehicles. The temporal evolution and empirical cumulative distributions of RMS delay spread are evaluated for thresholds ranging from -20 to -50 dB. Reducing the threshold increases the median RMS delay spread from approximately 140 to 168 ns and the 90th percentile from 200 to 210 ns. Both statistics converge at lower thresholds, with -40 dB providing a practical convergence level. The results emphasize the importance of threshold selection for reliable channel characterization and consistent comparison of measurements.

comments
5 pages, 7 figures
fields
Information Theory
pdf ↗ page arXiv:2610.36783 ↗

Fluid antenna systems (FAS) leverage continuous reconfigurable apertures to transcend the spatial limits of fixed-position arrays, yet acquiring full-aperture channel state information (CSI) from sparse observations remains a formidable bottleneck. In this paper, we establish a continuous-field formulation for FAS by recasting channel estimation as a stochastic inverse field-reconstruction problem constrained by the projected Helmholtz model. Through electromagnetic information theory (EIT)-based spectral leakage analysis, we demonstrate that finite-aperture reconstruction induces an intrinsic convolutional relationship in the spectral domain, rendering the estimation task a fundamental frequency-domain inverse deconvolution problem. To solve this, we propose a U-shaped Fourier neural operator (UFNO) specifically tailored for noisy, underdetermined wireless environments. Furthermore, we leverage the Landau–Pollak theorem to derive an aperture-dependent EDoF bound that guides the spatial mode budget, and employ a physics-aware Sobolev loss to promote consistency in both channel values and spatial variations. Unlike conventional learning-based approaches, a single frozen UFNO can be directly evaluated at unseen spatial–frequency resolutions without resolution-specific retraining or explicit output interpolation. Extensive simulations show that our framework significantly outperforms state-of-the-art benchmarks in accuracy while achieving orders-of-magnitude lower inference latency than iterative solvers at high spatial–frequency resolutions.

fields
Information Theory
pdf ↗ page arXiv:2610.36787 ↗

We study variable-length feedback (VLF) coding over $K$ additive white Gaussian noise (AWGN) channels whose noise variances are unknown to both terminals but lie in a known interval. The receiver selects one channel at each time, conveys the selection over noiseless feedback, and must learn the channel qualities while decoding, to avoid overusing suboptimal channels. This setting, which we call the VLF-Bandit problem, combines VLF coding, universal decoding, and bandit channel selection. We develop two joint coding and channel-selection schemes. The VLSF-Bandit scheme is a single-phase variable-length stop-feedback (VLSF) scheme based on a Gaussian random codebook, a score formed from a ratio of likelihood mixtures, and the asymptotically optimal $\text{kl-UCB}^{++}$ bandit algorithm applied to the received power. The VLF-Bandit scheme sends the message in packets by Schalkwijk–Kailath transmission, uses its normalized outputs as noise samples to learn the variances, and lets its exploration packets carry message bits. We prove second-order achievability bounds for both schemes, which are first-order optimal. In the logarithm of the achievable codebook size, the magnitude of the $\log N$ coefficient, with $N$ the average decoding time, is a baseline of one plus a channel-exploration term for VLSF-Bandit. For VLF-Bandit, this magnitude is smaller, with a universal-decoding term replacing the baseline and a smaller channel-exploration term. The VLF-Bandit bound improves on the universal VLF bound of Yavas and Tan (2025) for $K=1$ and on the bound of an explore-then-commit strategy whenever $K\geq2$.

comments
34 pages, submitted to Transactions on Information Theory
fields
Information Theory
pdf ↗ page arXiv:2610.36818 ↗

This work investigates how site-specific digital twins (DTs) can support wireless localization using simplified distributed infrastructure. We propose PRISM (Probabilistic Refinement using Inverse learning and a Site-specific Model), a DT-guided probabilistic localization scheme with single-antenna access points (APs). Localization is formulated as a probabilistic inverse problem in which a conditional proposal distribution over UE locations is learned from multi-node channel impulse responses (CIRs). The proposal generates candidate locations, with its spread regularized during training and DT consistency used to reweight candidates and refine the estimate at inference. The DT serves as an approximate forward-consistency model, avoiding assumptions of exact channel fidelity or differentiability with respect to user position. Numerical experiments demonstrate gains over data-driven and classical fingerprinting benchmarks, while achieving a median error of approximately $0.7$ m using 10 single-antenna APs. Without retraining, PRISM remains effective under degraded inference-time DT features. PRISM also yields uncertainty estimates from its proposal covariance, with empirical $χ^2$ calibration approaching ideal behavior as APs increase.

fields
Information Theory
pdf ↗ page arXiv:2610.36905 ↗

A Gaussian random wave field is sampled on a lattice within a bounded aperture. After displacement, re-indexing leaves a residual in the Voronoi cell. We study whether the empirical lag correlation has positive real part throughout the cell. For each realization in a single-output reciprocal link with noiseless over-the-air return, this is equivalent to the negative returned gradient being a descent direction for all residuals and differentiable losses with nonzero error signals. Return noise is treated exactly. For isotropic fields, ball-shaped apertures, unshifted lattice points, and dilations approaching the critical covering radius, every minimizer is a deep hole except with exponentially small probability. A Gaussian vector on the deep holes determines the limiting failure probability, with explicit covariance and lattice-independent limiting normalization. Square, hexagonal, simple cubic, and body-centered cubic lattices have two, three, four, and twelve sign classes. At fixed target failure probabilities below the critical limiting value, this covariance determines the exact leading aperture constant. When the margin and target failure probability vanish jointly, that constant is common to lattice shapes within each dimension. A cubic correction yields vanishing relative tail error in a moderate-deviation regime. A non-asymptotic two-sided aperture law has explicit constants. With a Rayleigh-faded specular path and positive diffuse margin, a failure floor occurs at every positive path power exactly when the cosine of its phase change is negative somewhere in the cell. The floor vanishes for all directions exactly when the covering radius is at most a quarter wavelength.

comments
22 pages, 1 figure. Submitted to the IEEE Transactions on Information Theory
fields
Information Theory
pdf ↗ page arXiv:2610.37174 ↗

In reciprocity-based over-the-air split learning, channel aging rotates and attenuates the derivative signal returned to the transmitter. Seven case studies combine numerical forecasts, mechanism checks, and competing predictions. Near the correlation zero, an archived SGD comparison finds truncated inversion 0.000813 lower in cross-entropy than polar correction, with a paired 95% interval for the difference from -0.001853 to 0.000060, and a retained-mode quadratic proxy slightly favors truncation. Selecting the return port with the largest received power falls short of the predicted five-point alignment advantage over random selection at both tested spacings, gaining at most 0.97 points, while the local inverse relation between amplitude and phase gradient is supported. Seven specified interference-induced accuracy-collapse predictions are rejected, and independent zero-mean interference preserves the conditional mean under a fixed linear chain. Per-subcarrier correction and spherical-wavefront compensation do not deliver their predicted accuracy gains. Focusing leaves both methods near chance in a single-mode model and adds no observed gain in a model with a line-of-sight component plus scattering. A successful port assignment under nonsymmetric motion refutes an exact-symmetry requirement. The competing positioning-error prediction is supported. Gaussian rail jitter with a standard deviation of 0.383 wavelengths yields high terminal test accuracy, whereas a fixed bias of the same magnitude yields near-chance accuracy. These cases distinguish valid local physical relations from unsupported training forecasts through task support, noise injection, and update moments.

comments
10 pages, 1 figure, 5 tables. Companion to arXiv:2610.23118, arXiv:2610.23135 and arXiv:2610.23137
fields
Information Theory
pdf ↗ page arXiv:2610.37188 ↗

The spatial chromatic model (SCM) is a novel entity-oriented spatial data model proposed in recent years. Spatial tessellations formed by SCM have been applied in preliminary analyses of spatial point pattern recognition,spatial topology, and Voronoi diagram generation. Unlike previous spatial tessellation models, a key feature of SCM is that each subspace (cell) possesses a chromatic code, which serves as the foundation for various theoretical analyses and applications of SCM. The uniqueness of cell cluster codes in SCM constitutes an important property of the model. Previous studies have demonstrated the uniqueness of the chromatic codes of 2-cell and 3-cell connected disjoint clusters. Building on this analysis of connected cluster code characteristics, this paper proves the uniqueness of the chromatic codes of 4-cell connected disjoint clusters. The proof results and the new concepts introduced in this study help further refine the spatial chromatic model and provide a theoretical foundation for its subsequent applications.

fields
Information Theory, Combinatorics
pdf ↗ page arXiv:2610.37395 ↗

We consider the problem of private information retrieval from MDS-coded databases with $T$ colluding servers, $U$ robust servers, and $B$ Byzantine servers, referred to as U-B-MDS-TPIR. In the U-B-MDS-TPIR setting, $M$ files are stored across $N$ servers, and in this paper each file is independently encoded using an $(N,K)$ GRS code. We characterize the capacity among linear, full support-rank U-B-MDS-TPIR schemes as $C=\frac{N-U-2B}{N-U}(1+\frac{K+T-1}{N-U-2B}+\cdots+(\frac{K+T-1}{N-U-2B})^{M-1})^{-1}$ when $N-U-2B\ge KT$. When $K=1$ or $T=1$, the same expression gives the capacity without the linearity and full support-rank restrictions. We further extend the construction to the retrieval of $P$ out of $M$ files. For $P \ge \frac{M}{2}$, the resulting multi-file scheme achieves capacity when $K=1$ or $T=1$.

fields
Information Theory
pdf ↗ page arXiv:2610.37412 ↗

This paper investigates cross-layer design for differentially private federated learning over cell-free massive MIMO networks with over-the-air (OTA) aggregation. Distributed access points (APs) collaboratively serve user equipments (UEs) through user-centric clustering, where clipped local model updates are perturbed with Gaussian noise and simultaneously transmitted over a shared uplink for OTA averaging at a central processing unit. Exploiting channel hardening, we show that thermal noise inherent in wireless reception can act as a privacy-masking resource, reducing or even eliminating the need for artificial noise. Based on this observation, we formulate a joint optimization problem over AP-UE associations, UE transmit powers, and artificial noise variances to minimize aggregation variance subject to per-UE zero-concentrated differential privacy constraints derived from an ergodic privacy accounting framework. Analysis reveals a regime-dependent structure: stringent privacy requirements favor low transmit powers and sparse AP associations to exploit thermal noise for privacy, whereas relaxed requirements favor larger clusters to improve aggregation quality. The resulting mixed-integer nonconvex problem is solved via alternating optimization combining difference-of-convex power allocation and greedy coordinate-descent clustering. Numerical results show up to 89% lower aggregation variance than conventional baselines and enable 97% test accuracy on MNIST with consistent gains on CIFAR-10. Under strict privacy budgets, competing schemes remain near random-chance performance, while under relaxed budgets they continue to lag significantly behind the proposed design.

fields
Information Theory
pdf ↗ page arXiv:2610.37424 ↗

In this paper, we study the minimum subpacketization of $(n,k)$ Reed-Solomon (RS) codes over $\mathbb{F}_{q^\ell}$ admitting cut-set-optimal exact $\mathbb{F}_q$-linear repair of every coordinate from any $d$ helpers. For $3\le k<d\le n-1$, write $s=d+1-k$ and denote this minimum by $\ell_{min}(n,k,d;q)$. Using divisibility and relative-degree constraints on cross-ratio fields of evaluation sets, we establish $\ell_{min}(n,k,d;q)\ge sG_s(d+1)$, where $G_s(m)$ is the minimum product of $m$ pairwise coprime integers, each at least $s$. Let $Δ_s(n,d)$ be the minimum of $lcm(a_1,\ldots,a_n)$ over positive integers $a_1,\ldots,a_n$ satisfying $lcm(a_j:j\in J\cup\{i\})\ge s~lcm(a_j:j\in J)$ for every $d$-subset $J\subseteq[n]$ and every $i\in[n]\setminus J$. We construct RS codes with subpacketization $sΔ_s(n,d)$ and further prove that $\ell_{min}(n,k,d;q)=sΔ_s(n,d)=sG_s(n)$ whenever $d(d+2)\geq 4(n+1)$, for every prime power $q$. In particular, this determines the minimum subpacketization in the full-helper case $d=n-1$, resolving the corresponding exact linear-repair case of Open Problem 20 posed by Ramkumar et al. in their survey on codes for distributed storage. We further give a finite formula for $G_s(n)$ and determine the exact threshold for its closed-form expression in terms of the first $n$ primes. For fixed $k$ and $d$, we prove $\log\ell_{min}(n,k,d;q)= Θ_{k,d}(\log n\log\log n)$ as $n\to\infty$, using divisor-function estimates and cover-free families. An affine-plane construction shows that $Δ_s(n,d)<G_s(n)$ for some parameters, so $sG_s(n)$ is not a universal formula. We conjecture that $\ell_{min}(n,k,d;q)=sΔ_s(n,d)$ for every prime power $q$ and all $3\le k<d\le n-1$.

comments
33 pages
fields
Information Theory
pdf ↗ page arXiv:2610.37522 ↗

Functional Analysis21

The symplectic interlacing theorem states that the symplectic spectrum of a $2n \times 2n$ real symmetric positive definite matrix satisfies a $2k$-step interlacing relation with the symplectic spectrum of every $2(n-k) \times 2(n-k)$ $s$-principal submatrix. We prove the converse in this paper. More precisely, given a positive global symplectic spectrum of size $n$ and a positive local symplectic spectrum of size $n-k$ satisfying the $2k$-step interlacing inequalities, we construct a $2n \times 2n$ real symmetric positive definite matrix having the prescribed global symplectic spectrum and a $2(n-k) \times 2(n-k)$ $s$-principal submatrix of it having the prescribed local symplectic spectrum. Our main tool is a reduction of $s$-principal compression to a rank-$k$ singular value perturbation. In the context of the Gaussian quantum marginal problem, our result implies that the symplectic interlacing inequalities are not merely necessary conditions on Gaussian marginals, but are also the complete spectral compatibility conditions for the global-versus-local-mode marginal problem.

fields
Functional Analysis, Information Theory, Mathematical Physics, Symplectic Geometry, Spectral Theory
pdf ↗ page arXiv:2610.35376 ↗

In this paper, we establish an analytical model for a class of hyponormal contractions $T$ satisfying $Ran[T^*,T] \subseteq \mathcal D_{T^*}\ominus \mathcal D_T$, where $\mathcal D_{T^*}$ and $\mathcal D_T$ are the associated defect spaces for $T$. We also show that, under a certain condition, a contraction $T$ with $\mathcal D_T \subseteq \mathcal D_{T^*}$ is hyponormal.

comments
Preliminary version. 15 pages
fields
Functional Analysis, Complex Variables
pdf ↗ page arXiv:2610.35880 ↗

We show that the Buzano-Dragomir inequality holds for quaternionic Hilbert spaces. Using this generalization, we derive the Dragomir inequality for the quaternionic numerical radius and Deutsch entropic uncertainty principle for quaternionic Hilbert spaces.

comments
14 Pages; 0 Figures, Quaternionic Kraus Conjecture and Quaternionic Crouzeix Problem are added
fields
Functional Analysis, Mathematical Physics
pdf ↗ page arXiv:2610.35881 ↗

A question of Bourgain and Brezis, later restated as Open Problem 7.1 by Brezis, asks whether the $H^{1/2}+W^{1,1}$ norm of a mean-zero scalar function is controlled by the ambient $H^{-1/2}+L^1$ sum norm of its gradient. We show that this estimate fails in every dimension $N\geq 2$. More precisely, for every integer $K\geq 16$, we construct a smooth real-valued mean-zero function $Φ_K$ on $\mathbb{T}^N$ such that $|\nablaΦ_K|{H^{-1/2}(\mathbb{T}^N)+L^1(\mathbb{T}^N)}\leq C$ and $|Φ_K|{H^{1/2}(\mathbb{T}^N)+W^{1,1}(\mathbb{T}^N)}\geq cK^{1/4}-C$, where the absolute constants $c,C>0$ are independent of $N$ and $K$. Equivalently, the ratio of the gradient-constrained decomposition cost to the ambient decomposition cost is unbounded. The construction applies the Hodge projection to a smooth approximation of a normal vector measure on a codimension-one sheet. Its potential records lacunary partial sums at geometrically separated distances from the sheet. A scale-path lower bound, combined with an odd critical Hardy inequality, excludes uniformly bounded $H^{1/2}+W^{1,1}$ decompositions.

comments
18 pages, no figures
fields
Functional Analysis, Analysis of PDEs
pdf ↗ page arXiv:2610.35882 ↗

We extend the wavelet and tent space characterization of $h^1(\mathbb{R}^n)$ to $0<p\le 1$ via a local dyadic tent space $t_p$, and derive the unconditional $L^2$-convergence of the resulting atomic decomposition, with an exact error identity refining Ding, Jiang and Zhu (2019).

comments
30 pages. Split from arXiv:2608.14960, which now carries the critical molecular theory and the maximal admissible class for Goldberg-type splittings
fields
Functional Analysis, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.35886 ↗

We determine the asymptotic order of the observable diameter of $\ell^p$-product spaces associated with a probability measure on $[0,1]$ whose density has a power-type interior zero. Different vanishing orders are allowed on the two sides of the zero. For every $1 \le p \le \infty$, we obtain sharp two-sided estimates and show that the asymptotic order is governed by the larger of the two vanishing orders. In the case $p=2$, the transition occurs exactly at the threshold for the spectral gap.

comments
32 pages
fields
Functional Analysis, Metric Geometry, Probability
pdf ↗ page arXiv:2610.35889 ↗

In this paper, we study the algebraic properties of $\mathcal{S}_θ$---the class of truncated Toeplitz plus truncated Hankel operators --- on the model space $\mathcal{K}_θ$. We also give a complete characterization of the intersection class of truncated Toeplitz and truncated Hankel operators under certain conditions. As applications, we describe the non-uniqueness of a decomposition $T=A+B$, show that self-adjoint operators admit self-adjoint decompositions, give a necessary and sufficient inclusion criterion for $\mathcal{S}_θ$ when $θ$ is a finite Blaschke product, and, from a first-order commutator characterization of $\mathcal{S}_θ$, describe the products of its elements and the operators commuting with $S_θ+ S_θ^{*}$.

comments
This is a preliminary version. Pages 25
fields
Functional Analysis, Complex Variables
pdf ↗ page arXiv:2610.35897 ↗

In this paper, we introduce and study the notion of character cyclic amenability for Banach algebras, which combines the cyclicity condition on derivations with the character-dependent module structure arising in character amenability. We establish several basic properties of this notion and investigate its behavior under standard Banach algebra constructions. In particular, we study character cyclic amenability for opposite algebras, projective tensor products, l1-direct sums, quotient algebras, dense homomorphic images, and theta-Lau products. We also provide examples illustrating the relationship between character cyclic amenability and other amenability-type properties, including examples showing that character cyclic amenability does not imply cyclic amenability. Finally, we characterize character cyclic amenability for module extension Banach algebras in terms of phi-cyclic derivations and derivations associated with the underlying Banach module.

fields
Functional Analysis
pdf ↗ page arXiv:2610.35901 ↗

We prove that a Lipschitz-free space $\F(M)$ fails Kwapień's property (K) whenever its underlying metric space $M$ contains a bi-Lipschitz copy of $C\times K$, where $C\subset\R$ is closed with positive measure and $K$ contains a Cantor metric space. Consequently, this holds for every subset of $\R^n$, $n\geqslant2$, whose closure has positive measure, and for a compact planar set of Assouad dimension one. These Lipschitz-free spaces embed into no strongly weakly compactly generated space and no $L_1$-space; in the separable case their duals are not isomorphic to $\ell_\infty$. Combining our result with a decomposition of Aliaga and Medina, we show that $\F(\R^n\times\Z^m)$ is strongly weakly compactly generated if and only if it has (K), and this occurs precisely when $n\leqslant1$. In particular, $\F(\R^2)$ does not embed isomorphically into $\F(\R\times\Z)$, answering a question of Aliaga. The planar examples nevertheless have the Dunford–Pettis property and Pełczyński's property $(V^*)$ and are weakly sequentially complete.

fields
Functional Analysis
pdf ↗ page arXiv:2610.36194 ↗

Noninjective multiplication symbols allow commuting operators to mix function values along their fibers, while Hölder regularity constrains this mixing at branch points. We study these operators on the Banach algebra $A_β$, $0<β<1$, defined by a restricted interval-variation condition. For $u(t)=t(1-t)$, we determine the full commutant $\{M_u\}'=\mathscr C_β$, including operators whose reflection coefficients need not belong to $A_β$. Separately, we characterize the operator-norm closure $J_β$ of the regular multiplication–reflection subalgebra of $\mathscr C_β$, obtain two-sided norm-distance estimates, and find a complemented copy of $\ell^\infty/c_0$ in the Banach-space quotient $\mathscr C_β/J_β$. Every $T\in\mathscr C_β$ satisfies $σ(T)=σ_{\mathrm{ap}}(T)=σ_{\mathrm{usf}}(T)=σ_{\mathrm{ess}}(T)$, and Fredholmness is equivalent to invertibility. For multipliers with finitely many full bi-Lipschitz branches, we also characterize their full commutants and establish the same spectral equalities for every commuting operator. For the regular finite-branch algebra, we establish an exact norm-closure criterion using weighted local Hölder defects and separate row-difference conditions, even when several folds share a level.

fields
Functional Analysis
pdf ↗ page arXiv:2610.36217 ↗

Let $1\leq p<\infty$. We prove an $L^p$-operator analogue of Raeburn's symmetric imprimitivity theorem for countable discrete groups. Let $A$ be an $L^p$-operator algebra with a contractive two-sided approximate identity, admitting a nondegenerate isometric representation on a $σ$-finite $L^p$-space. Given commuting free and proper actions of groups $G$ and $H$ on a second-countable locally compact Hausdorff space $Z$, together with commuting isometric actions on $A$, we show that the two corresponding full crossed products, and likewise the two reduced crossed products, are Morita equivalent as Banach algebras. The proof is organized through the linking groupoid. Its main analytic ingredient is a coefficient-valued spatial matrix dilation; in the reduced case we prove that this dilation preserves regularity after a Borel reindexing of the opposite corner. A pairing-valued approximate identity yields fullness and nondegeneracy.

fields
Functional Analysis, Operator Algebras
pdf ↗ page arXiv:2610.36677 ↗

We construct a bounded closed convex set in $c_0(\mathbb{C})$ on which no nonzero continuous complex-linear functional attains its maximum modulus. Consequently, this answers in the negative the two longstanding questions raised by Deville, Godefroy and Zizler in \cite[p. 35]{DGZ}. The construction uses a small perturbation of the identity on $c_0(\mathbb{R})$.

comments
Comments and suggestions are welcome
fields
Functional Analysis
pdf ↗ page arXiv:2610.36771 ↗

We study how boundary geometry affects $L^q$ approximation by shallow ReLU neural networks on convex bodies $K\subset\mathbb R^d$. The target classes are unit balls of weighted variation spaces that can enlarge classical spaces by incorporating domain geometry into constraints on output coefficients. We establish upper and lower bounds for arbitrary convex bodies and determine the optimal polynomial approximation exponents for all planar convex bodies and their products with polytopes, including cubes and circular cylinders. For $C^2$ convex bodies, we determine the optimal exponents in a dimension-dependent range of $q$ extending beyond $2$. When all principal curvatures are positive, the result extends to every finite $q\ge2$. For these classes, the dependence of the optimal exponent on the domain is captured by a single geometric growth exponent associated with small boundary caps. We also identify sharp weight thresholds for retaining the classical approximation exponent and improve previous weights on Euclidean balls while preserving the classical $L^2$ approximation rate. When the target is given as a finite ReLU network, the upper bounds can be realized by a compressed network using only the original neurons, without increasing the weighted coefficient sum. These results show that boundary geometry can change the optimal exponent even when dimension, error norm, and weighting rule are fixed.

comments
92 pages, 2 figures, 2 tables
fields
Functional Analysis, Metric Geometry
pdf ↗ page arXiv:2610.36854 ↗

The complex polynomial Bohnenblust–Hille inequality bounds the coefficient $\ell_{2m/(m+1)}$ norm of a complex $m$-homogeneous polynomial by its supremum norm, with a constant independent of the number of variables. For each fixed integer $N\ge3$, we characterize the choices of $N$ degree-$m$ monomials for which the inequality holds with optimal constant $1$ for every polynomial supported on them, provided $m$ exceeds an explicit threshold depending only on $N$. This holds precisely when their exponent vectors contain no nontrivial three-term arithmetic progression; equality for a nonzero polynomial then holds only for monomials. We also prove that, in the same degree range, every polynomial with at most $N$ nonzero coefficients that violates the constant-one inequality has a unique coefficient of maximal modulus, whose exponent is the center of such a progression in its support. Exact contractivity also follows from a comparability condition on the two largest coefficients, with a degree threshold that necessarily depends on the number of nonzero monomials.

fields
Functional Analysis
pdf ↗ page arXiv:2610.36887 ↗

Let $R_\infty(n)$ be the maximal Banach–Mazur distance from an $n$-dimensional real normed space to $\ell_\infty^n$. We prove that $R_\infty(n)\ge cn^{2/3}(\log(2n))^{-10}$. Together with the upper bound of Friedland, Naor, and Youssef, this gives $R_\infty(n) = n^{2/3+o(1)}$. The examples are polars of symmetric Gaussian polytopes with $2n$ generators, and the lower bound holds with exponentially high probability. A joint minimization over coefficient representations and positive weights yields an exact second-variation identity and sparse coefficients. A Gaussian projection argument controls all low-rank completions of the recorded matrix. An adaptive choice of exceptional generators then reduces the remaining case to a fixed-shift Gaussian small-ball estimate over two finite nets.

fields
Functional Analysis
pdf ↗ page arXiv:2610.36917 ↗

We characterize when a de Branges–Rovnyak space coincides with a harmonically weighted Dirichlet space. We obtain a necessary and sufficient condition for this identification, which settles a long-standing open problem. We illustrate the result with several examples, that include measures supported on uncountable sets.

fields
Functional Analysis, Complex Variables
pdf ↗ page arXiv:2610.36943 ↗

The aim of this paper is to obtain the characterization of Fourier transform for functions taking values in a separable Hilbert space $H$, defined on the groups $G=\R^n,\,G=\Z/n\Z$. In the first part of the paper, we consider the Fourier transform as a bijective mapping on $S_H$, the space of all $H$-valued smooth rapidly decaying functions on $\R^n$, which naturally extends bijectively to the dual space $S_H'$. Within this framework, we characterize the Fourier transform in terms of the convolution theorem, supplemented by additional structural properties of the Fourier transform. In the second part of the paper, we consider the discrete setting, where the Fourier transform is an operator on $L^1(\Z/n\Z, H)$, and we provide corresponding characterization based on its essential properties. In contrast to the classical setting, a key assumption in our framework involves the interplay between two distinct convolution operations. As a consequence, the structure of our proof diverges notably from the classical case and introduces additional complexities.

comments
26 pages
fields
Functional Analysis, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.36957 ↗

Let $\mathcal{B}_α$ denote the shift operator on the weighted Bergman space $A^2_α(\mathbb{D})$. We prove that, for every integer $k\geq1$, the Beurling-type theorem holds for $\mathcal{B}_α^k$ on $A^2_α(\mathbb{D})$ if and only if $-1<α\leq1$. We also show that, for every $α>-1$, there exists a finite Blaschke product $B$ such that $M_{B}$ admits a cyclic invariant subspace but fails the wandering subspace property.

comments
19 pages
fields
Functional Analysis, Complex Variables
pdf ↗ page arXiv:2610.37035 ↗

We study noncommutative polynomials and functions invariant under linear actions of infinite (compact Lie) groups, with a focus on $SU(2)$ acting on two noncommuting variables. For $SU(2)$, we give a complete description of the invariant noncommutative polynomials: there are no invariants of odd degree, and in degree $2n$ the invariants are exactly the span of the $S_{2n}$-orbit of $(XY-YX)^n$. We also construct an explicit basis recursively from the operation $(A,B)\mapsto XAYB-YAXB$, and show that the space of degree-$2n$ invariants has dimension equal to the $n$-th Catalan number. All proofs proceed by direct computation with the Haar integral.

fields
Functional Analysis
pdf ↗ page arXiv:2610.37185 ↗

Mappings contracting perimeters of triangles are Banach contractions whenever every point of the underlying metric space is an accumulation point. Consequently, applying the perimeter-contraction theorem on a complete metric space without isolated points yields an existence result already covered by Banach's fixed point theorem. This observation applies, in particular, to operators associated with differential and integral equations acting on standard nontrivial function spaces, such as spaces of continuous functions $C[a,b]$, Lebesgue spaces $L_p$, and Sobolev spaces $W^{k,p}$, equipped with their usual norm metrics. Thus, applications extending beyond Banach's theorem for the same operator and metric require a complete invariant domain containing some number of isolated points. Such a domain may be a closed subset of a standard function space, despite the absence of isolated points in the ambient space. The main goal of this paper is to illustrate applications of the fixed point theorem for mappings contracting perimeters of triangles in settings where the associated operators are not Banach contractions. The differential-equation examples admit interpretations as stylized models combining continuous evolution, nonlocal measurement, and quantized feedback. The integer and difference-equation examples illustrate the effects of rounding on equilibria and periodic responses.

comments
34 pages
fields
Functional Analysis, Analysis of PDEs, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.37644 ↗

We prove a new functional inequality for uniformly log-concave measures, stronger than the Poincaré inequality. This new inequality bounds the variance of differences of convex functions and is motivated by the stability theory of optimal transport: for a fixed uniformly log-concave source measure, it provides a sharp stability bound for Brenier potentials with respect to variations of the target measure. Applying this inequality to Gaussian measures, we recover the fact that the Bures-Wasserstein distance between two positive semidefinite matrices is equivalent to the Frobenius distance between their square roots. Our proofs hinge upon linearization arguments for the Monge-Ampère equation governing the quadratic optimal transport problem.

fields
Functional Analysis, Analysis of PDEs, Metric Geometry, Probability
pdf ↗ page arXiv:2610.37659 ↗

Numerical Analysis21

Monte Carlo PDE solvers enable mesh-free pointwise queries, but repeatedly traversing heterogeneous media is costly, particularly in scenes with reusable or dynamically changing materials. We introduce Walk on Multiscale Media (WoMM), which compiles fine-scale transport into reusable, editable and composable transport prefabs. A probabilistic interpretation of multiscale finite-element bases and condensed operators yields coarse transitions retaining fine-scale material effects. Each prefab combines port-level operators, entrance-dependent exit distributions, and Poisson-marked trajectory records, enabling incremental updates for overlapping absorption fields, gradient-field steering through path reweighting and transient diffusion. Integrated with conventional Walk-on-Spheres in a hybrid continuous–discrete solver, these prefabs accelerate repeated pointwise PDE queries and support interactive material edits without repeatedly solving the fine-scale problem.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.35952 ↗

We develop a dynamical tensor-train method for the Wigner–Fokker–Planck equation arising in continuous-variable open quantum systems. The Wigner function is evolved directly in tensor-train form using a projector-splitting integrator, while the nonlocal Wigner pseudo-differential operator is treated through low-rank representations of the potential difference and matrix-free exponential actions. The resulting method avoids full tensor-product computations in high-dimensional phase space. Numerical experiments in one, two, and three spatial dimensions demonstrate the accuracy and scalability of the method, including simulations in a six-dimensional phase space.

comments
35 pages, 8 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.36257 ↗

We study Monte Carlo power iteration algorithms in the context of reactor criticality simulations. Under mild conditions on the underlying physical system and the algorithmic structure, we derive explicit error bounds, which are to the best of our knowledge the first rigorous results for such methods which are ubiquitous in nuclear engineering. This is done by exploiting connections to Sequential Monte Carlo algorithms, and extending recent developments about stochastic neutron transport. We demonstrate the applicability of our results by considering a concrete version of the Monte Carlo power iteration algorithm as well as variance reduction schemes.

comments
29 pages
fields
Numerical Analysis, Probability
pdf ↗ page arXiv:2610.36266 ↗

Nearly incompressible materials arise in soft tissues, elastomers, gels, and other soft solids. Yet, numerical modeling remains challenging because displacement-based formulations can exhibit volumetric locking near incompressibility. Mixed displacement-pressure formulations address this difficulty but require compatible approximation spaces satisfying the inf-sup condition. In mixed formulations, discontinuous pressure spaces are attractive since pressure degrees of freedom can be condensed locally. In high-order isogeometric analysis, however, their Galerkin stability requires pressure-grid coarsening that increases with polynomial degree. Here, we develop a variational multiscale (VMS) stabilized mixed isogeometric formulation that retains this locality while substantially relaxing the restriction. The key observation is that VMS stabilization controls nonconstant elementwise pressure modes and is unconditionally stable with respect to its parameter, provided the elementwise constant pressure component satisfies the inf-sup condition. Consequently, for maximally continuous displacement splines, a pressure grid coarsened by a factor of two satisfies the constant-mode inf-sup criterion independently of polynomial degree. Local condensation then yields a displacement-only system, facilitating use in existing displacement solvers. The formulation is further extended to immersed isogeometric analysis with extended B-splines, pressure-cell aggregation, subdivision-based cut-cell quadrature, and weak enforcement of essential boundary conditions. Discrete inf-sup studies and convergence analyses in the Kirsch and Cook's membrane benchmarks demonstrate stability and show that the immersed treatment preserves mixed stability properties. The proposed framework therefore provides a local, high-order, and robust approach for nearly incompressible problems in mesh-conforming and immersed settings.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.36349 ↗

Based on the complex representation of quaternion matrices, this paper revisits the quaternion tensor singular value decomposition (QT-SVD) under the QT-product and establishes the corresponding $z$-block circulant matrix decomposition. We develop an efficient implementation of the QT-SVD. Numerical experiments show that this implementation is substantially faster than QT-SVD based on the quaternion matrix SVD approach \cite[Theorem~2.3]{AMLzhang2022} while maintaining high numerical accuracy. We further apply the QT-SVD to color video encryption and reconstruction. Experiments on five DAVIS video sequences show that the QT-SVD achieves higher reconstruction quality and shorter decomposition times than the quaternion T-SVD \cite{tsvdarxiv} under the same processing procedure. Both methods exhibit comparable ciphertext statistics and differential sensitivity, while the QT-SVD introduces less additional distortion under the tested input perturbations.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.36550 ↗

Stationary tail computation for high-dimensional reflected Brownian motions (RBMs) remains computationally challenging. Building on NN4MGF's shared-coordinate basic-adjoint-relationship (BAR) framework, we develop Holomorphic Cauchy Networks (HCNs). Native complex-valued rational Cauchy subnetworks enforce holomorphicity on pole-free domains, removing sampled Cauchy-Riemann penalties and their input-derivative graphs. Exact origin anchoring and boundary-coordinate exclusion enforce structural identities, while anchored interactions represent dependent transforms. Under holomorphic-log assumptions, we establish simultaneous value and first-derivative approximation for the constrained architectures and a mixed-derivative lower bound for coordinate-only representations under nonzero cross-covariance. Closed-form real-axis derivatives and sequential backpropagation further control memory. On the same NVIDIA H100 NVL, the tested 30D HCN configuration uses approximately tenfold less peak allocated memory and has 9.7-fold lower mean update time than NN4MGF; extrapolation to the respective scheduled update counts gives a projected full-schedule update-time ratio of 58.2. With 64 million sampled base inputs per high-dimensional run, HCN reaches 50 dimensions on a laptop GPU with 6.367% mean relative error over thresholds 1 through 8. Degree-matched comparisons yield lower reported mean relative errors over the primary evaluation ranges, including 1.422% versus 6.684% on the 2D Harrison non-product benchmark. An exploratory 10D correlated example illustrates the interaction extension.

comments
29 pages, 3 figures, 14 tables. Code available at https://github.com/KehanWang126/holomorphic-cauchy-networks-rbm
fields
Numerical Analysis, Probability
pdf ↗ page arXiv:2610.36558 ↗

Open Data Analysis Tool for Science and Engineering (ODAT-SE) is an open-source Python framework for inverse problems that are solved by repeated forward simulations. It succeeds 2DMAT, which was developed for surface-structure analysis with quantum-beam diffraction and has since been applied to other fields. In version 3, the search algorithms were separated from the application-specific solvers, which became independent packages, and the software was renamed; this paper describes the current version 4.0. The framework connects optimization and sampling algorithms to solvers through a common interface for parameter mapping, constraints, and objective evaluation, so that an existing simulation code can be used with any of the algorithms. Relative to the previously published 2DMAT, it provides two-level MPI parallelism, which distributes candidate evaluations over processes and each evaluation over a parallel solver, checkpoint and continuation facilities, random search, global optimizers from SciPy, tensor-train optimization, and a Bayesian workflow that estimates partition-function ratios, model evidence, effective noise levels, and marginal posterior distributions. We validate the Bayesian workflow against closed-form results for a linear regression, benchmark tensor-train optimization on continuous, binary, and mixed-integer problems, and verify the two-level parallelization by the same optimum for all process decompositions. Taking up the positron-diffraction example of the 2DMAT paper, the workflow identifies the correlated structural solutions and an effective noise scale close to the imposed perturbation, while exposing incomplete convergence of symmetry-related mode weights. Solver packages, templates, and a gallery of examples support the use of the framework as a general platform for inverse-problem analysis.

comments
41 pages, 8 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.36746 ↗

We propose a minimizing-movement framework for constructing parametric finite element methods (PFEMs) for anisotropic curve shortening flows. The cornerstone is an admissible modified energy: a smooth, convex, and iteration-dependent surrogate that majorizes the original anisotropic energy and agrees with it at the current iterate. For any fixed finite element degree $n\geq1$ and any positive surface energy density $γ\in C^2(\mathbb{S}^1)$, the unconstrained one-step problem has a unique solution for any time-step size whenever the parametric derivative of the current discrete curve does not vanish on any mesh element. It also satisfies a one-step discrete energy dissipation inequality. It therefore preserves the variational structure of the underlying $L^2$-gradient flow. The framework recovers a variational interpretation of Dziuk's isotropic scheme and applies directly to the weak anisotropies. For strong anisotropies, a convex splitting removes the weak-anisotropy restriction on the original density. A scalar constraint further extends the framework to area-preserving flows. Numerical experiments exhibit approximately the expected convergence for smooth isotropic and weakly anisotropic examples. They also confirm energy dissipation and area preservation. In the strongly anisotropic regime, the computations produce sharp facet-like profiles but reveal mesh-scale oscillations under the grid refinement.

comments
23 pages, 6 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.36753 ↗

This paper establishes a fully implicit model to investigate the spread of epidemics and studies its dynamic properties, extending the semi-implicit model in Yang et al.(Commun. Nonlinear Sci. Numer. Simul., 152 (2026) 109140 ). We first employ the split-step backward Euler method to address the non-adaptability arising from backward Euler discretization and ensure the global convergence of the numerical solution. By linearizing the model, the computational cost is reduced at the expense of a slight increase in error, effectively avoiding complex iterative solution processes. Then, we derive the existence, uniqueness, positivity, and boundedness of the numerical solution, and obtain its global convergence order. On this basis, the extinction and persistence of the disease are investigated under three different scenarios. The results indicate that environmental noise is a key factor leading to disease extinction. Finally, numerical simulations visually verify the obtained theoretical conclusions.

comments
19 pages, 8 figures
fields
Numerical Analysis, Probability
pdf ↗ page arXiv:2610.36924 ↗

Walk-on-Spheres (WoS) enables mesh-free solutions of elliptic PDEs, but conventional methods repeatedly discard sampled trajectories, limiting their efficiency in interactive applications. We introduce reusable transport skeletons, a persistent representation that organizes stochastic walks into a shared computational graph. Our framework enables efficient reuse across spatial queries and boundary conditions, while supporting progressive refinement, structure-preserving compression, and local adaptation to dynamic geometry. It further enables interactive editing of transport between selected regions. Experiments demonstrate substantial computational savings while preserving solution accuracy. By transforming transient stochastic trajectories into reusable transport structures, our method provides a flexible foundation for interactive Monte Carlo PDE solving.

comments
19 pages,15 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.36931 ↗

In this paper we propose a GMRES-accelerated version of the Method of Alternating Projections (MAP) for subspaces. It is proved that the accelerated version not only converges to a feasible solution, but also solves a best approximation problem. Comparisons are made with other accelerations proposed in the literature, for solving saddle point problems.

comments
18 pages, 3 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.36938 ↗

We develop a multilevel Monte Carlo (MLMC) estimator to compute statistical quantities of interest of solutions to parameter-dependent nonlinear eigenvalue problems. The main difficulty for this problem is that parametrized eigenpairs often lack a globally well-defined parameter-to-solution map due to crossings and bifurcations. These phenomena render eigenvalue-tracking infeasible in multi-dimensional parameter domains. To overcome this issue, we propose both an Eulerian and a Lagrangian perspective on parameter-to-solution maps for nonlinear eigenvalue problems, which remain well-defined even in the presence of local crossings and bifurcations. Their implementation is based on a straightforward modification of Beyn's contour integral approach. Building on these maps, we propose an MLMC estimator that does not rely on matching eigenvalues or eigenvectors across samples or discretization levels. Nested approximation spaces and inter-level prolongation are only required when statistical quantities of interest related to the eigenspaces are of interest, but not for quantities of interest related to the eigenvalues. We demonstrate the feasibility and validity of our approach by numerical examples using a quasi-Monte Carlo method, including a nonlinear formulation of the Laplace Dirichlet eigenvalue problem on random domains in three dimensions motivated by the work of Steinbach and Unger. To the best of our knowledge, this is the first work considering the approximation of statistical quantities of interest for random nonlinear eigenvalue problems.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.36955 ↗

This work proposes a hyper-reduced order model for coupled hydro-mechanical simulations of elastoplastic deformation and water flow in partially saturated soils. These simulations are used to monitor and assess the safety of geotechnical structures such as earthfill dams, where calibration, uncertainty quantification and scenario analysis require solving the same transient model many times for different parameter values. The displacement and the pore pressure are approximated in a field-split reduced basis obtained by proper orthogonal decomposition, and the Galerkin-projected equations are integrated with decomposed cubature rules, one per balance equation, built with the energy-conserving sampling and weighting (ECSW) method. Each residual is split into an affine part and a nonlinear correction: the affine part is computed exactly, at negligible online cost, so that the cubature rule only needs to approximate the correction, which is localized in space. The construction is independent of the constitutive model choices and is implemented in FEniCSx by assembling the original variational forms on the small set of elements selected by ECSW. The method is tested on two- and three-dimensional dams retaining tailings and water, simulating a raise of the dam and the subsequent consolidation phase, with the friction angle, stiffness and hydraulic conductivity of the dam material and the height of the raise as varying parameters. The hyper-reduced model reproduces the displacement and pressure fields with errors below $0.5\%$ with respect to the finite element solution, achieving speed-ups of up to $43\times$ in two dimensions and $77\times$ in three. These performance gains grow with the extent of the plastic yielding, and the selected elements concentrate in the plastified and unsaturated zones, whose size is set by the physics rather than by the mesh resolution.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.37060 ↗

The Bayesian approach to inverse problems provides a powerful framework for inferring unknown parameters from noisy observations. Among the computationally most challenging statistical inference problems are problems with intractable likelihoods, especially if no unbiased likelihood estimator is available. We focus on a class of such problems, where a biased likelihood estimator is obtained through a stochastic approximation of a forward map (stochastic solver). Obtaining an empirical posterior approximation in applications is very expensive due to the repeated stochastic solver evaluations required by standard sampling methods. To address this computational bottleneck, we develop a novel algorithm based on a density-tempered marginalized Sequential Monte Carlo (SMC) sampler. Our approach (hierarchical SMC) accelerates inference by leveraging a hierarchy of stochastic solvers with increasing accuracy both along the tempering sequence and within mutation stages of the SMC sampler. We establish, under additional assumptions, that the asymptotic speedup is proportional to the number of tempering stages $T$. We demonstrate computational gains on several examples, including a nonlinear, high-contrast inverse problem arising in photoacoustic medical imaging. On this problem, we demonstrate a speedup by a factor of 14.3, reducing the inference wall-clock time from about 4.6 days to 7.7 hours on an NVIDIA A40 GPU.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.37082 ↗

We develop a spectral Barron regularity framework for nonlinear operators between infinite-dimensional function spaces. Using Hilbert-valued Fourier representations of finite-dimensional input truncations, we establish bounds that are uniform in the input truncation dimension and prove arbitrary-order Barron regularity for a class of elliptic solution operators. Second, we use this operator Barron regularity to derive approximation error bounds for an FNO-type neural operator, with explicit dependence on the input truncation dimension, output reconstruction dimension, and network width. For a fixed input truncation dimension, optimizing the neural approximation bound over the Barron regularity order yields stretched-exponential decay of the approximation error in the network width. We also show some numerical experiments to support the predicted rate.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.37101 ↗

We construct explicit quantum circuits for the velocity-stress formulation of elastic wave equations in homogeneous media using Schrödingerisation. After rewriting the equations in symmetric matrix form(SMF), we use Fourier spectral discretization for source-driven periodic problems and staggered-grid finite differences for Dirichlet, Neumann, and impedance boundary conditions. The Schrödingerisation method transforms any linear PDEs into a system of Schrödinger-type PDEs -with unitary evolution- using the warped phase transformation that maps the equations in one higher dimension. For periodic problems, we develop circuits by combining controlled Fourier-mode rotations. For physical boundaries, interleaved encoding swaps the spatial and field-component registers, enabling a more efficient circuit construction using shift operators. We analyze $p$th-order Trotter-Suzuki formulas and derive splitting-error and CNOT-gate estimates, under the assumptions of smooth auxiliary initialization and solution-state recovery. Comparison with classical partitioned splitting methods identifies a polynomial improvement in spatial-resolution dependence for the quantum state-evolution subroutine. Noiseless circuit simulations of a forced periodic problem and an impedance problem illustrate the constructions and quantify their errors against exact solutions and classical reference computations.

comments
36 pages, 16 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.37247 ↗

This paper develops reliable a posteriori error estimates for finite element approximation of an evolution equation with a positive-type memory kernels on bounded convex polygonal or polyhedral domains, including both smooth and weakly singular kernels. For the weakly singular kernel $t^{α-1}/Γ(α),\; 0<α<1,$ the model captures the transition from parabolic diffusion to wave propagation, recovering the corresponding limiting equations as $α\to0^+$ and $α\to1^-$, respectively, with the memory term governing the intermediate dynamics. The spatial discretization employs time-dependent finite element spaces, while the backward Euler method combined with numerical quadrature is used for time integration. A unified \emph{a posteriori} analysis is established for both semidiscrete and fully discrete approximations. Elliptic reconstruction yields optimal spatial error estimates, while a memory-lag estimate preserves the essential cancellation in the convolution term, yielding a first-order temporal estimator whose kernel dependence is characterized by its integral. The resulting estimators are robust, with constants independent of the final time, spatial mesh size, and time step. Their kernel dependence involves only computable quantities, avoiding exponential growth in time. Moreover, for the weakly singular kernel, the $α$-dependent factors remain uniformly bounded for $α\in(0,1).$ Numerical experiments on a convex domain validate the theoretical results, and a space-time adaptive algorithm is also demonstrated on an L-shaped domain featuring a corner singularity and two temporal layers.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.37264 ↗

We develop a multifidelity black-box variational inference (MF-BBVI) approach for PDE-constrained Bayesian inverse problems in which forward solves are expensive and PDE gradients are unavailable. The method is built on score-function estimators for the evidence lower bound (ELBO), which avoid differentiation through the PDE model but are known to suffer from high variance. To address this high variance, we employ multifidelity approximate control-variate estimators that combine a small number of high-fidelity samples with a larger number of low-fidelity samples to reduce estimator variance and the number of high-fidelity evaluations required for optimization. We further develop a PDE-derivative-free second-order optimization strategy based on both gradient and Hessian score-function estimators. For low-fidelity model generation, we propose an on-the-fly strategy that builds and adaptively updates a reduced-order model (ROM) as the variational distribution evolves throughout the optimization. We investigate strategies for adapting the ROM and estimating the approximate control-variate gains, examining tradeoffs between estimator bias, variance reduction, and computational cost. The resulting approach facilitates derivative-free variational inference for large-scale inverse problems through a reduction in overall computational cost. We demonstrate the method on four numerical examples: a linear benchmark with an exact posterior and three PDE-constrained inverse problems.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.37359 ↗

We develop an $\mathcal{H}_2$ model reduction framework for polynomial bilinear systems with quadratic outputs. We define the associated $\mathcal{H}_2$ norm and show its consistency with the established norm in the original (non-polynomial) BQO case. We derive first-order necessary conditions for $\mathcal{H}_2$ optimal model reduction of these polynomial systems and use them to construct an efficient iterative algorithm for computing reduced BQO models that approximately satisfy these conditions. Numerical results on two examples demonstrate the accuracy and effectiveness of the proposed reduction approach.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.37387 ↗

We develop a phase-type (PH) representation of strong stability preserving (SSP) Runge–Kutta (RK) methods. This representation provides a probabilistic interpretation of the RK coefficients and translates the chain order conditions into absorption-time moment identities. Using moment ratio bounds for PH distributions, we prove the conjectured optimal SSP coefficients for general implicit RK methods of orders two and three. The equality characterizations also establish uniqueness of the optimal RK tableaux.

fields
Numerical Analysis, Probability
pdf ↗ page arXiv:2610.37425 ↗

Parametric PDEs in science and engineering are often posed on irregular meshes with geometries that may contain holes or complex boundaries, yet popular operator-surrogate architectures such as the Fourier Neural Operator (FNO) rely on structured geometries for fast computational kernels. We introduce Adaptive Modal Operator Refinement (AMOR), which replaces FNO's Fourier-based spectral convolutions with reduced-basis blocks and supports both data-driven and prescribed input/output bases through a common encode/decode interface. AMOR thus applies unmodified to arbitrary meshes and, because its spectral parameterization has no tensor-product dependence on the spatial dimension, it scales favorably to three dimensions, where FNO's spectral weights grow as $m^3$ with the number of spectral modes $m$. We prove that AMOR is a universal approximator of continuous operators. On two-dimensional structured-grid benchmarks, AMOR is competitive with FNO overall and surpasses PCANet: it outperforms FNO on nonlinear diffusion and Helmholtz problems, while a static snapshot basis keeps it from matching FNO's accuracy on the time-dependent Kolmogorov flow problem. On three-dimensional problems it matches or exceeds FNO's accuracy at lower inference cost on linear elastostatics while trailing FNO on a turbulent lid-driven-cavity flow and scales favorably to problems demanding high capacity. On irregular meshes, where FNO is not directly applicable, AMOR reduces the relative $L^2$ error over the FNO extension Geo-FNO by $31\%$ (airfoil) and $42\%$ (elasticity). Finally, AMOR maintains strong performance across three physically distinct, experimentally motivated loading regimes on a topologically nontrivial three-dimensional aircraft-rib mesh, demonstrating its applicability to realistic engineering problems without modification.

code
github.com/leonardvetter/AMOR repository not found
comments
50 pages, 13 figures, 11 tables. Submitted to Computer Methods in Applied Mechanics and Engineering. Code: https://github.com/leonardvetter/AMOR; data: https://doi.org/10.5281/zenodo.18815516
fields
Numerical Analysis
pdf ↗ page arXiv:2610.37485 ↗

Differential Geometry19

Let \(G\) be a compact Lie group acting Hamiltonianly on a symplectic manifold \(M\) with moment map \(μ:M\to\g^*\). For a singular value \(a\) of \(μ\), the symplectic quotient \(μ^{-1}(a)/G\) is stratified and the classical Kirwan map is not defined. We introduce a \emph{formal neighborhood equivariant cohomology} \(\cH_G^*(a)\) using the slice-groupoid Cartan sheaf and the closed embedding \(μ^{-1}(a)/G\hookrightarrow μ^{-1}(U_a)/G\) for a \(G\)-invariant neighborhood \(U_a\) of \(a\). We define a formal Kirwan map \[ κ_{\mathrm{formal}}:H_G^*(M)\longrightarrow \cH_G^*(a). \] We prove that \(κ_{\mathrm{formal}}\) is surjective under a local \emph{equivariant homotopy retraction} condition on the slice level sets \[ C_x=μ_x^{-1}(a|_{\g_x})\subset S_x, \] which is satisfied, for example, when \(a\) is a regular value of every slice moment map, or when \(a=0\) and \(C_x\) is a \(G_x\)-equivariant cone. The weakest condition we identify is the existence, for each \(x\inμ^{-1}(a)\), of a local quasi-isomorphism \[ \Car(S_x,G_x)\simeq \cC^\bullet_{C_x} \] in a \(G_x\)-equivariant neighborhood of the orbit, without requiring a global tubular neighborhood. As an application we recover, in the formal framework, the even-degree surjectivity theorem of Delarue–Ramacher–Schmitt for \(S^1\)-actions and clarify why odd-degree surjectivity fails in general.

comments
23 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.35936 ↗

Let g be a Lie algebra of dimension 2n+1, let beta be a contact form, and let alpha be a nonzero integrable 1-form. We study when the affine path alpha_t = (1-t)alpha + t beta consists of contact forms with the orientation determined by beta \wedge (d beta)^n. We first establish an algebraic criterion: affine deformability is equivalent to the non-negativity of a scalar obtained from the interaction form Theta(alpha, beta) = alpha \wedge (d beta)^n + n beta \wedge d alpha \wedge (d beta)^{n-1}. For unimodular Lie algebras this scalar is (n+1)alpha(Z_beta), where Z_beta is the Reeb vector. We also derive a formula involving the modular character tau(X) = tr(ad_X) in the non-unimodular case. On nilpotent contact Lie algebras, every integrable 1-form is closed; using the standard structural properties of nilpotent contact Lie algebras, this implies that every left-invariant foliation is an affine origin for every fixed contact form. An explicit three-dimensional example illustrates that the modular term can yield affine deformability even when alpha(Z_beta) = 0. We discuss the implications for left-invariant forms on Lie groups and compact quotients. We do not claim a general obstruction example in the non-unimodular case; establishing such an example requires a separate verified construction.

comments
5 pages, amsart
fields
Differential Geometry, Rings and Algebras
pdf ↗ page arXiv:2610.35940 ↗

In this paper, we establish spectral Bakry-Émery $Γ_2$ criteria and their modified counterparts for the $V$-Laplacian on closed Riemannian manifolds equipped with evolving weighted measures. The estimates yield jointly optimized spectral bounds and sharp endpoint constants. When the measures evolve according to the Fokker-Planck equation, these estimates yield monotonicity formulas for the Fisher information associated with Shannon and Tsallis entropies, together with quantitative propagation of Poincaré, logarithmic Sobolev, and nonlinear interpolation inequalities. In the stationary reversible setting, our results specialize to the Witten Laplacian and recover the unweighted Cai-Wang estimates \cite{Cai}.

fields
Differential Geometry
pdf ↗ page arXiv:2610.36091 ↗

We study conditions for the $L^p$ convergence of the AMV Laplacian $Δ_r$ on completions of Riemannian manifolds $(M,g)$. The condition invloves the Minkowski dimension of the set $\overline{M} \setminus M$ and how much the curvature is blowing up around $S$. We build a concrete family of sets that satisfies these conditions, and provide an example of non convergence when these conditions are not met.

comments
54 pages
fields
Differential Geometry, Analysis of PDEs, Metric Geometry
pdf ↗ page arXiv:2610.36125 ↗

We prove that if $M$ is a (compact, connected, and embedded) proper Dupin hypersurface in a sphere then, viewing $M$ as a submanifold of the ambient Euclidean space, its orthogonal projection to the normal space at any point is convex. This is a generalization of the convexity theorem of Terng for isoparametric submanifolds in Euclidean spaces.

comments
9 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.36366 ↗

We prove that a complete connected four-dimensional gradient steady Ricci soliton with positive curvature operator that is $κ$-noncollapsed at all scales admits an effective isometric $O(3)$ action preserving the potential function and having principal orbits diffeomorphic to $S^2$.

fields
Differential Geometry
pdf ↗ page arXiv:2610.36563 ↗

We establish a unified fixed-point Grassmannian framework for the classical pseudo-Riemannian symmetric spaces from Berger's classification. For a classical group $G=\mathop{\mathrm{Aut}}(V,b)$, we identify $G$ with the open graph-like locus inside an ordinary or maximal-isotropic Grassmannian $\overline{G}$. Natural involutions of $\overline{G}$ are constructed via classical forms, compatible scalar structures, and subspace configurations. We explicitly describe the full fixed loci of these involutions in terms of classical Grassmannians. The corresponding graph-like fixed loci (or their signature strata) are identified, respectively, with the moduli spaces of classical forms, compatible scalar structures, and subspace configurations; these moduli spaces coincide with classical pseudo-Riemannian symmetric spaces. In summary, the 54 classical series are organized into four families: 10 group-type models, 7 classical-form Grassmannians, 20 scalar-structure Grassmannians, and 17 subspace-configuration Grassmannians.

fields
Differential Geometry
pdf ↗ page arXiv:2610.36724 ↗

An exterior free-boundary minimal surface is a properly immersed minimal surface outside the unit ball whose compact boundary lies on the unit sphere and meets it orthogonally. By Mazet and Mendes, in $\mathbb{R}^3$ every such surface with one regular end is a catenoidal piece, and every stable one with parallel embedded regular ends is catenoidal. We construct connected embedded examples with three ends from the Costa surface $M$, placed so that its central saddle point is at the origin. For every sufficiently small $\varepsilon>0$ we perturb $\varepsilon^{-1}(M\setminus B_\varepsilon)$ into a properly embedded exterior free-boundary minimal surface of genus one with one boundary circle and three ends, two catenoidal and one planar. The construction is a small-hole perturbation. The only error of the initial surface is in the boundary condition. The symmetry group of the Costa surface removes all obstructions: at the hole it forbids the logarithmic and the tilting modes, and globally it kills the bounded Jacobi fields of the Costa surface, which is nondegenerate by Nayatani. In the conformal compactification the Jacobi operator is smooth also at the ends, so the linear analysis takes place on a compact torus with one small hole. The same construction applied to the Costa-Hoffman-Meeks surfaces gives such examples of every genus $k\geq1$, of index at least $2k+3$. We also report numerical experiments with two independent implementations of the Costa surface. They confirm the period condition, the symmetries, the boundary error, the index and nullity of the Jacobi operator and the absence of an equivariant kernel. They also compute the free-boundary correction near the hole, which gives pictures of the surfaces.

comments
26 pages, 7 figures. Interactive 3D models: https://surfaces.igorkozyrev.com/#costa . The scripts for the numerical experiments are attached as ancillary files (anc/scripts)
fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2610.36741 ↗

We establish Cartan-Ambrose-Hicks (CAH) theorems for statistical manifolds and characterize statistical symmetric spaces by dual geodesic symmetries. These theorems determine statistical structures up to local statisticmorphism by comparing the curvature of an alpha-connection and the skewness tensor using radial geodesics and parallel displacement of that connection. The Levi-Civita formulation characterizes local statisticmorphisms by comparing the radial Jacobi operators of Riemannian curvature and the skewness tensor using Levi-Civita parallel displacement. The same characterization holds with statistical curvature in place of Riemannian curvature. For Hessian manifolds, comparison of the skewness tensor alone is necessary and sufficient. For connected statistical manifolds, we prove the equivalence between the homogeneous definition of statistical symmetric spaces and the existence of global dual geodesic symmetries at every point. The CAH theorems give the necessary and sufficient conditions for maps constructed from exponential maps of dual connections to be local dual geodesic symmetries, preserving the metric and interchanging the dual connections. Such local symmetries exist at every point if and only if the Riemannian curvature and skewness tensors are Levi-Civita parallel. For Hessian manifolds, this condition is equivalent to the equality of the Hessian and dual Hessian curvature tensors.

comments
55 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.36803 ↗

By Chern–Weil theory, rational Pontrjagin classes are represented by curvature forms. Sprays, which require no metric and include all Finsler and Riemannian geodesic sprays, provide a natural setting for the vanishing problem. Shen and Zhao proved that if a manifold admits a locally projectively flat spray, then all of its rational Pontrjagin classes vanish. We ask whether this rigidity persists for the wider class of projectively affine sprays. We prove that if a smooth manifold admits a locally spherically symmetric projectively affine spray, then all rational Pontrjagin classes vanish. Stronger, the Pontrjagin forms of the associated Berwald spray vanish identically as differential forms, a genuine geometric rigidity. Hence a manifold with nonzero rational Pontrjagin class cannot carry such a spray; in particular, it admits no Riemannian metric whose geodesic spray is locally spherically symmetric. We also construct non-Berwald, non-projectively-flat examples with identically vanishing Pontrjagin forms, showing the phenomenon is not a byproduct of projective flatness.

comments
19 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.36841 ↗

For a Ricci non-negative non-parabolic manifold $(M^{n\ge3},g)$, assuming the minimal normalized Green function admits the asymptotic expansion $G=r^{2-n}+H_p+O_1(r)$, we prove the scaling invariant defect $H_pR^{n-2}$ controls the geometry of the geodesic ball quantitatively, $\frac{1}{R}d_{GH}(B_p(R),B_0^{\mathbb{R}^n}(R))\le C_n \left(H_p R^{n-2}\right)^{\frac{1}{88(n+1)(4n+1)}}$.

comments
35 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.36998 ↗

We study compact non-Kähler Bismut Hermitian-Einstein threefolds whose connected full group of holomorphic isometries acts with cohomogeneity one. We prove that the Lie algebra of this group is isomorphic to $\mathbb{R}^3\oplus\mathfrak{su}(2)$ or $\mathbb{R}^2\oplus\mathfrak{su}(2)$, unless the Bismut connection is flat. We then focus on the first case, where we determine the possible singular isotropy algebras and reduce the Bismut Hermitian-Einstein condition to a system of ordinary differential equations. We obtain explicit solutions satisfying suitable smoothness conditions, that yield families of compact and complete non-compact non-Kähler Bismut Hermitian-Einstein threefolds with non-constant scalar curvature and non-zero Bismut curvature, including examples on $S^3\times S^3$ and $S^3\times S^1\times\mathbb{R}^2$, respectively. We also investigate the topology of the compact examples and determine the asymptotic scalar curvature of the non-compact ones. Moreover, we prove the existence of a smooth deformation of a Bismut-flat Samelson structure on $S^3\times S^3$ through non-Kähler Bismut Hermitian-Einstein structures that are not Bismut-flat. The regularity of the constructed compact examples is also investigated.

fields
Differential Geometry
pdf ↗ page arXiv:2610.37102 ↗

Let $(M^n,g,f)$ be a complete connected shrinking gradient Ricci soliton with constant scalar curvature, satisfying \[ \Ric+\nabla^2f=\frac12g. \] We prove that if the focal set of $f$ is totally geodesic, then $M$ is rigid. Equivalently, its universal cover is a product of a compact Einstein manifold and a Euclidean space.

fields
Differential Geometry
pdf ↗ page arXiv:2610.37151 ↗

For a rotationally symmetric Ricci flow $g(t)=ds^2+f(s,t)^2g_{\mathbb S^{n-1}}$, $n\ge3$, we study the scale-invariant curvature anisotropy $A=1-f_s^2+ff_{ss}$ introduced by Angenent–Knopf. At the critical power $p_*=(n-1)/(n-2)$, with $Φ=sgn(A)|A|^{p_*/2}$, we prove the distributional identity \[ (\partial_t-Δ)|A|^{p_*} =-\frac4{n-1} \left|Φ_s+(n-1)\frac{f_s}{f}Φ\right|^2. \] Integration against a conjugate heat kernel gives an anisotropic entropy formula. As an application, we obtain an entropy threshold for Type-I singularities on $\mathbb S^n$: if the entropy associated with a singular-time kernel is at most $1$ at an earlier regular time, every tangent flow at the corresponding singular point is a round shrinking sphere. This criterion follows from the $0/1$ quantization of the singular-time anisotropic density.

fields
Differential Geometry
pdf ↗ page arXiv:2610.37154 ↗

In this paper, we investigate the classes of \lq\lq unicorns" of general $(α,β)$-metric types, namely, Landsberg general $(α,β)$-metrics that are non-Berwaldian. We provide explicit solutions for these metrics, including several new classes. Furthermore, we identify and exclude the cases that yield degenerate metrics to guarantee the geometric validity of the established solutions.

comments
17 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.37241 ↗

We compute Goldman volume on the moduli space of convex real projective structures on a closed oriented surface of genus at least two, with exponential weight defined by the shifted Blaschke area. For every positive parameter, the integral converges and is a finite sum of inverse powers of that parameter. Its coefficients are intersection numbers involving the bundle of regular cubic differentials on the moduli stack of stable curves. The proof establishes absolute convergence and vanishing signed boundary integrals of the transgression under simultaneous plumbing degeneration. The pushforward of Goldman measure by the shifted Blaschke area therefore has a polynomial density on $[0,\infty)$, and its sublevel volumes are polynomial in the threshold.

comments
Preliminary version; all comments are welcome
fields
Differential Geometry, Algebraic Geometry, Geometric Topology
pdf ↗ page arXiv:2610.37317 ↗

Deep learning algorithms have become widely used and successful across a wide range of tasks. A key challenge is determining whether these algorithms truly capture the intrinsic geometry of their training data in order to better understand what they learn from data. In this theoretical work, we explore a new geometric structure defined on the data space by a trained neural network through the Data Information Matrix (DIM), a variation of the Fisher Information Matrix. This new geometric structure is a foliation and the main focus of this work is to explore the non-smooth case of ReLU activation functions, where traditional geometric tools often fail. The foliation is defined through a distribution spanned by the columns of the DIM. We first prove that this distribution is non-smooth in the sense of distributions, rendering the existence of an associated foliation much more complex. To circumvent this issue, we introduce a smoothed version of this distribution that differs from the original only on a set of measure zero. Finally, we prove that this smoothed distribution induces a singular foliation on the data space, which is based on the data the network has been trained with. Our theoretical contributions provide a geometric framework to better understand how neural networks organize their input space in non-smooth settings, bridging differential geometry and deep learning theory.

comments
18 pages, 4 figures, 2 appendix
fields
Differential Geometry
pdf ↗ page arXiv:2610.37370 ↗

Let $σ(t)=\operatorname{sl}(\varpi t)$ be the lemniscatic sine, normalized to have period $2$. We study the zero set $\mathscr{M}$ of $S(x)=σ(x_1)σ(x_3)-σ(x_2)σ(x_4)$ in $\mathbb{R}^{4}$. We show that $\mathscr{M}$ is a complete embedded minimal hypersurface with infinitely many isolated singularities, which are exactly the vertices of the regular honeycomb of $\mathbb{R}^{4}$ by $24$-cells. Near every singular point $\mathscr{M}$ is an analytic image of the Clifford cone, and $\mathscr{M}$ is a stationary integral varifold with density $π/2$ at each singularity. The isometry group of $\mathscr{M}$ is generated by the translations of the root lattice $D_4$ and a hyperoctahedral group of order $384$ fixing the centre of a $24$-cell; it has index three in the symmetry group of the honeycomb, acts transitively on the singular points, and contains a symmetry which is not visible from the defining equation. Seen from the centre of a $24$-cell, $\mathscr{M}$ is the zero set of a theta series of a translate of the Hurwitz quaternions, which makes the whole symmetry group manifest; the sections of $\mathscr{M}$ by a family of parallel hyperplanes are triply periodic surfaces with cubic symmetry which change topology through conical points. Finally, $\mathscr{M}$ is obtained from a union of $2$-planes by attaching cones over $2$-tori, and its quotient by the lattice of periods is a compact minimal hypersurface in a flat $4$-torus with exactly three singular points and infinite Morse index; we show that its regular part is one-sided and compute its Euler characteristic and integral homology.

comments
39 p
fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2610.37453 ↗

A Brownian particle observed at time $t$ carries information about where it started. On a closed surface this makes the surface itself a statistical model, whose Fisher information pulls back to a heat-smoothed metric $h_t=2tJ_t$; its volume $V_t$ is a normalized Jeffreys volume, and $\log(V_t/2t)$ is the model-dependent constant in Rissanen's minimum-description-length complexity. We ask whether maximizing this statistical size among metrics of fixed area in a conformal class recovers the uniformizing metric. First, $9(A-2πχt-V_t)/t^2$ converges to the curvature energy $\int K^2\,dA$ (the Calabi energy up to a topological constant), uniformly on classes of bounded geometry; hence, within a fixed-area conformal class of bounded geometry containing the uniformizer, approximate maximizers of $V_t$ converge to the constant-curvature metric (on the sphere, to the round metrics). Second, near every flat torus we prove a sharp finite-scale version: for small perturbations whose oscillations are resolved by the diffusion, $t|k|^2\leθ<τ_*$, the flat-subtracted defect is comparable to $\|Δf\|_2^2$, uniformly in the frequency cutoff and over noncollapsing lattices. The explicit threshold $τ_*=4.408\ldots$ is optimal, and it is exactly the observation time at which the information about one curvature mode is maximal. Third, the flat structure is seen only at an exponential scale: the volume loss of a flat torus measures the ambiguity of winding and is comparable to $t^{-1/2}e^{-\operatorname{sys}^2/(16t)}$, which selects the hexagonal lattice as $t\to0$ but not at all times.

comments
49 pages, 6 figures
fields
Differential Geometry
pdf ↗ page arXiv:2610.37520 ↗

Dynamical Systems18

Rigorous existence results for periodic solutions of state-dependent delay differential equations (SDDEs) are scarce outside perturbative regimes, where the equation is close to an ordinary or constant-delay differential equation. We develop a computer-assisted framework that proves the existence of periodic solutions of SDDEs directly. The numerical approximation is a trigonometric polynomial, and a true periodic solution nearby is obtained by a Newton–Kantorovich argument in the space of $C^1$ periodic functions. The main difficulty lies in the terms in which the solution is evaluated at a time that itself depends on the solution. We obtain rigorous enclosures of their Fourier coefficients by adapting validated discrete Fourier transform techniques for analytic nonlinearities. We apply the framework to three representative examples of increasing complexity: a periodically forced toy model, an autonomous scalar equation with two state-dependent delays, and a threshold-type delay model from operon dynamics in gene regulation. To our knowledge, these are the first computer-assisted proofs of existence of periodic solutions of SDDEs outside perturbative regimes. This opens the way to computer-assisted proofs for SDDEs more broadly. All computer-assisted proofs, including the requisite codes and the approximate objects, are accessible on GitHub at \cite{code_proof}.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.36216 ↗

In this paper we study the set $\mathcal{M}_M$ of roots of power series with coefficients in $\{0, \pm 1, ..., \pm M \}$ and constant coefficient $1$. We prove that every nonreal root in the unit disk of a polynomial of this form is an interior point of $\mathcal{M}_M$. Moreover we establish a dimension formula and show the corresponding Hausdorff measure is finite and positive for a family of attractors parameterized by boundary parameters $λ\in \partial \mathcal{M}_M \setminus \mathbb{R}$. Finally we show that $\partial \mathcal{M}_M$ has Lebesgue measure zero.

fields
Dynamical Systems, Complex Variables
pdf ↗ page arXiv:2610.36279 ↗

We give a new and elementary\footnote{That is: without functional analysis.} proof that certain smooth, complete, expanding maps from the interval to itself are mixing.

comments
7 pages, 2 figures
fields
Dynamical Systems
pdf ↗ page arXiv:2610.36292 ↗

Iterated Weierstrass P functions exhibit relationships between period lattices and the dynamics. We connect the dynamical properties to the occurrence of integer and rational points on associated real elliptic curves. Defining a complex number z to be an integer point if both P(z) and P'(z) are integers, we restrict to P when the lattice is real and its invariants g2 and g3 are integers. We study when an integer point lies in the Fatou set of P and when it lies in the Julia set, and if there is any dynamical significance to this. Applying established results on integer solutions to cubic Diophantine equations to the dynamical properties of iterated elliptic P functions, we show integer points at poles of P lie in the Julia set. Integer points can also occur in the Fatou set of P.

fields
Dynamical Systems, Algebraic Geometry
pdf ↗ page arXiv:2610.36323 ↗

For closed negatively curved surfaces, we obtain formulas for the derivatives of the harmonic entropy of its geodesic flow and the linear drift of its Brownian motion along conformal deformations. We deduce that the harmonic entropy of the geodesic flow of a closed surface of non-constant negative curvature (resp. linear drift of the Brownian motion on its universal cover) is strictly decreasing (resp. increasing) along the normalized Ricci flow. We also prove that any hyperbolic metric on a surface admits a neighborhood in the space of conformal negatively curved metrics in which the linear drift is greater or equal to the Liouville entropy, with equality if and only if the metric is hyperbolic.

comments
12 pages
fields
Dynamical Systems, Differential Geometry, Probability
pdf ↗ page arXiv:2610.36354 ↗

We expand our discussion of the almost stability theorem of Dicks and Dunwoody in the context of probability measure preserving equivalence relations.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.36451 ↗

This note describes a novel approach to systematically characterize dynamics based on finite information. We present an algebraic framework for dynamics from which it is possible to deduce more traditional descriptors associated with the qualitative structures of the geometric theory of dynamical systems. The purpose of this paper is to indicate how this approach can be used to obtain interpretable characterizations of dynamics that are robust with respect to perturbations, in the context of models of multiscale and data-driven systems.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.36532 ↗

We investigate an impulsive Hopfield-type neural network with a generalized piecewise constant argument, including the nonimpulsive case in which all jump terms vanish. The piecewise constant argument is given by $γ(t)=ζ_k$ on each interval $[t_k,t_{k+1})$, where $t_k\leqζ_k\leq t_{k+1}$, $k\in\mathbb{N}_0$. Depending on the position of $ζ_k$, the argument may be advanced, retarded, or of mixed type. By representing periodic solutions through a Green function and applying a fixed-point argument, we obtain explicit and verifiable sufficient conditions for the existence and uniqueness of an $ω$-periodic solution. We then use a Gronwall-type estimate, together with period averages, to prove that this solution is globally exponentially attractive. These conditions can be checked directly from the system parameters, without requiring explicit knowledge of the periodic solution. In the course of the analysis, we also revisit and correct some estimates presented by Pinto et al. \cite{Pinto_ejqtde}. The case $ζ_k=t_k$ reduces to their setting, whereas our results cover the more general choice of $ζ_k$ within each interval. We also present examples illustrating different types of piecewise constant arguments.

comments
32 pages, 6 figures
fields
Dynamical Systems
pdf ↗ page arXiv:2610.36560 ↗

Building on the seminal work of Eskin, Mirzakhani and Mohammadi in the context of dynamics on moduli spaces, we introduce linear loci and linear measures to provide a relatively basic description of orbit closures and invariant measures in strata of translation surfaces. We use this to discuss geometrical properties of orbit closures and invariant measures from the perspective of geometric structures.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.36893 ↗

The Wilson-Cowan model is a classic example of a ``neural mass model'' - a simplified yet powerful way to describe the activity of large networks of neurons. It is particularly interesting because, even in its relatively simple form, it can produce a surprising variety of behaviours, from steady rhythms to complex oscillations. A key factor influencing these dynamics is the presence of time delays, reflecting the time it takes for signals to travel between neurons. This article offers an accessible introduction to the Wilson-Cowan model, focusing on how these time delays affect its behaviour. We explore the core concepts behind the model, survey the existing literature, and introduce some numerical techniques specifically designed for analysing such delay differential equations. We also provide a practical, hands-on tutorial on using the DDE-BIFTOOL software package, demonstrating its application to the Wilson-Cowan model. Downloadable code allows the reader to reproduce our results and to explore the model's dynamics further, including MATLAB code for all computations, using DDE-BIFTOOL for the continuation, and Python versions of simulation codes.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.36920 ↗

We establish a complex one-dimensional fractal transference principle connecting density combinatorics with the fractal geometry of complex continued fractions. Our framework applies to a broad class of conformal iterated function systems indexed by lattice points, including those associated with the Hurwitz continued fraction, and transfers positive-density combinatorial patterns to digit sets while preserving the maximal Hausdorff dimension. As applications, we obtain fractal counterparts of the multidimensional and polynomial Szemerédi theorems, as well as their relativized versions yielding a counterpart of the multidimensional Green–Tao theorem.

comments
34 pages, 3 figures
fields
Dynamical Systems, Number Theory
pdf ↗ page arXiv:2610.37119 ↗

In a recent preprint, we found counterexamples to a number of auxiliary conjectures in a 2012 paper of Mirtabatabaei and Bullo, concerning convergence in two fundamental models of heterogeneous Hegselmann-Krause dynamics, the synchronized bounded confidence (SBC) and synchronized bounded influence (SBI) models. Building on our work, Etesami and Touri presented an example of an opinion vector which doesn't converge under SBC dynamics, resolving the main conjecture from the 2012 paper in that case. Here we complete this project by presenting an example of an opinion vector which doesn't converge under SBI dynamics. The example was found with the help of AI.

comments
6 pages, 1 figure
fields
Dynamical Systems, Combinatorics
pdf ↗ page arXiv:2610.37246 ↗

In this paper, we develop a general criterion for the K-property of equilibrium states within the Climenhaga-Thompson framework. Under suitable pressure-gap and non-uniform expansiveness assumptions, we show that the unique equilibrium state has the K-property, provided that the good collection of orbit segments satisfies an appropriate non-uniform specification property, as well as an appropriate non-uniform distortion control property for the potential function. These results do not require any additional lambda-decomposition structure and apply to a wide range of systems, including those with non-trivial central directions, such as the frame flow over closed rank-one manifolds with bunched nonpositive curvature. Moreover, we also construct several counterexamples to demonstrate that generally the assumptions in our criteria cannot be weakened while keeping the K-property.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.37392 ↗

We establish the variational characterization of the measure of maximal entropy on noncompact manifolds of nonpositive sectional curvature. Assuming that the deck transformation group contains an axial isometry with a rank-one axis and has a finite critical exponent, we prove that the supremum of the entropies of invariant probabilities equals the critical exponent and is attained if and only if the Bowen-Margulis-Sullivan (BMS) measure is finite. In that case, the measure of maximal entropy is unique and equals the normalized BMS measure. The proof combines conditional-entropy rigidity on the regular set with an HTS-based entropy exclusion on the singular set, overcoming both non-compactness and the loss of uniform hyperbolicity due to zero curvature. We also obtain mixing under a non-arithmetic length-spectrum assumption. For nonpositively curved surfaces of genus at least two, the existence of a closed Jacobi rank-one geodesic and the non-arithmeticity of the axial length spectrum are automatic.

comments
40 pages, 1 figure
fields
Dynamical Systems, Differential Geometry
pdf ↗ page arXiv:2610.37439 ↗

We establish uniform multicritical pseudo-Siegel bounds for polynomials of Zakeri type, where all free critical points are on the boundary of an invariant Siegel disk of bounded type. This generalizes the unicritical result~\cite{DL:wZ}. As an application, we show the optimality of the Brjuno condition under the Zakeri restriction: as the rotation number becomes non-Brjuno, the Siegel disk evolves into the Cremer Hedgehog. Combined with~\cite{BCR26}, this implies the optimality of the Brjuno condition for all neutral polynomials, resolving a conjecture by Douady from the 1980s.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.37468 ↗

This article primarily investigates the Assouad dimension of graphs of continuous functions. Initially, the Assouad spectrum of the graph of a Hölder and count-Hölder continuous function has been derived. It has been shown that the Assouad dimension of the graph of a continuous function $f:[a,b]\to \mathbb{R}$ with the count-Hölder condition is at least \( 1 + \sup \left\{ \overline{\dim}_{\mathrm{B}}L_f(y) : y \in \mathscr{R}_f[a,b]\right\} \) where $\overline{\dim}_{\mathrm{B}}$ denotes the upper box dimension, $L_f(y)$ is the level set of $f$ at level $y$, and $\mathscr{R}_f[a,b]$ is the range of $f$ on $[a,b]$. As a consequence, non-trivial lower bounds are obtained for the Assouad dimension of the graphs of classical functions such as the Weierstrass and Takagi functions. We have also employed a unified projection solely in the direction of the $y$-axis to establish lower and upper bounds for the Assouad dimension. On this basis, we have shown that the lower bound is sharp in terms of level sets, namely the upper box dimension cannot be replaced by the Assouad dimension. This provides a partial positive answer to Question 17.11.1 posed by Fraser in~\cite{Fraser2021}, and to Question 2.2 posed by Yu in~\cite{Yu2020}. Building on this result, the analysis has been further extended to a broader class of Weierstrass-type functions and some self-affine functions like fractal interpolation functions. Moreover, we have discussed certain circumstances under which the Assouad dimension can reach 2 with several examples constructed. A supplementary explanation of why Brownian motion has the full Assouad dimension has also been provided based on the preceding work of Howroyd and Yu \cite{Howroyd2019}.

fields
Dynamical Systems, Classical Analysis and ODEs, Combinatorics
pdf ↗ page arXiv:2610.37477 ↗

In his global theory of one-frequency Schrödinger operators~\cite{Av15}, Avila asked whether the acceleration of a real-analytic $\SL(2,\R)$ cocycle over an irrational circle rotation is bounded below by the absolute value of its topological degree and whether equality in nonzero degree characterizes premonotonicity. We answer this question affirmatively for every irrational frequency, without arithmetic restrictions. Here premonotonicity means that some iterate becomes strictly monotonic in the base variable after a periodic change of coordinates~\cite{AK15}. In the positive Lyapunov exponent regime, this provides a nonzero-degree counterpart to Avila's characterization of uniform hyperbolicity by vanishing acceleration. In nonzero degree, we also characterize minimal acceleration through signed connections and invariant positive matrix-valued measures. In the positive-degree, positive Lyapunov exponent case, the proof constructs an analytic positive connection using invariant positive matrix-valued measures and a transfer–scattering representation. A complex-analytic construction then produces a periodic gauge. As a further consequence, for each fixed irrational frequency and nonzero degree, the premonotonic locus is path connected: every cocycle in this locus can be joined to the linear rotation of the same degree by a path preserving premonotonicity.

fields
Dynamical Systems, Spectral Theory
pdf ↗ page arXiv:2610.37479 ↗

We review the following question: when do two polynomial automorphisms of C^2 have the same maximal entropy measure? While the answer has been known for some time when both automorphisms are compositions of Hénon maps, the general case is more delicate. In this (mostly expository) note we explain how recent results in random dynamics give an almost complete solution to the question, and point out a number of related open problems.

fields
Dynamical Systems, Complex Variables
pdf ↗ page arXiv:2610.37600 ↗

Representation Theory14

We prove an explicit correspondence between the irreducible finite dimensional representations of the mirabolic quantum group $MU(n)$ and those of the mirabolic Hecke algebra $MH_d$ given by the mirabolic quantum Schur-Weyl duality. In particular, this resolves a generalization of a conjecture by the second author. The proof is via a careful analysis of the $MU(n)$-weight spaces and the eigenvalues of the Jucys Murphy elements of $MH_d$ on the mirabolic tensor space.

comments
35 pages. This is the revised version of the second half of arXiv:2510.07469v1 , which was split into two papers, the revised first half is arXiv:2510.07469v2
fields
Representation Theory, Quantum Algebra
pdf ↗ page arXiv:2610.35943 ↗

We study the behavior of various delooping levels in the setting of Morita context algebras with zero bimodule homomorphisms. Building on the notion of syzygies for modules over such algebras, we establish conditions under which the delooping level (see [6]) of a Morita context algebra is finite, provided that the component algebras also have finite delooping levels. We further extend these results to other invariants, the global delooping level (see [2]) and the derived delooping level (see [8]), showing how their finiteness is preserved in the Morita context. Several examples illustrate both the applicability of our criteria and the necessity of the hypotheses, highlighting cases where the delooping level of the whole algebra can be smaller than that of its components.

fields
Representation Theory
pdf ↗ page arXiv:2610.36157 ↗

We prove multiplicity at most one for the unitary similitude Ginzburg–Rallis model over local fields of characteristic zero, for both the quasi-split and non-quasi-split groups. The theorem holds for all irreducible smooth representations over non-Archimedean fields and all irreducible Casselman–Wallach representations over the real field. We first describe the decomposition into parabolic double cosets and compute the required stabilizers. Over a non-Archimedean field we use character obstructions, localization, and transpose symmetry. Over the real field we use a Casimir identity and explicit normal vectors to prove the corresponding statement for distributions with prescribed infinitesimal character. The Gelfand–Kazhdan criterion and a dualizing involution then give the multiplicity bound.

fields
Representation Theory
pdf ↗ page arXiv:2610.36294 ↗

For a $d$-dimensional vector space $V$ over a field of characteristic zero, we show that $GL(V)$-stable ideals of the universal enveloping algebra of the free two-step nilpotent Lie algebra $\mathfrak n_2(V)$ are in natural correspondence with filters in Young's lattice of integer partitions with at most $d$ rows. The same classification holds for $GL(V)$-stable Poisson ideals of $S(V\oplusΛ^2(V))$ with its natural Poisson bracket. As applications, we determine all weak polynomial identities of the $n$-th Weyl algebra relative to its standard generating subspace and of the universal enveloping algebra of the $(2n+1)$-dimensional Heisenberg Lie algebra relative to the Lie algebra itself.

comments
21 pages, comments are welcome
fields
Representation Theory, Combinatorics, Rings and Algebras
pdf ↗ page arXiv:2610.36638 ↗

We use Gauss decomposition to study the singular vectors in the regular representation on the coordinate superalgebra of the big cell of the queer supergroup. This gives explicit bases for the highest weight spaces in queer Howe duality. As an application, the same method recovers the explicit highest weight vectors in the general linear case obtained by Cheng and Wang.

comments
27 pages, comments welcome
fields
Representation Theory
pdf ↗ page arXiv:2610.36939 ↗

Compact generation allows large triangulated categories to be studied through suitably small objects. For homotopy categories arising from exact model structures, a natural question is how this intrinsic notion of smallness relates to finite presentation in the underlying exact category. In this paper, we investigate this relation for hereditary projective model structures generated by finitely presented Gorenstein projective objects. We establish compact generation and study the stronger requirement that every compact object admits a finitely presented cofibrant representative. We call objects with such representatives perfect. Under local coherence and the existence of enough projectives among the finitely presented objects, the finitely presented cofibrant objects form a Frobenius category whose stable category describes the perfect objects and whose idempotent completion recovers all compact objects. The relative $Q$-shaped setting, for suitable small categories $Q$, makes explicit how these finiteness properties depend on the choice of exact structure. The pure exact structure yields a compactly generated homotopy category, called the $Q$-shaped pure derived category. On the other hand, the split exact structure yields the $Q$-shaped homotopy category, which is compactly generated only in the pure semisimple case, analogously to the classical homotopy category of complexes. Finally, for acyclic $Q$ and for left artinian rings, we show that every compact object admits a strictly perfect representative. In particular, we establish that compact and perfect objects coincide, proving the conjecture of Holm and Jorgensen under these hypotheses.

comments
36 pages. Comments are welcome
fields
Representation Theory, Category Theory
pdf ↗ page arXiv:2610.36966 ↗

The amplituhedron was introduced by Arkani-Hamed and Trnka (2014) to geometrically encode scattering amplitudes in high-energy physics. It can be viewed as the generalization of a cyclic polytope into the Grassmannian, and is defined to be the projection of the totally nonnegative part of one Grassmannian with respect to a totally negative element of a second Grassmannian. We construct a Lie-theoretic generalization of amplituhedra called 'flag amplituhedra', by replacing the pair of Grassmannians in the definition with an arbitrary pair of partial flag varieties for a reductive group $G$. We study in detail an example of a flag amplituhedron which is not a Grassmannian amplituhedron; it is a curved hexagon in $\mathbb{P}^1\times \mathbb{P}^1$.

comments
26 pages
fields
Representation Theory, Mathematical Physics, Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.37017 ↗

Let $U$ be a support $τ$-tilting module over a finite dimensional algebra $A$, and let $\mathcal{V}\subseteq \text{Fac}U$ be any torsion class. The heart $\mathcal H=\text{Fac} U\cap \mathcal{V}^\perp$ inherits the exact structure from mod$A$ and admits an Ext-progenerator $P$. Let $B=\text{End}_A(P)$. We define the relative derived delooping level on the exact heart $\mathcal H$ and compare it with the relative delooping level on $\mathcal {H}$ and the derived delooping level of $B$. We show that $\text{Findim} B^{op}$ is bounded by the relative derived delooping level on the $\mathcal H$. For a wide heart, the relative derived delooping level on the heart equals the derived delooping level of $B$. We obtain explicit bounds from representation-finite relative syzygy strata and characterize global first-order finiteness using an Ext-injective cogenerator.

fields
Representation Theory
pdf ↗ page arXiv:2610.37050 ↗

We extend our previous results on cluster structures in the Grothendieck rings of categories $\mathcal{O}^{\mathrm{sh}}$ for shifted quantum affine algebras to the non-simply laced types. We also show that, for all types, the formal power series expressions of the cluster monomials in terms of classes of positive prefundamental representations can be regarded as limits of cluster expansions, thus establishing their positivity.

comments
79 pages, 17 figures
fields
Representation Theory, Quantum Algebra
pdf ↗ page arXiv:2610.37184 ↗

We derive a formula for the nilpotent Kac polynomial and define an explicit generalized Kac-Moody Lie algebra (by generators and relations) and conjecture that its graded dimensions recover the constant terms of the nilpotent Kac polynomial. We outline empirical evidence supporting the conjecture and prove it for two special classes of quivers: quivers without oriented cycles except loops and quivers with "negative enough" Euler form.

comments
18 pages
fields
Representation Theory, Algebraic Geometry
pdf ↗ page arXiv:2610.37207 ↗

We present a geometric framework for realizing quantum Schur algebras with unequal parameters, via constructible sheaves on flag varieties arising from non-split groups of Lie type. In the special cases of general unitary groups and minus-type orthogonal groups, this framework yields the iSchur algebras with unequal parameters, which are closely related to the quasi-split iquantum groups of type AIII. Such a geometric realization enables the construction of canonical bases, integral forms, and monoidal categorification for these unequal-parameter iSchur algebras and the corresponding iquantum groups. This material is not a formal generalization from the equal-parameter case to the unequal-parameter case. A nontrivial arithmetic structure is imposed on the geometric structure of split groups, giving rise to many new phenomena. For example, the support varieties of generators of an integral form may not be rationally smooth, the categorification is provided by the category of descent data of semisimple complexes fixed by the twisted Frobenius, and the positivity of the canonical basis is established using a refined covering.

comments
81 pages
fields
Representation Theory, Group Theory, Quantum Algebra
pdf ↗ page arXiv:2610.37291 ↗

We study quotients of the closures of the nilpotent orbits of type A by finite groups of inner automorphisms. Assuming the nilpotent orbit is not minimal, we show that the Q-factorial terminalizations of quotients of nilpotent orbit closures are precisely the quotients of the symplectic resolutions of the orbit closures. This identifies the Mori fan of each Q-factorial terminalization of the quotient with the Mori fan of a symplectic resolution of the orbit closure.

comments
31 pages
fields
Representation Theory, Algebraic Geometry
pdf ↗ page arXiv:2610.37396 ↗

This note establishes a rank reduction for RLL-realization of quantum affine supergroups of orthosymplectic type. In the rational limit, this provides a more conceptual proof of the analogous result for orthosymplectic superYangians from arXiv:2311.18818. We also determine the exact scalar prefactor appearing in the evaluation of the universal $R$-matrix, refining the main result of arXiv:2408.16720.

comments
v1: 28pp, comments are welcome!
fields
Representation Theory, Quantum Algebra
pdf ↗ page arXiv:2610.37460 ↗

Let $\mathbf{G}$ be a connected unramified quasi-split reductive group defined over a nonarchimedean local field $F$. This paper concerns the Iwahori Whittaker functions attached to the unramified principal series of $\mathbf{G}(F)$. We introduce Demazure-Whittaker operators and show that in the non-split case they generate a Hecke algebra with unequal parameters that are determined by $\mathbf{G}$. We then connect these operators to the Whittaker functions, establishing formulas for calculating all values for a basis of Iwahori Whittaker functions and giving a new proof of the Casselman-Shalika formula for the spherical Whittaker function. In doing so we build on the work for split groups of Brubaker, Bump and Licata; the new feature here is the appearance of a Hecke algebra with unequal parameters.

fields
Representation Theory, Number Theory
pdf ↗ page arXiv:2610.37614 ↗

Statistics Theory12

Rectified flow (RF) is an ODE-based generative modeling framework that has achieved remarkable success in practice over the past few years. By design, RF seeks to produce flows that have a high degree of straightness, in the sense that the resulting ODE trajectories can be obtained with relatively few (ideally, one) sampling steps. However, a rigorous analysis of straightness or near-straightness of RF flows remains missing in the literature. In this paper, we study settings in which the coupling generated by two-step RF is provably straight at the population level. We provide a complete characterization in two popular classes of data distributions: (i) smooth, strongly log-concave densities and (ii) Gaussian mixture models. Our main results yield sharp conditions for the coupling generated by two-step RF to be straight. Non-straight couplings are constructed when these conditions fail, which demonstrate the sharpness of these conditions. Using the theory developed on the straightness of two-step RF, we propose a notion of approximate straightness that measures the straightness of the learned couplings and can be estimated from the data. We describe a procedure for constructing asymptotic confidence intervals for our proposed notion of approximate straightness and propose a data-driven method for selecting the number of reflow rounds in multiple iterations of RF.

comments
76 pages, 8 figures
fields
Statistics Theory
pdf ↗ page arXiv:2610.36090 ↗

We study score matching over finite-dimensional approximation spaces generated by gradients of eigenfunctions of the infinitesimal generator of a forward diffusion. We compare polynomial approximation under one-dimensional Ornstein-Uhlenbeck flow with trigonometric-polynomial approximation under heat flow on the torus. In the Ornstein-Uhlenbeck setting, we show that for every non-Gaussian initial distribution uniformly comparable to the standard Gaussian, the asymptotic score-estimation variance grows exponentially along a subsequence of polynomial degrees. Averaging the score-matching loss under the diffusion semigroup, which we call operator-informed score matching (OISM), reduces the variance at each fixed degree but does not eliminate this exponential instability at sufficiently small times. A local minimax lower bound shows that the same high-degree dependence arises when estimating the polynomial projection of the score, rather than being an artifact of OISM. In contrast, under heat flow on the torus, OISM over trigonometric polynomials admits finite-sample control of the score error along the forward diffusion and yields the classical minimax rate for estimating sufficiently smooth densities under expected Kullback-Leibler loss.

comments
45 pages, no figures
fields
Statistics Theory
pdf ↗ page arXiv:2610.36205 ↗

We study the mean estimation of Gaussian location model with star-shaped constraints under adversarial contamination, where at most $ k<N/2 $ among $ N $ observations are corrupted by an oracle adversary. We characterize the minimax squared $ \ell_2 $ risk throughout the full range of contamination level $ \varepsilon = k/N \in [0,1/2) $. The minimax risk is \begin{equation*} \mathfrak{M}\asymp \big( δ_\varepsilon ^{*2} \vee σ^2\varepsilon ^2\log \frac{e}{1-2\varepsilon } \big) \wedge d^2 , \end{equation*} where $ δ_\varepsilon ^* := \sup \big\{ r \geq 0: (1-2\varepsilon )\frac{Nr ^2}{σ^2} \leq \log M^\mathrm{ loc }_K(r ,c) \big\} $, $ \log M^\mathrm{ loc }_K(r ,c) $ is the local entropy of constraint set $ K $, $ d $ is the diameter of $ K $ (set $ d=\infty $ if $ K $ is unbounded), $ σ^2 I_n $ is the covariance of the Gaussian noise, and $ c $ is a sufficiently large constant. Our result sheds light on the interaction between contamination error and information theoretic complexity in the regime of proximity to the breakdown point $ \varepsilon \uparrow 1/2 $.

comments
33 pages
fields
Statistics Theory
pdf ↗ page arXiv:2610.36283 ↗

This paper studies randomised quasi-Monte Carlo (RQMC) methods for Metropolis–Hastings (MH) algorithms, whose accept–reject discontinuity complicates RQMC analysis. To circumvent some of these complications, we study Array-RQMC for MH, with $N$ replicated chains evolved over a fixed time horizon $T$, where we make use of Hilbert sorting for joint multidimensional updates at each step. For one-dimensional MH and bounded-variation test functions, we prove a sampling mean squared error bound of $\mathcal O_T(N^{-4/3})$ relative to the finite-horizon expectation. For product targets, coordinatewise Array-RQMC Metropolis-within-Gibbs retains this exponent for additive marginal bounded-variation tests, with constants that may depend on dimension. With Gaussian random-walk proposals, the bounds extend to scalar and additive marginal polynomials with an additional factor $(\log N)^\ell$, where $\ell$ is the maximum polynomial degree. For joint updates in state dimension $p$ using $p$ proposal uniforms, we provide a conditional $\mathcal O_T(N^{-1-1/(p+2)})$ framework under regularity assumptions that are not verified for the joint multidimensional examples. Numerical experiments support the verified one-dimensional theory and demonstrate variance reduction for joint Hilbert-sorted updates, including Bayesian logistic regression, while illustrating the limitations of these gains as dimension increases.

fields
Statistics Theory
pdf ↗ page arXiv:2610.36431 ↗

We develop a large-$n$, large-$T$ theory for the unsmoothed fixed-effects quantile regression estimator, paralleling Hahn and Newey (2004) for smooth nonlinear panel models. An exact rank-score identity for the profiled criterion, with a convexity argument, delivers the leading incidental parameter bias of the slope estimator, of order $T^{-1}$, in closed form. When $n$ and $T$ grow at the same rate, the estimator is asymptotically normal but in general not centred, and conventional intervals are invalid. Admissible growth of $n$ relative to $T$, up to $n=o(T^3)$, depends on the regularity of the conditional error densities. A split-panel jackknife removes the leading bias throughout this range and yields valid inference. In serially dependent panels whose dependence decays geometrically or polynomially, the bias acquires a term from the dependence between regressors and errors of different periods, the jackknife remains valid with contiguous but not with interleaved half-panels, and clustered standard errors are valid.

fields
Statistics Theory
pdf ↗ page arXiv:2610.36443 ↗

The Sinkhorn divergence, a debiased form of entropic optimal transport, has been widely used in various scientific disciplines due to its computational and statistical tractability. Unlike the Wasserstein distance, however, the Sinkhorn divergence is not a metric and it does not naturally admit a formal Riemannian structure. These properties are central to many applications of optimal transport methods that utilize interpolation, barycentric averaging, parallel transport or other geometric operations. Recent work addresses these shortcomings by constructing a formal Riemannian tangent space and a metric tensor on the space of probability measures that quadratically approximates the Sinkhorn divergence. In this work, we extend this framework with two major threads of contributions. For a suitable class of finite-smoothness cost functions, we prove Hadamard differentiability of the Sinkhorn metric tensor and we derive the Levi-Civita covariant derivative in this geometry; we then use the covariant derivative to obtain the geodesic and parallel transport equations, each of which amount to an ODE on a fixed Reproducing Kernel Hilbert Space (RKHS) with smooth coefficients. Our second contribution is to linearize the solutions to the geodesic and parallel transport equations using standard tools from Banach space ODE theory. We show that under standard regularity conditions these linearizations imply $\sqrt{n}$-limiting Gaussian processes centered at population values for the plugin estimators of the exponential map, the logarithmic map, parallel transport and the induced Sinkhorn distance. Moreover, we show that the nonparametric bootstrap is consistent for all objects mentioned. To the best of our knowledge, these are the first $\sqrt{n}$--limit theorems and bootstrap consistency results for canonical geometric operations in a transport-based geometry on the space of probability measures.

fields
Statistics Theory
pdf ↗ page arXiv:2610.36488 ↗

Dimension-reduction methods such as PCA identify a subspace capturing most of the variation in the data, but not a distinguished basis within it. Selecting a basis in which the samples are sparsely represented is a recurring task in unsupervised learning, addressed by methods in factor analysis, dictionary learning, and subspace clustering. We pose this task as Wasserstein approximation of the data distribution by measures supported on the axes of an orthogonal frame, which, in turn, reduces to a nonsmooth optimization problem over the orthogonal group $\mathrm{O}(d)$. We show that, under mild assumptions, both empirical risk minimizers and a locally initialized stochastic gradient method recover the population-optimal frame at the near-parametric rate $\widetilde{O}_d(n^{-1/2})$, where $n$ is the sample size. The main difficulty is that, although the population objective is smooth, the empirical objective is not-this mismatch precludes standard covering arguments. We overcome this issue by replacing Lipschitz continuity of the empirical gradient with a coarsely Lipschitz bound, a tool that may be of independent interest. Numerical experiments support our theoretical findings.

comments
66 pages, 5 figures
fields
Statistics Theory, Optimization and Control
pdf ↗ page arXiv:2610.36581 ↗

Randomized unbiased estimation can be highly efficient for approximating expectations of functionals of stochastic differential equations. However, in practice, the optimal computational efficiency is rarely attained due to errors arising from parameter estimation and truncation. The effects of these errors on computational efficiency have not been investigated. In this article, we study the error analysis and optimal sample allocation for the single term estimator. Specifically, under mild assumptions on the estimated parameters, we develop a unified error analysis that incorporates both sources of error, and establish the relevant convergence rates for the computational efficiency loss. Based on this analysis, we further investigate the optimal allocation of the computational budget for parameter estimation, and propose several efficient sample allocation strategies. Numerical experiments support the theoretical convergence rates and demonstrate the effectiveness of our proposed methods.

fields
Statistics Theory, Optimization and Control, Probability
pdf ↗ page arXiv:2610.36699 ↗

Recent work established minimax rates for pointwise estimation of the conditional average treatment effect (CATE), but the fundamental limits under global losses remained unresolved. We derive matching minimax lower and upper bounds under the $L_2$ and $L_\infty$ risks on certain covariate-density and support classes. For the lower bounds, we develop a many-fuzzy-hypotheses method that replaces each fixed alternative in a multiple-hypothesis construction with a mixture over nuisance functions that leaves its CATE unchanged. A localized product-mixture $χ^2$ bound is used to control the divergence of each resulting mixture from a common baseline, enabling multiple well-separated CATE hypotheses to be handled simultaneously. For the upper bounds, we analyze a modified local polynomial R-learner. The $L_2$ argument extends the pointwise bias–variance analysis, while the $L_\infty$ argument combines concentration inequalities for U-statistics with a continuity argument enabled by the use of a continuous kernel and additional regularity of the second-order basis. The resulting rates show that the pointwise elbow phenomenon persists globally: the $L_2$ transition threshold is unchanged, while the $L_\infty$ rate has an additional logarithmic factor and a transition threshold adjusted by $1-\log\log n/\log n$.

fields
Statistics Theory
pdf ↗ page arXiv:2610.37496 ↗

For each time $t$, the regression function $m_t(x)=E[Y_t\mid X_t=x]$ of an outcome on contemporaneous covariates can be regarded as a field on covariate space. We study how this field evolves when the covariates of individual units move as a diffusion and the outcome attached to each unit changes according to its own dynamics. Under mild regularity we show that $m_t$ satisfies a second-order evolution equation whose diffusion term is $\frac12\mathrm{tr}(Σ\nabla^2 m_t)$, the trace of the Hessian of the regression function weighted by the instantaneous covariance $Σ$ of the covariate increments, whose transport term is the drift of the time-reversed covariate process, and whose source term is the conditional mean rate of change of the outcome. The equation is a regression-theoretic reading of a classical transport law, and it yields a Feynman-Kac representation of $m_t$ along reversed covariate paths. We connect the Hessian appearing in the equation to the Hessian estimated by quadratic regression (under Gaussian design the best quadratic approximation has Hessian $E[\nabla^2 m(X)]$ exactly) and to the leading bias of kernel smoothers. For binary outcomes governed by bistable replicator dynamics with imitation among units with similar covariates, the closed equation is of Nagumo type, its planar travelling fronts are exactly logistic in $(x,t)$, and a single logistic regression of the outcome on covariates and time identifies the ratio of diffusion to reaction and the direction of invasion. We discuss identification under panel versus repeated cross-sectional designs and a two-stage estimation strategy, and we illustrate the results with simulations.

comments
13 pages, 2 figures. Code for the symbolic checks and simulations is included as ancillary files
fields
Statistics Theory
pdf ↗ page arXiv:2610.37500 ↗

This paper considers model-based clustering of diffusion processes using high-frequency data. We formulate a finite mixture model for diffusion processes and develop a clustering method based on their constant diffusion coefficients. Using a quasi-likelihood, we propose estimators of the number of clusters, the diffusion coefficients and the cluster labels, and establish their consistency and the asymptotic normality of the diffusion coefficient estimators under high-frequency asymptotics with the number of observed processes fixed. Simulation studies are conducted to investigate the finite-sample performance of the proposed method.

fields
Statistics Theory
pdf ↗ page arXiv:2610.37514 ↗

We determine the intrinsic Fisher completion of the full-covariance wrapped Gaussian family on a fixed torus and relate its geometry to uniform estimation. The completion is homeomorphic to the total-variation closure. Every added law is, in suitable integer torus coordinates, a product of a uniform law and a lower-dimensional wrapped Gaussian. All these densities are positive and real analytic, but in dimension at least two no added point has a compact neighborhood. Small balls remain contractible, and explicit minimizing geodesics pass through added points. In dimension at least two, every fixed Fisher neighborhood of an added point has a positive minimax risk at every sample size. For subfamilies with a common positive lower bound on the variance in every nonzero integer direction, we prove that compactness of the completed closure is equivalent to uniform consistency in either $L^1$ or completed-Fisher loss. Equivalently, the Fourier moments must decay uniformly over the subfamily. A fixed covariance floor guarantees this condition without an upper covariance bound and gives $L^1$ risk at most $C_{\varepsilon,d}n^{-1/2}\{\log(n+1)\}^{d/4}$. A common-phase example illustrates how partially uniform limits can retain relative-phase information while remaining uniformly estimable.

comments
33 pages, 5 figures
fields
Statistics Theory, Differential Geometry, Metric Geometry
pdf ↗ page arXiv:2610.37526 ↗

Group Theory9

We classify the minimal faithful Alt(6)-modules with an additive dimension in characteristic not 3. We also highlight the remaining work needed to complete the classification of minimal faithful Alt(n)- and Sym(n)-modules with an additive dimension

fields
Group Theory
pdf ↗ page arXiv:2610.35941 ↗

For $n\ge2$, let $F_n$ be the free group of rank $n$, and let $\IA(F_n)$ be the group of automorphisms of $F_n$ acting trivially on its abelianization. Two natural filtrations of $\IA(F_n)$ are its lower central series $γ_c(\IA(F_n))$ and the Andreadakis filtration $\mathrm{I}_{c+1}\mathrm{A}(F_n)$, consisting of automorphisms acting trivially on $F_n/γ_{c+1}(F_n)$, $c \geq 1$. Andreadakis conjectured that these filtrations coincide. We disprove this conjecture for every $n\ge3$ and $c\ge4$.

fields
Group Theory
pdf ↗ page arXiv:2610.36302 ↗

We study the subalgebra generated by degree-one mod-$2$ cohomology for metacyclic $2$-groups $G=\langle a,b\mid a^{2^α}=1,\ b^{2^m}=a^c,\ b^{-1}ab=a^r\rangle$ with unique normal forms, where $α,m\ge2$ and $r\equiv-1\pmod4$. Presentation compatibility forces $c=0$ or $2^{α-1}$ and $G^{\mathrm{ab}}\cong C_2\times C_{2^m}$. If $c\equiv0\pmod4$, the degree-one-generated algebra is $\mathbb{F}_2 [x,y]/(y^2,x^2+xy)$, a four-dimensional algebra shared by the four nonmodular types of Aaboun–Zekhnini. The unique exceptional case $α=2$, $c=2$ is modular and has infinite-dimensional degree-one algebra $\mathbb{F}_2[x,y]/(y^2,x^2y)$. The regular algebra is PBW, Koszul, universally Koszul, and strongly Koszul, whereas the modular algebra is not Koszul. We realize the regular algebra by inflation from the common quotient $C_4\rtimes_{-1}C_4$ and identify its quadratic dual with the characteristic-two Jordan plane. For the inversion tower $C_4\rtimes_{-1}C_{2^m}$, we give uniform presentations of the full mod-$2$ cohomology rings and determine the kernels and images of inflation in every degree. Their direct limit yields the continuous cohomology ring of $C_4\rtimes_{-1}\mathbb Z_2$. We also record an arithmetic consequence for Galois groups arising over cyclotomic $\mathbb Z_2$-extensions of real quadratic fields.

fields
Group Theory, Rings and Algebras
pdf ↗ page arXiv:2610.36401 ↗

Ghosh, Meiri and Sarnak showed that the commutator equation over $\mathrm{SL}_2(\mathbb{Z})$ satisfies a profinite local to global principle, and that for a prime $\ell\equiv\pm1$ mod $5$ it fails for $\mathrm{SL}_2(\mathbb{Z}[1/\ell])$ with a reciprocity obstruction. We extend the failure to surface groups. For the genus-$2$ surface group $Γ_2$ we construct an explicit representation $ρ:Γ_2\to\mathrm{SL}_2(\mathcal{O}_{K,S})$, where $K=\mathbb{Q}(\sqrt{5})$ and $\mathcal{O}_{K,S}=\mathbb{Z}[\frac{1+\sqrt{5}}{2},\frac{1}{10}]$, whose quotients modulo the maximal ideals are the full groups $\mathrm{SL}_2(\mathbb{F}_q)$ except in residue characteristic $3$, where the quotient has order $24$. For every $k\ge2$ the word $[a_1,b_1]^k$ is not a commutator in $Γ_2$, while its image is a commutator in every one of these quotients. The same holds for every surface group $Γ_g$ with $g\ge2$.

comments
11 pages
fields
Group Theory, Number Theory
pdf ↗ page arXiv:2610.36497 ↗

Let $π$ be a set of primes. We extend the theory of left and right $π$-nilpotency in skew braces beyond the nilpotent type setting, and introduce the study of central $π$-nilpotency for finite skew braces. We investigate the relationships between these three notions, and a characterisation of central $π$-nilpotency in terms of chief factors is also established. This leads to the introduction of the $π$-Fitting ideal of a skew brace, for which we prove a local analogue of Fitting's theorem and provide a description in terms of centralisers of chief factors. Finally, we introduce the $π$-core of a skew brace and investigate its relationship with $π$-ideals and central $π$-nilpotency.

comments
19 pages
fields
Group Theory
pdf ↗ page arXiv:2610.37042 ↗

We demonstrate that the conjugacy growth series for the Heisenberg groups $H_n(\mathbb{Z})$ is never algebraic, irrespective of the generating set. This makes progress towards the conjecture that all finitely presented groups that are not virtually abelian have irrational conjugacy growth series. Contrary to the group $H_3(\mathbb{Z})$, where the conjugacy growth is $n^2\log(n)$, the conjugacy growth of the other Heisenberg groups is polynomial, making these groups the first known nilpotent groups where the irrationality does not follow from the asymptotic growth.

comments
17 pages, 4 figures
fields
Group Theory
pdf ↗ page arXiv:2610.37334 ↗

We study two obstructions to a finite group $G$ having a generating $d$-tuple whose Cayley graph has spectral gap at least $ε$. Lubotzky and Weiss proved that such a gap forces $\mathrm{ab}_k(G)$, the largest order of the abelianization of a subgroup of index at most $k$, to be $\exp(O_ε(dk))$, while de la Harpe, Robertson and Valette proved that it forces $\mathrm{rep}_k(G)$, the number of irreducible complex representations of degree at most $k$, to be $\exp(O_ε(dk^2))$. We show that the first bound implies the second, and in fact implies it with $k^2$ replaced by $k\log k$. For this, we prove that every finite $d$-generated group satisfies the inequality $\mathrm{rep}_k(G)\leq\exp((d-1)k\log k+O(dk))\cdot\mathrm{ab}_k(G)$, regardless of its spectral gap. The leading term $(d-1)k\log k$ is attained by powers of alternating groups, and it disappears for groups without large alternating sections. The proof reduces to counting the finite primitive subgroups of $\mathrm{GL}_m(\mathbf{C})$, and we show that there are at most $\exp(O(\log^4m))$ of them up to conjugacy and scalars. Wigderson asked whether $k^2$ in the second bound can be lowered to $k$, so our result answers his question up to a logarithm. We do not know whether the logarithm is actually needed. For powers of alternating groups, the natural candidates, we show that a spectral gap keeps the number of factors a fixed power below the maximum allowed by $d$-generation, and that random generators have spectral gap tending to $0$ already for superpolynomially many factors.

fields
Group Theory, Combinatorics, Representation Theory
pdf ↗ page arXiv:2610.37399 ↗

Throughout this paper, all groups are finite. Let $\mathrm{form}(G)$ denote the formation generated by a group $G$. A formation $\mathfrak{F}$ is said to be $S_n$-closed if every subnormal subgroup of every group in $\mathfrak{F}$ belongs to $\mathfrak{F}$. A. N. Skiba and L. A. Shemetkov posed the problem of proving that $\mathrm{form}(G)$ has only finitely many $S_n$-closed subformations for every group $G$ (Problem 14.81 of the Kourovka Notebook). In this paper we propose a negative answer to this problem. We prove that for the group $G_0=(\mathrm{SL}_2(5)\rtimes C_2)\times\mathrm{SL}_2(5)$ of order 28800, where $C_2$ acts on $\mathrm{SL}_2(5)$ by an outer automorphism, the formation $\mathrm{form}(G_0)$ has an infinite strictly descending chain of $S_n$-closed subformations. The main tool is a family of central-valued multilinear maps, defined for every group in $\mathrm{form}(G_0)$ by its outer actions on its components; their vanishing is preserved under taking epimorphic images, subdirect products and subnormal subgroups. The members of the chain are separated by central quotients of the groups $\mathrm{SL}_2(5)^{2^d-1}\rtimes C_2^d$, $d\geq 2$. Using Burichenko's description of the lattice of subformations of $\mathrm{form}(\mathrm{SL}_2(5)\rtimes C_2)$, we also prove that every non-empty $S_n$-closed subformation of $\mathrm{form}(G_0)$ is generated, as an ordinary formation, by a single group, and that the set of such subformations is countably infinite. Finally, we determine the intersection of the chain and give an explicit generator of it.

comments
18 pages
fields
Group Theory
pdf ↗ page arXiv:2610.37462 ↗

We study systems of moves and states in the sense of [JNW21] on certain bipartite graphs, and the associated homomorphisms from the commutator subgroups of right-angled Coxeter groups to $\mathbb Z$. We give examples of right-angled Coxeter groups whose graphs do not satisfy the legal system criterion, but contain finite index right-angled Coxeter subgroups whose graphs admit legal systems. We also provide a family of examples of graphs without legal systems, where no sequence of the considered graph operations yields a graph with a legal system. Moreover, we show that for a bipartite graph and an orbit of non-empty states under the bipartite system of moves, the associated homomorphism from the commutator subgroup of the right-angled Coxeter group to $\mathbb Z$ has finitely generated kernel if and only if the system is legal.

comments
19 pages, 9 figures. Comments are welcome
fields
Group Theory
pdf ↗ page arXiv:2610.37641 ↗

Metric Geometry9

We establish a sharp reverse isoperimetric inequality for convex subsets of a triangle under a prescribed area constraint. Given a triangle $T$ and an area fraction $η\in(0,1)$, we determine the set with maximal perimeter among convex bodies $K\subset T$ satisfying $|K|=η|T|$ and characterize the equality cases. Every maximizer is a triangle, and the optimal configuration is obtained by retaining a longest side of $T$ and placing the third vertex on a shortest side, at the position determined by the area constraint. This characterization yields an explicit expression for the maximal perimeter in terms of $η$ and the side lengths of $T$.

fields
Metric Geometry
pdf ↗ page arXiv:2610.36220 ↗

The main result of this paper uses an approximate inner ball condition to characterize $C^{1,α}$ convex bodies for $0<α\leq 1$. In particular, we show that a convex body $K$ is of class $C^{1,α}$ if and only if there exist constants $C,\varepsilon_0>0$ such that for every $0<\varepsilon\leq\varepsilon_0$ and for every $x\in\partial K$ there exists $h^\varepsilon(x)$ satisfying $B(h^\varepsilon(x), C\varepsilon^\frac{1-α}{1+α})\subset K$ and $\operatorname{dist}(x,B(h^\varepsilon(x), C\varepsilon^\frac{1-α}{1+α}))\leq\varepsilon.$ We then give a new proof of the regularity of the convex envelope by showing that $\operatorname{conv}(f)\in C^{1,α}_{\rm loc}(\mathbb{R}^n)$ whenever $f\in C^{1,α}_{\rm loc}(\mathbb{R}^n)$ and $f$ satisfies the coercivity condition $f(x)\to\infty$ as $|x|\to\infty$. Finally, we use the approximate inner ball characterization of $C^{1,α}$ convex bodies to prove that the Minkowski sum of a $C^{1,α}$ convex body and a nonempty closed convex set is a $C^{1,α}$ convex domain.

fields
Metric Geometry, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.36238 ↗

In 2000, Birbrair and Fernandes proved that the Hölder Semicomplex of the branches of a germ of a curve with its contact orders is a complete invariant of outer bi-Lipschitz equivalence of definable sets in polynomially bounded o-minimal structures. They also proved a realization theorem for it, i.e., for each Hölder semicomplex $G$, there is a germ of a semialgebraic curve $(X,x_0)$ such that the Hölder Semicomplex of $(X,x_0)$ coincides with $G$. In 2000, Birbrair proposed the problem of finding and characterizing all the Hölder semicomplexes that can be realized as a Hölder Semicomplex of a real analytic curve. Solving this problem in its full generality is a notoriously hard task; however, in this paper, we provide a complete answer to this problem when we restrict it to tangential nodal curves: curves defined as the union of the zero set of a harmonic function and its tangent cone at a fixed point.

fields
Metric Geometry, Complex Variables, Number Theory
pdf ↗ page arXiv:2610.36381 ↗

A packing of balls in Euclidean space is a set of balls with non-overlapping interiors. For a finite packing of balls, its average contact number is the average number of balls tangent to a ball in this packing, that is, twice the total number of tangencies divided by the number of balls. The average kissing number of the space is defined as the supremum of average contact numbers over all finite ball packings. In this paper, we present a new semidefinite programming approach to average kissing numbers and improve upon known upper bounds in all dimensions \(3\) through \(30\). In particular, we provide the first improvements to the upper bounds in dimensions \(10\) through \(30\) obtained from kissing numbers. We also fully solve the average kissing number problem in dimensions \(8\) and \(24\) by showing that the average kissing numbers coincide with the kissing numbers, 240 and 196560, respectively. These values are limits of average contact numbers of finite subpackings from the \(E_8\) and Leech lattices.

fields
Metric Geometry, Combinatorics
pdf ↗ page arXiv:2610.36675 ↗

The article is devoted to the construction of dimension-raising maps of cubes with the highest possible Holder exponent. In particular, a square-to-cube mapping with a Holder exponent of $2/3$ is constructed, which is a solution to the V.I.Arnold problem.

comments
36 pages, 5 figures
fields
Metric Geometry
pdf ↗ page arXiv:2610.36766 ↗

For $n\geq 2$, we study the optimal function $\varphi_n$ in the Brunn-Minkowski inequality for $θ$-convolution bodies $|K+_θL|^{1/n}\geq\varphi_n(θ)(|K|^{1/n}+|L|^{1/n})$ for any convex bodies $K,L\subset\mathbb{R}^n$. We improve the known lower bound for every $0<θ<(3/4)^n$ and determine $\varphi_n$ on an interval strictly containing $[(2/3)^n,1]$. For $θ<1$ in this interval, equality holds if and only if $K$ and $-L$ are translates of the same $n$-dimensional simplex. The proof reduces the geometric problem to a sharp one-dimensional estimate for nonnegative concave functions under a weighted integral constraint.

fields
Metric Geometry
pdf ↗ page arXiv:2610.37066 ↗

We study properties of compact Besicovitch sets and other Kakeya-type sets which are typical in the sense that the exceptions form a meager set in the Baire category sense in a suitable metric space. The main motivation comes from the known fact that typical compact Besicovitch sets in $\mathbb{R}^n$ have minimal Hausdorff dimension among all Besicovitch sets of $\mathbb{R}^n$. Thus, in order to estimate the smallest possible Hausdorff dimension of an arbitrary Besicovitch set, it is enough to estimate the Hausdorff dimension of a typical (compact) Besicovitch set, or analogously of a Besicovitch set satisfying our list of typical properties. Among other results, we prove that a typical Besicovitch set $K$ in $\mathbb{R}^n$ has the following properties: it is not sticky, any line intersects $K$ in a set of ($1$-dimensional) measure zero or in a unit line segment and a set of measure zero, it has no more than $n$ different unit line segments in any direction, and for each $1\le j\le n$ the Hausdorff dimension of the set of those directions in which $K$ has exactly $j$ unit line segments is $n-j$. For compact metric spaces $X$ and $Y$, let $\mathcal{C}(X,Y)$ denote the set of compact subsets of $X\times Y$ whose projection onto the first coordinate is $X$, equipped with the Hausdorff metric. Kakeya-type compact sets can be coded by elements of such spaces, allowing typical properties to be studied through comeager subsets of $\mathcal C(X,Y)$. Among other results, we prove that, under suitable conditions on $X$ and $Y$, a typical $A\in \mathcal{C}(X,Y)$ is totally disconnected and purely unrectifiable, its projection to the second coordinate has Hausdorff dimension zero, and the dimension identities $\dim_H(A)=\dim_H(X)$ and $\dim_P(A)=\dim_P(X)+\dim Y$ hold.

fields
Metric Geometry
pdf ↗ page arXiv:2610.37419 ↗

Let $\mathcal{N}(k,n)$ be the largest size of a family of affine $k$-planes in $\R^n$ at pairwise separation distance one. We develop a unified treatment of upper bounds, dimension reduction, and product constructions. Orthogonal-projection coordinates give a rank-stratified distance identity; a Veronese lifting converts its polynomial consequences into simultaneous total isotropy. Connected Nash decomposition and an analytic diagonal argument replace the linear-independence step in Höfer's three-dimensional method. An explicit cylindrical decomposition estimate yields $\mathcal{N}(k,n)\leqslant 2^{\,2^{\,3(k+1)(n-k)}}$. For configurations whose direction sums have dimension at most $k+1$, we prove the exact reduction $\mathcal{N}_{\leqslant k+1}(k,n)=\mathcal{N}(1,n-k+1)$. Every codimension-two equidistant configuration splits off a common linear factor, giving $\mathcal{N}(n-2,n)=\mathcal{N}(1,3)$. We also determine the sharp bound $2(n-k)$ when all directions lie in a fixed $(k+1)$-dimensional subspace. A transverse product construction gives multiplicative inequalities and makes $S(c)=\mathcal{N}(c-1,2c-1)$ supermultiplicative. In particular, $S(c)\geqslant 2^{c\bmod 2}7^{\lfloor c/2\rfloor}$. The structural results and constructions do not depend on the exact value of $\mathcal{N}(1,3)$.

comments
20 pages, comments are welcome
fields
Metric Geometry, Combinatorics
pdf ↗ page arXiv:2610.37421 ↗

Since its introduction in 2003, Wilking's Connectedness Lemma has become a powerful tool in the study of manifolds with positive sectional curvature. In recent work of the authors, this has been extended to orbifolds. The purpose of this note is to exhibit counterexamples in the class of Alexandrov spaces for the Connectedness Lemma in both its ordinary and dual formulation. This answers a 2007 question of Wilking.

comments
6 pages
fields
Metric Geometry, Differential Geometry
pdf ↗ page arXiv:2610.37476 ↗

Geometric Topology8

Relying merely on B.I. (a.k.a. Bowditch's intelligence), we show that whenever a group $G$ admits a convergence action on a compact perfect metrizable space $M$, there is a hyperbolic space $X$ along with a proper isometric action of $G$ on $X$ inducing a $G$-equivariant homeomorphism $f:M \rightarrow \partial X.$ This answers a question by Misha Kapovich.

fields
Geometric Topology, Group Theory
pdf ↗ page arXiv:2610.36211 ↗

We prove that every closed, orientable $3$--manifold contains hyperbolic knots with arbitrarily large essential geodesic systole. That is, for any $3$--manifold $M$, and $R> 0$ there is a hyperbolic knot $K \subset M$ such that every closed geodesic in the knot complement which is non-null-homotopic in $M$ has length at least $R$. The construction also produces filling knots, in the sense of Freedman-Krushkal [FreeKrus2023]. This answers two questions of Biringer.

comments
28 pages, 8 figures, comments welcome. No AI use
fields
Geometric Topology, Group Theory
pdf ↗ page arXiv:2610.36259 ↗

We show that there is no algorithm to decide whether or not a closed, connected, orientable, locally flat PL surface in $S^4$ of genus at least 18 (resp. 19) is TOP (resp. PL) unknotted. We prove analogous results for non-orientable surfaces, and for surfaces in any compact PL 4-manifold. We also show that there is an orientable surface in $S^4$ of genus 18 whose group has unsolvable word problem, and an orientable surface in $S^4$ of genus 40 whose group contains an isomorphic copy of every recursively presented group. Similar results hold for non-orientable surfaces.

comments
17 pages
fields
Geometric Topology
pdf ↗ page arXiv:2610.36331 ↗

We construct homological lifts of Arnold's strangeness $\operatorname{St}$ for long and closed plane curves. For long curves, we introduce a new $q$-variable quantization polynomial $\operatorname{St}_q$, an edge state sum, designed for categorification and construct a triply graded homology that recovers $\operatorname{St}_q$. We prove strictness by exhibiting long curves with the same number of crossings and the same $\operatorname{St}_q$ (in particular, the same rotation number and $\operatorname{St}$) that are distinguished by the homology. For closed curves, we introduce a corresponding new polynomial $\operatorname{St}_q$ and construct an edge homology. In this case $\operatorname{St}_q$ is recovered from a signed $q$-character $\operatorname{ch}_q$ defined on the canonical cycles of the homology rather than directly from a graded rank. Again, we exhibit closed curves with the same number of crossings and the same $\operatorname{St}_q$ (in particular, the same rotation number and $\operatorname{St}$) that are distinguished by the homology.

comments
21 pages, 6 figures
fields
Geometric Topology
pdf ↗ page arXiv:2610.36387 ↗

We study the $d$-invariant from Heegaard Floer homology for Brieskorn spheres $Σ(p,q,r)$. Using surgeries on two singular fibers, we establish new relations of the $d$-invariants among various Brieskorn spheres. For $p=2$, we show that $d$ never vanishes if $q/r$ lies in certain intervals on the rational line. Altogether, we completely classify Brieskorn spheres $Σ(2,q,r)$ with $d=0$. The zero set is expressed using submonoids of $\mathrm{GL}(2,\mathbb{Z})$ and the Stern–Brocot tree. As an application, we prove that most Brieskorn spheres $Σ(2,q,r)$ do not bound a rational homology $4$-ball.

comments
26 pages, 8 figures, comments welcome
fields
Geometric Topology, Algebraic Geometry, Number Theory
pdf ↗ page arXiv:2610.36472 ↗

We show that if a three-manifold $M$ admits a veering triangulation, then the dual spine of every essential triangulation of $M$ can be smoothed to a branched surface. We use this criterion to identify a large collection of cusped hyperbolic three-manifolds that do not admit veering triangulations. We then use the same criterion to construct the first infinite family of cusped hyperbolic three-manifolds without veering triangulations.

comments
33 pages, 19 figures. Comments welcome
fields
Geometric Topology
pdf ↗ page arXiv:2610.37057 ↗

We investigate improved bounds on the kissing number of a hyperbolic surface under the assumption that the systole set contains pants decompositions with volume growth constraints. Namely, when the dual graph of the pants decomposition has polynomial volume growth, we obtain a power-savings improvement for the upper bound on the kissing number in terms of systolic length. We also construct super-exponentially many (in $g$) pants decompositions with exponential volume growth for which our techniques still imply stronger upper bounds for kissing numbers. The proofs use the technology of ``stem systems'' and ``tulips'' developed by the first and second author to study $1$-systems of curves.

comments
13 pages, 3 figures
fields
Geometric Topology
pdf ↗ page arXiv:2610.37551 ↗

Inspired by work in Seiberg–Witten and Heegaard Floer homology, Dunfield, Schütz, and the author constructed an abelian "local equivalence group" built from pairs of chain complexes, and a homomorphism from the smooth concordance group to this group that uses the odd Khovanov complex and the Bar-Natan deformation of the even Khovanov complex. Lewark independently gave a related local equivalence group and concordance invariant, using only the Bar-Natan deformation. In this paper, we construct $(\mathbb{Z}/2\mathbb{Z})^\infty$ subgroups of Lewark's local equivalence group and the kernel of the map from our local equivalence group to Lewark's, as well as $\mathbb{Z}^\infty$ direct summands of Lewark's group and this kernel, invisible to the $s$-invariants. The construction gives at least six $\mathbb{Z}$-valued concordance homomorphisms linearly independent from $s$, $τ$, and Dai–Hom–Stoffregen–Truong's invariants.

comments
40 pages. Research largely carried out by Anthropic's Claude models; paper written by the author. See methodology section and AI disclosure
fields
Geometric Topology
pdf ↗ page arXiv:2610.37595 ↗

General Mathematics7

We prove that there are exactly $2^\mathfrak{c}$ simultaneously bireflective and bicoreflective subconstructs of $\mathbf{FTS}$. By the Lowen–Wuyts bijection, these correspond to the total fuzzy topologies contained in $M$, the fuzzy topology consisting of all nondecreasing left-continuous self-maps of the unit interval. The upper bound follows from a cardinality estimate. The matching lower bound is obtained by encoding $2^{\mathfrak c}$ ultrafilters on $\mathbb N$ into total fuzzy topologies contained in $M$.

fields
General Mathematics
pdf ↗ page arXiv:2610.35876 ↗

Let $F\colon A\to\mathcal{P}(X)$ be a soft linear subspace of a linear space $X$ over $\mathbb{K}$. Its soft elements form the product $\prod_{t\in A}F(t)$. We use this representation to characterize selection spaces, decompose soft norms and soft linear operators into their sectionwise parts, and compare soft, uniform, and componentwise convergence. We also characterize completeness and continuity in the box, extended-supremum, and product settings. Neumann-series and fixed-point results are obtained as applications.

comments
13 pages, no figures
fields
General Mathematics
pdf ↗ page arXiv:2610.35879 ↗

In this paper, we present a Radon-Nikodym theorem for dominated fuzzy number multimeasures defined on $Σ$ and taking values in $\mathcal{F}_{cwk}(X)$, where $(Ω, Σ, μ)$ is a complete finite measure space and $\mathcal{F}_{cwk}(X)$ is the family of generalized fuzzy numbers on a Hausdorff locally convex topological vector space $X$, i.e., a map $u: X \to [0,1]$ is a member of $\mathcal{F}_{cwk}(X)$, if and only if for each $α\in (0,1]$ the $α$-cut $[u]^α$ of $u$ is a member of the family $cwk(X)$ of all nonempty convex and weakly compact subsets of $X$, ($[u]^α = \{x \in X: u(x) \geq α\}$).

comments
7 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.35883 ↗

Let $(X,τ)$ be a noncompact metrizable space and let $\langle T_n\rangle$ be a sequence of nonempty closed sets converging to the empty set in the sense of Kuratowski–Painlevé. The subsets of $X$ whose closures meet at most finitely many terms form a bornology $\mathcal B_τ(\langle T_n\rangle)$. We give an internal criterion for inclusion between two such bornologies and recover each of them from a canonical free filter of closed sets. In fact, a nontrivial bornology is sequence-induced exactly when it has the enlargement property and its closed polar has countable character. Our principal result says that the bornology of relatively compact sets is sequence-induced precisely when $X$ is locally compact and separable; in the locally compact case the character of its closed polar is the density of $X$. In the penultimate section, we make a clever use of the sequence $T_n=\{n\}$ in the real line to give a negative answer to a conjecture posed in 2001 about a coarsest topology making the associated bornology a nontrivial metric bornology. Finally, we end the paper with a more general result: a coarsest such supertopology exists if and only if the original topology already makes the induced bornology a nontrivial metric bornology, in which case the coarsest supertopology is the original topology itself.

fields
General Mathematics
pdf ↗ page arXiv:2610.35907 ↗

Every neighborhood assignment on a charming space $X$ has a countable kernel whose closure in $X$ is Lindelöf $Σ$. Consequently, for any two neighborhood assignments $M$ and $N$ on $X$, there is a closed discrete set $D\subseteq X$ such that $M[N[D]]=X$. We give an exact one-step formulation of the remaining $D$-space problem and identify the universal pointwise square-refinement condition with the classical $z$-space condition. For continuous closed surjections, we prove a fixed-assignment lifting equivalence. We also show that every uncountable $\mathrm{HFC}_{\mathrm w}$ subspace of $[ω_1]^{\leqω}$ is dually closed countable but not charming. For a fixed Lindelöf $Σ$ kernel $Y$, we iterate the Hu–Yin boundary reduction by a transfinite closed-core process. At every stage, the current iterate is a $D$-space if and only if $X$ is a $D$-space. The iteration stabilizes before $hL(Y)^+$ at the largest closed subspace in which the kernel trace and its complement are both dense. The bound is sharp: for every infinite cardinal $κ$ and every $γ<κ^+$, there is a compact example with a dense $σ$-compact Lindelöf $Σ$ kernel $Y$ satisfying $hL(Y)=|Y|=κ$ whose first stabilization stage is $γ$.

comments
28 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.35909 ↗

Advances in artificial intelligence (AI) raise the prospect of substantial changes in how research is conducted and who can direct discovery. This perspective argues that universities should prepare for these changes by developing AI as well as using it, rather than relying primarily on capabilities designed elsewhere. The relationship is reciprocal: AI can advance disciplinary research, while scientific questions, experimental evidence, and observed failures can guide AI development. The proposal distinguishes three complementary commitments: shared operation of existing AI models, disciplinary AI development, and foundational AI research. Their resource requirements differ, and no single university needs to reproduce a frontier AI company. Shared computational services should be connected to sustained technical teams, laboratory experimentation, and knowledge records that make findings traceable and reusable. Independent evaluation, alignment research, transparent resource allocation, and inclusive academic governance are integral to this infrastructure. Consortia and national facilities can support capabilities beyond individual campuses while preserving room for investigator-led inquiry. Initial programs in experimental and mathematical research would test scientific value, the effort required to validate results, and improvements to AI before expansion. The argument concerns research rather than educational reform and calls for building institutional capacity before the extent and timing of disruption are settled.

fields
General Mathematics
pdf ↗ page arXiv:2610.35919 ↗

We give negative answers to Questions 4.4 and 4.5 in Cao's survey on the Wijsman topology. We construct a metrizable space $Y$ and a compatible metric $d$ such that every finite power $Y^n$ is Baire, while $(\mathrm{CL}(Y),τ_W(d))$ is not Baire. The construction realizes the Vietoris hyperspace of a suitable metric space as an open subspace of a ball hyperspace. We also study the set $S_d$ of points having an open $σ$-$d$-totally bounded neighborhood. We prove that Baireness of $(\mathrm{CL}(X),τ_W(d))$ implies Baireness of $S_d$, and that, whenever $S_d$ is dense in $X$, the following three conditions are equivalent: $(\mathrm{CL}(X),τ_W(d))$ is Baire, $X$ is Baire, and $S_d$ is Baire. As consequences, we obtain complete characterizations for locally $σ$-$d$-totally bounded metric spaces, countable metric spaces, and $σ$-compact metrizable spaces. In our counterexample, $S_e$ coincides with a non-dense discrete component for every compatible metric $e$, showing that the density assumption is essential.

fields
General Mathematics
pdf ↗ page arXiv:2610.35921 ↗

Rings and Algebras7

A ring $R$ with non-zero identity element is said to be centrally essential if for any non-zero element $a$ of $R$, there exist non-zero central elements $x$ and $y$ such that $ax = y$. In the paper, we prove that the minimal order of a finite non-commutative centrally essential ring is equal to $64$. A partial classification of such rings is presented, and the existence of non-isomorphic non-commutative centrally essential rings of minimal order is proved.

comments
11 pages
fields
Rings and Algebras
pdf ↗ page arXiv:2610.36174 ↗

We consider finitely generated free modules and their endomorphisms over an Artinian local PIR, also known as an SPIR. We characterize submodules of such modules by constructing minimal generating sets from given bases. We prove that any endomorphism possessing a square-free annihilating polynomial modulo the maximal ideal admits a principal annihilating ideal, and establish that every free invariant submodule under such an endomorphism has an invariant complement. Moreover, for such endomorphisms, we detail the construction of invariant submodules and compute the number of free invariant submodules when the ring is finite. Furthermore, we introduce canonical matrices associated with these endomorphisms, yielding a complete characterization of matrix similarity via invariant polynomials in this context. Finally, we address the non-derogatory case, obtaining a sharper characterization of submodule invariance.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.36447 ↗

We study the Kothe radical (the upper nil radical) in the variety of Novikov algebras. We prove that the class of weakly right nilpotent Novikov algebras coincides with the class of r-nil algebras and forms a radical class. We establish that the Kothe radical is a hereditary radical in the variety of Novikov algebras. Furthermore, we obtain a structural description of the Kothe radical as the intersection of ideals whose corresponding quotient algebras are prime K-semisimple algebras. Finally, we construct an example of a prime r-nil Novikov algebra over a field of characteristic p>0, which implies that the Baer radical is strictly contained in the Kothe radical in this case.

comments
13 pages
fields
Rings and Algebras, Quantum Algebra
pdf ↗ page arXiv:2610.36903 ↗

We study $δ$-derivations of the finite-dimensional perfect Lie algebra $\mathfrak{saff}_{n+1}=\mathfrak{sl}_{n+1}(\mathbb{C})\ltimes\mathbb{C}^{n+1}$, where $\mathbb{C}^{n+1}$ is the natural $\mathfrak{sl}_{n+1}(\mathbb{C})$-module. For $n>1$ and $δ\ne1$, we prove that every nonzero $δ$-derivation occurs at $δ=\frac12$ and is a scalar multiple of the identity. For $n=1$, we explicitly describe the four-dimensional space of $(-2)$-derivations.

comments
7 pages
fields
Rings and Algebras
pdf ↗ page arXiv:2610.37027 ↗

All traditional encryption schemes share a fundamental operational vulnerability: encrypted data must be decrypted before it can be used. Fully homomorphic encryption eliminates this vulnerability by enabling computation directly on encrypted data, but existing FHE schemes impose substantial practical constraints that limit deployment. This paper presents a formal security analysis of the Gribov-Kahrobaei-Shpilrain (GKS) private-key FHE scheme. GKS requires no bootstrapping or de-linearization, produces bit-exact decryption, supports native encrypted search, and scales to unbounded depth.

comments
41 pages
fields
Rings and Algebras
pdf ↗ page arXiv:2610.37061 ↗

The commuting graph of a noncommutative ring has its noncentral elements as vertices, with two distinct vertices adjacent when they commute. Akbari, Ghandehari, Hadian and Mohammadian conjectured that a full matrix ring over a finite field is determined, among rings with identity, by its commuting graph. In this paper, we prove the conjecture for every even matrix size.

comments
4 pages
fields
Rings and Algebras
pdf ↗ page arXiv:2610.37150 ↗

In this paper, we give an affirmative answer to the open question posed by G.Birkhoff and R.S.Pierce in 1956 concerning the existence of a compatible lattice order on the complex field $\mathbb{C}$. Using valuation rings theory, basis extension in infinite-dimensional vector spaces and extension of field embeddings to algebraic extensions, we obtain that for every real $t>1$ transcendental over $\mathbb{Q}$, the compatible lattice order can be chosen such that $0\prec1$, $0\vee i=t$, and $1\wedge t=0$. Consequently, $\mathbb{C}$ admits at least $2^{\aleph_0}$ distinct compatible lattice orders, each extending the usual order on $\mathbb{Q}$.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.37155 ↗

Classical Analysis and ODEs6

We introduce a new logarithmic digit variation for Vilenkin–Fourier partial sums and establish two-sided estimates for the corresponding Lebesgue constants with universal constants. These estimates hold for both bounded and unbounded Vilenkin systems. As an application, we show that bounded logarithmic variation yields uniform boundedness of the corresponding partial-sum operators on \(L_1\), \(L_1\)-norm convergence for every integrable function, and almost everywhere convergence.

fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.35900 ↗

We prove a separated entropy bump theorem for two-weight $L^2$ inequalities for positive operators defined by Carleson sequences on a probability space with a general filtration. The bump conditions use conditional averages of martingale maximal functions and involve the first power of the entropy penalty, removing the square appearing in the separated condition of Treil and Volberg. As a consequence, we obtain separated Orlicz bump estimates for a class of doubling Young functions. These results apply, in particular, to sparse operators. The proof uses principal sequences, a natural global extension of principal sets, and combines a good-$λ$ estimate with a weighted Carleson embedding to obtain the required testing inequalities.

fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.35949 ↗

In this paper, we study the spherical maximal operator $\mathcal{M}_{S^{d-1}(A)}$, which is associated with the sphere $S^{d-1}$ embedded in the horizontal plane of some nilpotent group $\mathbb{G}_A$. Here, $\mathbb{G}_A$ is the group identified with $\mathbb{R}^{d+1}$, and its group law is determined by the $d\times d$ real matrix $A$. Specifically, when $A$ is a skew-symmetric matrix $J$, which is the case of the Heisenberg group, it was known that the spherical maximal operator $\mathcal{M}_{S^{d-1}(J)}$ is bounded on $L^p(\mathbb{G}_J)$ for $p$ in the range $(\frac{d}{d-1},\infty]$. In this paper, we extend the results from the skew-symmetric matrix $J$ to a more general class of invertible matrices $A$. We aim to classify the range of $p$ for which the operator $\mathcal{M}_{S^{d-1}(A)}$ is bounded on $L^p(\mathbb{G}_A)$ based on the eigenvalues of $A$.

journal
Studia Math (2025) 282(1) 25-26 Published at Studia Math 282(1) 25-67 2025
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.36828 ↗

For one-dimensional oscillatory integral operators with real-analytic phases, the seminal work of Phong and Stein established that the Newton polygon of the mixed Hessian of the phase determines the sharp local \(L^2\)-decay. The local theory relies on spatial localization near the origin and therefore does not capture the global geometry of the real zero set of the Hessian. For globally defined operators, unbounded real zero curves of the Hessian can make the global decay strictly slower than the local decay or destroy \(L^2\)-boundedness altogether. We establish a global analogue of the Phong–Stein theory for quasi-homogeneous polynomial phases. We introduce a family of Newton polygons adapted to the regions near and away from the real zero curves of the Hessian, together with the corresponding box-area scales. The relative geometry of these Newton polygons and area scales characterizes global \(L^2\)-boundedness and determines the sharp \(L^2\)-estimates in the high- and low-frequency regimes. As an application, we determine the exact range of anisotropic homogeneous \(L^2\)-Sobolev estimates for the associated global Radon transform.

fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.36969 ↗

We establish sharp $L^p$ maximal estimates for averages over smooth nondegenerate curves in $\mathbb R^n$ for $p>n$. More precisely, we show that for any nondegenerate curve $γ$ in $\mathbb R^n$ and the associated maximal operator \begin{equation} M f(x):=\sup_{t>0} \Big|\int_{-1}^1f(x-tγ(s))χ(s)\,\mathrm{d}s\Big|, \end{equation} the estimate \begin{equation} \| Mf\|_{L^p(\mathbb R^n)}\lesssim_{γ,p,n} \|f\|_{L^p(\mathbb R^n)} \end{equation} holds for all $p>n$. A new ingredient is a weighted Kakeya-type estimate.

comments
49 pages
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.37250 ↗

We prove an uncertainty principle for subsets of $\mathbb R$ under weak branching assumptions. We impose a nonlinearity condition on the phase; without it our uncertainty principle fails e.g. for arithmetic progressions. The key ingredient in the proof is Shmerkin's nonlinear projection theorem. This paper is a part of a work in progress, but it is released earlier than the rest of that paper since it is also used in a paper of Cohen-Li-Tao. It also has an overlap with our recent paper [arXiv:2609.30261]; see the end of the introduction for more details.

comments
37+eps pages, 2 figures. Note the overlap with arXiv:2609.30261 explained in the paper
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.37549 ↗

Complex Variables6

We study the explicit bounded pseudoconvex domain \[ Ω=\{(z_1,z_2,w)\in\mathbb C^3: |w|^2+|z_1z_2(z_1-z_2)|^2+|z_1|^8+|z_2|^8<1\}. \] Its boundary is real analytic and has finite D'Angelo type everywhere. We prove two assertions about the invariant metrics of this domain. First, $(Ω,k_Ω)$ is not Gromov hyperbolic; in fact arbitrarily fat Kobayashi geodesic triangles occur in every neighborhood of the type-$8$ point $(0,0,1)$. Second, the Carathéodory–Reiffen and Kobayashi–Royden metrics are not uniformly comparable. Along an explicit family of tangential directions the Carathéodory metric tends to zero while the Kobayashi metric has a positive limit. For the corresponding distances there are points $x_j,y_j\to(0,0,1)$ such that $c_Ω(x_j,y_j)\to0$ while $k_Ω(x_j,y_j)\to\infty$. This gives, among bounded real-analytic pseudoconvex domains, an affirmative answer to Problem~7.3.2 of the Several Complex Variables Problem List.

fields
Complex Variables
pdf ↗ page arXiv:2610.36124 ↗

This paper provides a human digestion of the recent counterexample to Smale's mean value problem found by GPT-6 Astra. We give a geometric mechanism for constructing counterexamples. An outward normal that re-enters an analytic Jordan domain, subject to a curvature condition, yields a holomorphic model whose normalized divided differences exceed one in modulus on the boundary. Polynomial perturbations then concentrate every critical point near the boundary whilst preserving the strict inequality. This turns boundary geometry into counterexamples in every sufficiently large degree. We also locate these critical points to first order and show that their limiting distribution is the logarithmic equilibrium measure on the boundary. We end with an explicit example of degree $29\,984$, verified by interval arithmetic.

fields
Complex Variables
pdf ↗ page arXiv:2610.36158 ↗

We establish the long time existence of the inverse Monge–Ampère flow on a klt $\mathbb{Q}$-Fano variety $X$. We further prove that $X$ admits a Kähler–Einstein metric if and only if the flow converges in $L^1$.

fields
Complex Variables, Analysis of PDEs, Differential Geometry
pdf ↗ page arXiv:2610.36414 ↗

Using different approach we consider some problems for the class $\mathcal{S}_{ex}^{\star}$ defined by the condition $\frac{zf'(z)}{f(z)}\prec e^{αz}$, $0<α\leq1.$ Our method is more simple than that used by Das and Sarkar in their work: \textit{Sharp coefficient estimates for the exponential starlike class $S^\star_{ex}$} (\cite{DS26}).

fields
Complex Variables
pdf ↗ page arXiv:2610.36737 ↗

Given a radial doubling weight $μ$ on the unit disk $\mathbb{D}$ and $f(z)=\sum_{n=0}^\infty \hat{f}(n)z^n\in\mathcal{H}(\mathbb{D})$, the fractional derivative of $f$ induced by $μ$ is $$ D^μ(f)(z) = \sum_{n=0}^\infty \frac{\hat{f}(n)}{μ_{2n+1}}z^n, $$ where $μ_{2n+1}=\int_0^1 r^{2n+1}μ(r)\,dr$ are the odd moments of $μ$. The fractional integral of $f$ induced by $μ$ is $I^μ(f)(z)=\sum_{n=0}^\infty μ_{2n+1}\hat{f}(n)z^n$. We study the fractional Volterra-type operator $$ V_{μ,g}(f)(z)=I^μ(f\cdot D^μ(g))(z),\quad f\in\mathcal{H}(\mathbb{D}), $$ with symbol $g\in\mathcal{H}(\mathbb{D})$, introduced by Bellavita, Moreno, Nikolaidis and Peláez. First, we estimate the essential norm of $V_{μ,g}$ acting on $H^p$ and $A^p_α$. Later, we study the same questions for $V_{μ,g}:H^p\to H^q$ and $V_{μ,g}:A^p_α\to A^q_α$, characterizing their boundedness and compactness and estimating their essential norms for all $0<p,q<\infty$.

fields
Complex Variables
pdf ↗ page arXiv:2610.37548 ↗

In this note, using J.-P. Demailly's $L^2$ existence theorem for possibly incomplete Kähler metrics, we establish $L^2$ estimates and vanishing theorems for a class of domains in $\mathbb P^n$ without any boundary regularity assumptions. In particular, we give an affirmative answer to M.-C. Shaw's question concerning the vanishing of the $L^2$ Dolbeault cohomology groups of projective Hartogs triangles.

comments
12 pages
fields
Complex Variables
pdf ↗ page arXiv:2610.37572 ↗

Logic6

We give a direct proof of Infinity in the no-empty version of New Foundations, denoted NF-0, by establishing that every model of the specified theory contains an object that is not finite in a precise, internally defined sense. The version considered has extensionality, at least two distinct objects, no empty objects, and inhabited stratified comprehension: a stratified extension is supplied whenever it has a member. Finiteness combines a two-sided internally represented well-order with the exclusion of internally represented nonterminating strict transitive relations on inhabited subsets. Assuming all objects finite, partial bijections yield comparison and finite extrema of positive Frege cardinals. Positive exponentiation, singleton type-raising, and the least orbit of the genuine two-cardinal lead to the equation n = T(n) + 1. We prove separately, and in detail, that adjoining one point reverses internally defined parity: the key ingredients are a graph-rewiring proof of invariance under adjoining two points and one instance of stratified induction derived from the all-finite assumption. Since type-raising preserves parity, the cardinal equation is impossible. We identify the corresponding Lean 4 declarations and make explicit the scope of the formal verification. No model-existence or consistency theorem for ordinary NF is claimed.

comments
10, a 2-page short copy present also
fields
Logic
pdf ↗ page arXiv:2610.35928 ↗

Universally Baire sets are a standard tool in modern set theory, where they are closely connected with forcing and absoluteness. In this paper we study universal Baireness instead as purely a regularity notion for subsets of Polish spaces. Our aim is not to develop the set-theoretic theory, but to place the notion on a systematic descriptive-set-theoretic footing and to clarify the several nonequivalent definitions that occur in the literature. A subset of a Polish space is called \emph{universally Baire} if all its continuous preimages possess the property of Baire. However, it is important to specify what the domain of these continuous functions can be. In fact, a plethora of definitions are used by various authors in the literature, the class of possible domains is sometimes the class of Polish spaces, compact metrizable spaces, all topological spaces, compact topological spaces, etc. We clarify this situation by demonstrating that the above four options are pairwise nonequivalent under the Continuum Hypothesis, we determine all the implications between these four notions that are provable in $ZFC$, and also show that all the notions used in the literature are equivalent to one of these four. Furthermore, we use Miller forcing to show that actually the cases of Polish and compact metrizable spaces are consistently equivalent. We show that numerous definitions using Borel isomorphisms, changes of topologies and Boolean algebras are all equivalent to the case of Polish domains. We present an argument that the notion using Polish domains is the natural dual to the notion of universally measurable sets. Finally, as a byproduct, we prove that the following statement is independent of $ZFC$: if a subset of a Polish space has the property of Baire relative to every nowhere dense Cantor subspace, then the subset has the property of Baire in the whole space.

fields
Logic, General Topology
pdf ↗ page arXiv:2610.36265 ↗

The fibrational framework of Licata, Shulman, and Riley gives a uniform account of substructural and modal logics for single-conclusion judgments. This thesis extends it to a multiple-output setting in which cartesian resources may be used freely, while linear resources occur exactly once on either side of the turnstile. Judgments are indexed by descriptors: directed acyclic graphs generalizing the tree-shaped expressions of the original framework. Descriptor $2$-cells encode structural transformations, and explicit gates provide weakening, contraction, coweakening, and cocontraction at linear modes. We define a sequent calculus generated by two connectives, $F$ and $U$. We prove identity, invertibility, and cut admissibility. The mode and type theories present linear-non-linear $2$-polycategories, and the projection is a local discrete fibration and a fibration, with opcartesian lifts for one-output base morphisms. Two adequacy results validate the framework: its cartesian fragment recovers the Licata–Shulman–Riley calculus, and its purely linear fragment presents classical multiplicative linear logic. Finally, we relate indexed and fibrational semantics. After developing representability correspondences for multicategories and polycategories, we establish an equivalence between multi-indexed categories over a fixed representable base and multicategorical opfibrations over it, together with its biclosed bifibrational version. We also give a one-way construction from planar poly-indexed data to a planar polycategory.

fields
Logic
pdf ↗ page arXiv:2610.36552 ↗

In this paper, we consider conservation results for iterated $β_n$-model reflection principles. Using these principles, we characterize the $Π^1_{m+2}$-, $Σ^1_{m+2}$- and $\mathsf{B}(Π^1_{m+2})$-consequences of $Σ^1_{n+1}$-$\mathsf{DC}_0$ and $Σ^1_{n+1}$-$\mathsf{TDC}_0$ for $m \leq n$. Here, $\mathsf{B}(Π^1_{m+2})$ denotes the set of Boolean combinations of $Π^1_{m+2}$-formulas, and $\mathsf{TDC}$ stands for transfinite $\mathsf{DC}$. We also show that, in each case, the stronger theory has a super-exponential speed-up over the corresponding conservative subtheory.

fields
Logic
pdf ↗ page arXiv:2610.36982 ↗

We study algebras for the $\mathbf{BM}$-, $\mathbf{BI}$-, and $\mathbf{BMI}$-fragments of combinatory logic, where $\mathbf{B} xyz \to_{w} x\left(yz\right)$, $\mathbf{M} x \to_{w} xx$ and $\mathbf{I} x \to_{w} x$. We prove that every finite $\mathbf{BM}$-algebra $\mathcal{A}$ contains an element $Y$ satisfying $Yx = x\left(Yx\right)$ for all $x\in A$. Moreover, a single closed $\mathbf{BM}$-term represents such an element in every finite $\mathbf{BM}$-algebra. On the other hand, we show that there exists an infinite $\mathbf{BMI}$-algebra with no such element. Consequently, the variety of $\mathbf{BM}$-algebras is not generated by finite $\mathbf{BM}$-algebras. We also characterise finite $\mathbf{BI}$-algebras as finite semigroups with two distinguished left units $B$ and $I$, and finite $\mathbf{BMI}$-algebras as finite idempotent semigroups with three distinguished left units $B$, $M$ and $I$.

fields
Logic
pdf ↗ page arXiv:2610.37108 ↗

We prove Szemerédi-Trotter-type incidence bounds for algebraic relations in any field $F$ admitting a valuation with finite residue field. This includes $\mathbb F_q(t)$, $\mathbb F_q((t))$, and all finitely generated extensions of finite fields. Bays and Martin show that any algebraic $K_{d,s}$-free relation $E\subseteq F^k\times F^\ell$ admits a power-saving in the classical Zarankiewicz bound of Kővári, Sós, and Turán. Answering a question of the authors, we show that this power-saving is uniform in $F$: There is some $ε>0$ depending only on $d$ and the complexity of $E$ (and not on $F$) such that for $A\subseteq F^k$ and $B\subseteq F^\ell$, the number of edges $|E(A,B)|$ is at most $O(|A|^{1-ε}|B|^{1-\frac{1-ε}{d}}+|A|+|B|)$. For the case where $A$ is a set of points and $B$ is a set of lines in $F^2$ and $E$ is the incidence relation, we prove the explicit bound $O(|P|^{\frac{8}{9}}|L|^{\frac{5}{9}}+|P|+|L|)$. As the central ingredient of the proof, we introduce the model-theoretic notion of distality relative to a definable set. We begin developing this notion and prove that the theory of algebraically closed valued fields is distal relative to the residue field. An application of a result of Chernikov, Peterzil, and Starchenko, based on work of Chernikov, Galvin, and Starchenko, yields our incidence bounds.

comments
30 pages
fields
Logic, Combinatorics
pdf ↗ page arXiv:2610.37532 ↗

History and Overview5

We give a geometric introduction to Schubert polynomial multiplication through point-line flags in the complex projective plane. Moving a prescribed point onto a prescribed line gives a family with local equation $au=λ$, whose limiting curve has two components. The incidence equations and their local multiplicities give a two-term divisor product, using the standard intersection product on classes of algebraic cycles. Permutation labels and Schubert polynomials express the same decomposition as an instance of Monk's rule. For three incidence conditions, each allowing either a point on a prescribed line or a line through a prescribed point, the product has degree six. Under explicit general-position assumptions, direct construction and local equations give six distinct solutions of multiplicity one.

comments
15 pages, 3 figures
fields
History and Overview, Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.35944 ↗

The story of frieze patterns started in 1970s with a paper by Coxeter and two short papers by Conway and Coxeter, one listing questions and the other providing brief answers. During the first three decades it was followed by only a handful of papers observing some further properties of the same objects. Then Propp (2005) and Caldero and Chapoton (2006) established connections between frieze patterns and the recent theory of cluster algebras introduced by Fomin and Zelevinsky in early 2000s. Since then the theory of frieze patterns exploded with new results, and frieze patterns found a wide spectrum of connections in various domains of mathematics, including combinatorics, geometry, representation theory and integrable systems. The literature on frieze patterns grew vastly and continues to expand. In this note we give a brief overview of some of the directions emerging from the papers by Conway and Coxeter and single out core ideas which were later replayed in different contexts.

comments
This survey was written in November 2025 for the prospected book with a tentative title 'The Conway Book', in preparation
fields
History and Overview, Combinatorics, Rings and Algebras
pdf ↗ page arXiv:2610.35968 ↗

This paper presents a proof of the law of quadratic reciprocity using the Vandermonde matrix and Zolotarev's lemma.

journal
The American Mathematical Monthly, 133(4), 2026
fields
History and Overview
pdf ↗ page arXiv:2610.35990 ↗

Mathematical software has particular needs that are not all covered by the existing literature on software engineering or algorithm design. This paper looks at some of these aspects, in the concrete context of the computer algebra system GAP. It also studies the potential impact AI systems could have on the development process for mathematical software.

journal
Jahresber. Dtsch. Math. Ver., 2026
fields
History and Overview
pdf ↗ page arXiv:2610.36300 ↗

A proof answers a precise question, but mathematical understanding also grows through the relationships that its answer makes visible. A point measurement likewise differs from a photograph or a three-dimensional reconstruction of a building. In this work, we argue that, with the help of AI, mathematics should move from a discipline focused on isolated investigations of mathematical objects towards one capable of mapping them systematically. We therefore introduce the idea of "Mathematical Reality Capture", inspired by digital surveying and the reconstruction of a scene from several viewpoints. An elementary example shows how comparing algebra and geometry can illuminate the same phenomenon, provided that we distinguish diversity of explanations from validity of proof. From this analogy we derive criteria for building verifiable atlases, an experiment with which to assess their usefulness, and a reflection on the role of mathematicians in interpreting them.

fields
History and Overview
pdf ↗ page arXiv:2610.37260 ↗

Quantum Algebra5

We study the Duflo–Serganova functor for vertex algebras of chiral differential operators on classical invariant rings, introduced in \cite{ADS}. Our main result is a general cancellation theorem showing that a common bosonic and fermionic summand can be removed by Duflo–Serganova reduction. We apply this theorem to obtain Duflo–Serganova reduction isomorphisms for two families of simple orthosymplectic affine vertex superalgebras at levels $-2n$ and $n$, which are not covered by the existing results of Gorelik and Serganova. As a consequence, we obtain supercharacter identities, including the conjecture of Adamović and Milas as a special case.

comments
7 pages, comments are welcome
fields
Quantum Algebra, Representation Theory
pdf ↗ page arXiv:2610.34954 ↗

We present AlgebrasTQD, an open-source code written in GAP (Groups, Algorithms, and Programming), for computing condensable algebras and the corresponding topological boundaries and interfaces of twisted quantum doubles $D^ω(G)$. These algebras can be used to classify 1+1d gapped and gapless phases whose symmetry is related by gauging to a finite group $G$ with anomaly $ω\in H^3(G,U(1))$, to study 2+1d anyon proliferation and to construct logical gates in non-Abelian topological quantum codes. The functions in AlgebrasTQD compute all condensable algebras $A(H,F,γ,ε)$ with their algebra structure and anyons, sort them into isomorphism classes and identify the twin algebras that are inequivalent but share the same anyon decomposition. AlgebrasTQD also determines all the subalgebra relations, the corresponding Hasse diagram and the types of 1+1d phases for any chosen, possibly non-invertible, group-theoretical symmetry. For non-maximal algebras, the code computes the reduced topological order with its $3$-cocycle and the map of anyons across the topological interface. Finally, AlgebrasTQD includes functions that construct certain non-Clifford logical gates via non-Abelian surface codes. We summarize the formulas that are implemented in the AlgebrasTQD code, and we illustrate its use with Abelian and non-Abelian groups including $\mathbb{Z}_2^3$ with $ω_\text{III}$, $D_8$, $A_4$ with and without anomaly, twin algebras and twin phases, and the Clifford-hierarchy gates in quantum doubles of dihedral groups. All arithmetic is exact and AlgebrasTQD can be used with any finite group $G$ and 3-cocycle $ω$, limited only by the time and size of the calculation. The code is available at https://github.com/AlisonWarman/AlgebrasTQD

comments
38 pages, companion paper to the code at https://github.com/AlisonWarman/AlgebrasTQD
fields
Quantum Algebra
pdf ↗ page arXiv:2610.36029 ↗

We prove that for a simple and $C_2$-cofinite vertex operator algebra, the rationality is equivalent to the modular invariance of the space of complete ordinary one-point trace functions of its irreducible modules. This proves a conjecture given by Dong-Ren in \cite{DongRen}, which is also the inverse theorem of Zhu's modular invariance theorem \cite{Zhu}.

fields
Quantum Algebra
pdf ↗ page arXiv:2610.36565 ↗

For twisted quantum affine algebras, we compare Lusztig's restricted integral form with the integral form generated by group-like toral elements and divided powers of real Drinfeld current modes. The two forms coincide over $\mathbb{Z}[q,q^{-1}]$ outside type $A_{2n}^{(2)}$. In type $A_{2n}^{(2)}$, the smallest ordinary localization over which they coincide is $\mathbb{Z}[q,q^{-1}][Φ_{4r}(q)^{-1}\mid r\geq1]$ when $n=1$, and $\mathbb{Z}[q,q^{-1}][Φ_{8r}(q)^{-1}\mid r\geq1]$ when $n\geq2$, where $Φ_k(q)$ are cyclotomic polynomials. In particular, no localization obtained by inverting finitely many nonzero Laurent polynomials suffices in this family.

comments
61 pages
fields
Quantum Algebra, Representation Theory
pdf ↗ page arXiv:2610.36605 ↗

We study the relationship between semisimplicity and non-degeneracy for finite-dimensional Hopf triplets over a field of characteristic zero. The non-degeneracy condition appears in the construction of \(4\)-manifold invariants from Hopf triplets, and it was not known whether non-semisimple examples satisfying this condition exist. We prove that a Hopf triplet is non-degenerate if and only if it is semisimple.

comments
10 pages
fields
Quantum Algebra
pdf ↗ page arXiv:2610.36632 ↗

Algebraic Topology3

The Maxwell graph of a directed acyclic graph is obtained from its iterated basin and hill boundaries. The unweighted counts of extrema at these levels need not satisfy Morse inequalities. We assign multiplicities by the reduced homology of lower and upper order links and retain the vertices with nonzero local homology. The resulting graph preserves both boundary hierarchies, the local groups, and the homology of all sublevel and superlevel order complexes. Its descending and ascending Morse numbers satisfy the weak and strong Morse inequalities over any field. We also construct a Morse chain complex and compute the local groups in several examples. Finally, we consider homology on the original Maxwell graph and prove analogous inequalities for path homology using relative local groups.

comments
15 pages
fields
Algebraic Topology, Combinatorics
pdf ↗ page arXiv:2610.36372 ↗

Zigzag persistence modules serve as an essential tool in topological data analysis. The fundamental theorem of persistent homology states that any zigzag persistence module decomposes into a direct sum of simple building blocks called interval modules. We consider an interval cover, a possibly redundant family of interval submodules spanning every vertex space of a finite zigzag persistence module. We give an elementary descent argument that replaces a redundant cover with a minimal one that is an interval decomposition. This provides an optimisation perspective on persistence modules and their interval decomposition. As an application, we give an estimate of the generalised rank invariant in terms of a cover.

comments
9 pages, 1 figure
fields
Algebraic Topology
pdf ↗ page arXiv:2610.36843 ↗

We determine the multiplicative structures of the homotopy operations on derived Lie algebras, derived restricted Lie algebras and partition Lie algebras, as well as the relation between these. Under Koszul duality, this follows from a calculation of the homotopy operations on derived rings, p-truncated algebras and divided power algebras over $\mathbb{F}_p$.

comments
57 pages
fields
Algebraic Topology
pdf ↗ page arXiv:2610.36942 ↗

General Topology3

In this paper, we prove that a one-dimensional Peano continuum $X$ admits a weak homotopy equivalence $X\to C$ to a countable space $C$ if and only if the wild set $\mathbf{w}(X)$ is countable.

comments
14 pages, 5 figures
fields
General Topology, Algebraic Topology
pdf ↗ page arXiv:2610.35875 ↗

Working with tools from infinite-dimensional topology and using measure-theoretic ideas we analyze numerous well-known fami\-lies of copu\-las and quasi-copulas from a topological perspective, prove that most of them are homeomorphic to the Hilbert cube $\mathcal{H}$ and hence have the Fixed Point as well as the Disjoint Cells property. Working with so-called $Z$-sets, we then study how large one set is within another. Doing so, in order to show that a set $A$ is small/topologically negligible within a compact metric space $(Ω,m)$, we construct a homeomorphism $h: [0,1] \times \mathcal{H} \to Ω$ mapping $\{0\} \times \mathcal{H}$ to $A$. The obtained results are to a certain extent surprising since various supposedly small families are topologically large.

fields
General Topology, Functional Analysis, Probability
pdf ↗ page arXiv:2610.35912 ↗

We exhibit sober spaces $X$ and $Y$ such that the function space $[X \to Y]_K$ of continuous maps, equipped with the compact-open topology, is not sober; the codomain can be chosen to be the Sierpiński space $S$. Our construction builds on Isbell's complete lattice $M$, whose Scott space $ΣM$ is non-sober. Let $H = Q(M)$ be the set of nonempty compact saturated subsets of $ΣM$, ordered by reverse inclusion. Since $H$ is well known to be a spatial frame, the space $X = \mathrm{pt}(H)$ is sober with $\mathcal{O}(X) \cong H$. The key technical result is that sobriety of the space $H$ would entail sobriety of the original Scott space $ΣM$. This contradicts Isbell's example, and the desired counterexample follows.

fields
General Topology
pdf ↗ page arXiv:2610.36551 ↗

Operator Algebras3

Consider the countably infinite reduced free product $D$ of copies of $(\mathrm C([0,1]),λ)$, where $λ$ is Lebesgue measure. It is proved that finite and countable reduced free products of unital separable nuclear UCT $\mathrm C^*$-algebras with respect to a chosen family of faithful traces are classified, after adjoining $D$, by the Elliott invariant. In other words, an isomorphism between the naive K-theory invariants of two such algebras is induced by an isomorphism between the algebras.

comments
99 pages
fields
Operator Algebras
pdf ↗ page arXiv:2610.36459 ↗

For a deterministic sequence of normal unital completely positive maps on a von Neumann algebra, convergence of normal states, loss of memory, and convergence of Cesàro averages are different properties. This paper separates these properties in the predual norm and identifies the precise role of composition order. Uniform loss of memory for right-appended products on the predual implies uniform convergence to a normal state without any compactness or trace assumption; the convergence rate is bounded by the image diameter. In contrast, on every infinite-dimensional $σ$-finite von Neumann algebra there are right-appended products with pointwise loss of memory but no convergent normal-state trajectory, even though all nontrivial state images equal a fixed compact line segment. Consequently, finite dimensionality is characterized by the universal implication from pointwise backward memory loss to mixing. An attained normal-functional minorization coefficient extends the common-lower-bound construction for matrix channels to arbitrary von Neumann algebras and yields quantitative contraction and unique normal entrance laws. Further results distinguish pointwise from uniform relaxation, Cesàro convergence from mixing, fixed-start from time-uniform contraction, and stationary from moving asymptotic states. As an application, inhomogeneous transfer states on spin chains, including matrix product states, are shown to have a unique thermodynamic limit, uniformly in the boundary condition, with an arbitrary von Neumann algebra as bond algebra; boundary dependence and correlations are controlled by the same product coefficients. All counterexamples are explicit and extend to algebras of types I, II, and III.

fields
Operator Algebras
pdf ↗ page arXiv:2610.37143 ↗

We extend results on tracial approximability of crossed products from the setting of topological dynamical systems to C*-dynamical systems. For actions of discrete amenable groups on unital C*-algebras, we introduce a weak Rokhlin property which leads to a weak tracial approximation of the crossed products by means of finite direct sums of matrix algebras over hereditary subalgebras of the original algebra. Examples are given.

fields
Operator Algebras
pdf ↗ page arXiv:2610.37558 ↗

Spectral Theory3

We consider the Robin Laplacian on a planar exterior domain with a negative boundary parameter. For a negative boundary parameter of a sufficiently large absolute value this operator has at least two negative eigenvalues below the bottom of the essential spectrum. We prove that under a fixed area constraint and convexity assumption on the bounded complement of the exterior domain the second Robin eigenvalue is maximized by the exterior of the disk. The inequality is strict for every exterior of a non-circular domain whenever the comparison disk has a negative second eigenvalue. The proof modifies the parallel-coordinate method by allowing the angular phase of a trial function to vary with the distance from the boundary. This result resolves the first item in a 2024 conjecture by the authors under an additional convexity assumption. Relaxing the convexity and the perimeter constraint remain open problems.

comments
12 pages
fields
Spectral Theory, Mathematical Physics, Analysis of PDEs, Optimization and Control
pdf ↗ page arXiv:2610.34320 ↗

Let $M$ be a noncompact finite-area hyperbolic surface. We give a lower bound for the $L^2$ norm of eigenfunctions and Eisenstein series in a compact part of the surface. This extends a result of Soundararajan [arXiv:0901.4060] to the nonarithmetic setting. The crucial components of the proof are a long-time Egorov's theorem designed for this setting, a nonlinear uncertainty principle of Cohen–Dyatlov [arXiv:2609.30261], and an application of Weyl's method for exponential sums.

comments
73 pages, 7 figures
fields
Spectral Theory, Analysis of PDEs, Dynamical Systems
pdf ↗ page arXiv:2610.37573 ↗

Brains do not simply add up every signal they receive: they select which routes dominate. Yet connectomes are usually analyzed with tools that sum over all paths and place neural states in Euclidean spaces, blurring the feedback circuits that may sustain internal states. Here we treat neural signaling as routing instead of diffusion and describe it with max-plus algebra, in which signals accumulate along a path and the strongest route prevails. We propose a model in which the functional connectome of Caenorhabditis elegans emerges from the time-varying combination of electrical, synaptic, monoaminergic, neuropeptidergic and hormonal layers, and analyze five networks within this framework. Each collapses onto a handful of dominant feedback loops, which recover well-characterized circuits and point to new candidates for experimental study. Unlike linear spectra, these structures keep the biological meaning of the data, and their order-based description suggests that brain states live in lattices rather than Euclidean spaces.

fields
Spectral Theory
pdf ↗ page arXiv:2610.37587 ↗

Symplectic Geometry2

Given a compact surface $S$ admitting a Liouville form, a finite set $Q \subset S$, a line field on $S$, and a local system of dg-categories on $S \setminus Q$ together with a section of the monodromy functor around each interior $q \in Q$, we construct a Fukaya category. We give examples where such Fukaya categories are quasi-equivalent to derived categories of projective completions of line bundles and of iterated blow-ups of the affine plane; and to wrapped Fukaya categories of cotangent bundles of spheres. We define two versions of the Fukaya category, a topological, which is constructed combinatorially from a full arc system, and a geometric, which includes all Lagrangian branes supported on immersed curves. We prove that the topological category embeds into the geometric category in an essentially unique way. To define Lagrangian branes we develop a generalisation of the Chekanov–Eliashberg algebra incorporating dg-coefficients and wrapping along the boundary.

comments
91 pages, 29 figures
fields
Symplectic Geometry, Algebraic Geometry, Category Theory
pdf ↗ page arXiv:2610.36119 ↗

We study the geometry and topology of transversely vanishing eigenforms of the curl operator on $3$-manifolds, placing them in the broader framework of near-contact forms. We prove a Gray-type stability theorem for the kernels of near-contact forms, which we call near-contact structures, and derive a local neighborhood theorem around their singularities. As applications, we obtain a stability result for the contact topology of curl eigenforms, a classification of near-contact structures in terms of formal homotopical data, and a proof of the existence of contact structures on closed orientable 3-manifolds based on the spectral theory of the curl operator. We also study the symplectization of curl eigenforms as a possible bridge to symplectic dynamics. Finally, we characterize those transversely vanishing $1$-forms which are curl eigenforms for some smooth Riemannian metric. We deduce a generalization of the classical correspondence, due to Etnyre and Ghrist, between Reeb-like fields of contact forms and nowhere-vanishing rotational Beltrami fields to the setting of transversely vanishing vector fields.

comments
69 pages, 4 figures
fields
Symplectic Geometry, Differential Geometry, Dynamical Systems, Spectral Theory
pdf ↗ page arXiv:2610.36212 ↗

Category Theory1

We propose a definition of a finite linear 2-category, as a 2-categorical analogue of the notion of a finite abelian category of Etingof-Ostrik. We go on to prove that finite linear 2-categories with adjunctible projectives are precisely the Morita 2-categories of (all) algebras in a finite multitensor category. Thus, finite linear 2-categories with adjunctible projectives can be viewed as `non-semisimple' generalizations of the finite semisimple 2-categories of Douglas-Reutter. Secondly, we study finite multitensor 2-categories, that is, finite linear 2-categories equipped with a monoidal structure for which every objects has duals. We extend several results about fusion 2-categories to our present setting. In particular, we investigate the properties of the 2-category of bimodules over a rigid algebra $\mathcal{A}$ in a finite tensor 2-category $\mathfrak{C}$. We show that, if $\mathfrak{C}$ has adjunctible projectives, then the 2-category of $\mathcal{A}$-$\mathcal{A}$-bimodules in $\mathfrak{C}$ is a finite multitensor 2-category. This construction provides many examples of finite tensor 2-categories. For instance, the Drinfeld center of a finite tensor 2-category with adjunctible projectives is a finite tensor 2-category.

comments
39 pages, comments welcome!
fields
Category Theory, Quantum Algebra
pdf ↗ page arXiv:2610.37016 ↗

Commutative Algebra1

How many fixed points does belief propagation have on a binary pairwise model? Over complex parameters, the admissible equations of a finite loopless multigraph form an algebraic family whose generic degree we determine: it is a graph state sum over edge sets without vertices of degree one, each vertex of positive degree contributing one less than its degree. We give two geometric realizations of this state sum. The first is the total Betti number of a universal edge-incidence ring constructed without parameters. Its full graded Betti table is the state sum, its Tor algebra is natural for noncontracting graph maps, the alternating numerator recovers the graph polynomial of Watanabe and Fukumizu, and an Artin reduction by a regular sequence realizes the one-variable specialization, corrected for tree components. The second lies in the actual family: at the strong-field boundary each state factor is a ramification index, and at sufficiently strong field the state charts exhaust the admissible complex fiber. For fixed nonzero real couplings, every Eulerian message-sign sector contains a real fixed point at every external field, and strong fields attain this lower bound: one admissible real fixed point per element of the binary cycle space and one positive fixed point. For graphs of maximum degree at most three, this minimum equals the number of minimal module generators of the edge-incidence ring. One state decomposition connects the complex degree, graded syzygies, and an attained minimum real count. The degree, selected ring and resolution statements, and the real counts are formalized in Lean 4.

code
github.com/ywatanabex/bp-degree-syzygies-lean ∎ Lean project (lakefile.toml, lean-toolchain) is too large to build within the check's time limit at 2c097b1, checked 2026-10-28 build details
comments
33 pages (main text and supplementary material SM1-SM6 in one PDF), 1 figure. Lean 4 formalization of the main results: https://github.com/ywatanabex/bp-degree-syzygies-lean
fields
Commutative Algebra, Information Theory, Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.36964 ↗

K-Theory and Homology1

Let $K$ be a field of characteristic $2$, let $ρ\in K^\times$, and let $R_ρ$ be the corresponding Liu-Schulz algebra. We give a complete description of the Gerstenhaber bracket and the BV structure on the Hochschild cohomology ring $\operatorname{HH}^*(R_ρ)$.

comments
20 pages
fields
K-Theory and Homology
pdf ↗ page arXiv:2610.36277 ↗

Revised papers243

The main result in this paper is motivated by the Mézard-Parisi ansatz which predicts a very special structure for the distribution of spins in diluted mean field spin glass models, such as the random K-sat model. Using the fact that one can safely assume the validity of the Ghirlanda-Guerra identities in these models, we prove hierarchical exchangeability of pure states for the asymptotic Gibbs measures, which allows us to apply a representation result for hierarchically exchangeable arrays proved in arXiv:1301.1259. Comparing this representation with the predictions of the Mézard-Parisi ansatz, one can see that the key property still missing is that the multi-overlaps between pure states depend only on their overlaps.

comments
Corrected minor errors and allowed more general overlap intervals in the definition of clusters
journal
Probability Theory and Related Fields, 161 no. 3 (2015) 619-650
fields
Probability, Mathematical Physics
pdf ↗ page arXiv:1307.2207 ↗

We solve the double coset problem for arbitrary parabolic subgroups of the $n$-strand braid group $B_n$, by reducing it to the simultaneous conjugacy problem in the larger group $B_{n+1}$. The resulting decision algorithm is uniform in the braid index and the supplied subgroup generators. We also formulate the reduction for general group embeddings, and apply it to additional commuting constraints, several other spherical Artin types, and finite direct products.

comments
Submitted for publication
fields
Group Theory
pdf ↗ page arXiv:1402.5541 ↗

We revise and extend the algorithm of F. G. Scholtz and B. Chakraborty [J. Phys. A: Math. Theor. 46 (2013) 085204] to compute the finite Connes' distance between normal states on the Moyal plane and fuzzy sphere, using the Hilbert-Schmidt operatorial formulation. The original expression gives only a lower bound. The revised expression involves an infimum over the "transverse" component $Δρ_\perp$ of the algebra element in addition to the "longitudinal" component $Δρ$, identified with the difference of density matrices representing the states. Determining the transverse component for which the infimum is reached remains non-trivial, but under rather generic conditions the distance is proportional to the Hilbert-Schmidt norm of $Δρ$. We also determine an upper bound by adapting the approach of P. Martinetti and L. Tomassini [Commun. Math. Phys. 323 (2013) 107-141], and look for an optimal element saturating it. For the Moyal plane, such an element is obtained through the limit of a sequence of finite-dimensional projections of the representative of a multiplier-algebra element onto subspaces spanned by Dirac eigen-spinors. In contrast, on the fuzzy sphere the upper bound, given by the geodesic distance on a commutative sphere, is never reached for any finite $n$-representation of $SU(2)$. For maximal non-commutativity ($n=1/2$), the finite distance coincides with the lower bound and the transverse component plays no role. This changes from $n=1$ onwards, where the revised algorithm improves the estimate and gives an almost exact result. The contrasting features of these spaces become transparent through the analysis in the eigen-spinor bases of their Dirac operators.

comments
33 pages, 2 figures. v2: substantive edits, matching published version, including author list
journal
Int. J. Geom. Methods Mod. Phys. 15, No. 12 (2018), 1850204
fields
Mathematical Physics
pdf ↗ page arXiv:1608.05270 ↗

Let $n$ points be in crescent configuration in $\mathbb{R}^d$ if they lie in general position in $\mathbb{R}^d$ and determine $n-1$ distinct distances, such that for every $1 \leq i \leq n-1$ there is a distance that occurs exactly $i$ times. For the planar problem, Erdős conjectured that there exists $N$ such that no crescent configuration exists on $n$ points whenever $n\geq N$. In the same paper, he also gave constructions for $n$ up to 8, with $n=5$ due to Pomerance and $n=7,8$ due to Palásti, but nothing is yet known for $n \geq 9$. Most recently, Burt et al. proved that a crescent configuration on $n$ points exists in $\mathbb{R}^{n-2}$ for $n \geq 3$. In this paper, we classify planar crescent configurations by their complete distance-equality patterns, up to relabeling the vertices. We restrict attention to the strict distance classes defined below. There are three such classes on four vertices, all of which admit planar crescent realizations. On five vertices, there are $49$ strict classes, exactly $31$ of which admit planar crescent realizations. On six vertices, there are $6837$ strict classes, exactly six of which admit planar crescent realizations. The classification combines finite enumeration with exact algebraic verification of realizations and nonexistence. The distance-geometric criteria also apply in higher dimensions.

comments
31 pages; updates made to algorithms and 6 points configurations added
fields
Combinatorics
pdf ↗ page arXiv:1610.07836 ↗

We give an explicit characterization of all principally polarized abelian varieties $(A,Θ)$ such that there is a finite subgroup of automorphisms $G$ of $A$ that preserve the numerical class of $Θ$, and such that the quotient variety $A/G$ is smooth. We also give a complete classification of smooth quotients of Jacobians of curves.

comments
Final version, with a Corrigendum in appendix
journal
Mosc. Math. J. 22 (2022), 225-237
fields
Algebraic Geometry
pdf ↗ page arXiv:1901.07049 ↗

We analyze the expected running time of WalkSAT, a well-known local search procedure for satisfiability solving, on satisfiable instances of the k-XOR SAT problem. We obtain estimates of this expected running time by reducing the problem to a setting amenable to classical techniques from drift analysis. A crucial ingredient of this reduction is the definition of (new, explosive) hypergraph versions of interacting particle systems, notably of coalescing and annihilating random walks as well as the voter model. The use of these tools allows to show that the expected running time of WalkSAT depends on structural parameter (we call odd Cheeger drift) of the dual of the formula hypergraph.

comments
Journal of Complexity (to appear)
fields
Combinatorics, Probability
pdf ↗ page arXiv:1909.12353 ↗

In this paper we suggest a definition for a C*-algebra attached to an injective morphism of some Étale groupoid. We take into account all the peculiarities of such objects and present some interesting relations with already well-known C*-algebras from the literature.

comments
17 pages
fields
Operator Algebras
pdf ↗ page arXiv:2005.03746 ↗

A celebrated conjecture of Hardy and Littlewood provides with an asymptotic formula for the counting function of the twin primes. We give an unconditional proof of such a formula by means of a finite Ramanujan expansion of the counting function expressed in terms of the von Mangoldt function and its incomplete form. In a completely analogous way, we solve the conjugate conjecture on the representations of any even integer as the sum of two prime numbers.

comments
Lemma 2 is false
fields
General Mathematics
pdf ↗ page arXiv:2006.04547 ↗

A disc is cut by $n$ chords through a common interior point $O$, which produces $2n$ curvilinear sectors. We determine when the two sums of alternate sectors have equal area for arbitrary chord directions, which need not be equally spaced. We first derive a closed form for the area of a single sector. We then show that the answer depends on the parity of $n$. For even $n$, at fixed orientation the condition is affine in $r_0^2$, where $r_0$ is the distance from $O$ to the centre; it involves only the total angle of the odd sectors and one trigonometric moment of the directions. For odd $n$, the condition is a nonlinear alternating sum of circular-segment functions of the distances of the chords from the centre. We treat four, six and eight pieces in detail. We also hold the chord directions fixed and translate the whole pencil to an arbitrary interior point, and we characterise the direction sets that give equal sums at every such point. For four chords only equally spaced directions have this property, which gives a converse of the pizza theorem; for an odd number of chords no direction set has it. A fair two-chord cut requires the acute angle between the chords to lie within $(r_0/a)^2$ radians of a right angle, where $a$ is the radius. For equally spaced chords the results reduce to the known theorems of Goldberg and of Mabry and Deiermann.

comments
12 pages, 2 figures
fields
General Mathematics
pdf ↗ page arXiv:2110.12559 ↗

In this paper, we consider the stationary Stokes equations in an exterior domain three-dimensional under a slip boundary condition without friction. We set the problem in weighted Sobolev spaces in order to control the behavior at infinity of the solutions. In this work, we try to investigate the existence and uniqueness up to an element of a nontrivial kernel of the weak solutions related to this problem in $L^p-$theory when $p>3$. Our proofs are based on obtaining inf-sup conditions that play a fundamental role.

fields
Analysis of PDEs
pdf ↗ page arXiv:2111.05822 ↗

To address feasibility issues in model predictive control (MPC), most implementations relax state constraints by using slack variables and adding a penalty to the cost. We propose an alternative strategy: relaxing the initial state constraint with a penalty. Compared to state-of-the-art soft constrained MPC formulations, the proposed formulation has two key features: (i) input-to-state stability and bounds on the cumulative constraint violation for unbounded disturbances; (ii) close-to-optimal performance under nominal operating conditions. The idea is initially presented for open-loop asymptotically stable nonlinear systems by designing the penalty as a Lyapunov function, but we also show how to relax this condition to: i) Lyapunov stable systems; ii) stabilizable systems; and iii) utilizing an implicit characterization of the Lyapunov function. In the special case of linear systems, the proposed MPC formulation reduces to a quadratic program, and the offline design and online computational complexity are only marginally increased compared to a nominal design. Numerical examples demonstrate benefits compared to state-of-the-art soft-constrained MPC formulations.

comments
Extended version of accepted paper in Automatica. Appendix contains additional details
journal
Automatica, 196 (2027), 113336
fields
Optimization and Control
pdf ↗ page arXiv:2207.10216 ↗

We study the circular Riesz gas, a one-dimensional system of particles with power-law repulsion. For the long-range gas, we prove that the correlation between two gaps at distance $d$ decays like $d^{s-2}$, up to an arbitrarily small loss in the exponent, and that this exponent is sharp. As a consequence, the microscopic point process converges as the number of particles tends to infinity, and its limit is hyperuniform. We also prove the decay of correlations and the convergence of the microscopic point process for the hypersingular gas. The proof analyzes the Helffer-Sjöstrand equation, an elliptic system in high dimension that encodes covariances. It combines the deformation of Combes and Thomas with discrete elliptic regularity estimates and a factorization around the equally spaced configuration.

fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2209.00396 ↗

Stochastic first-order methods are widely used for large-scale optimization and machine learning, where their performance often depends on the learning rate. We study NAG-GS, a stochastic momentum method based on a semi-implicit Gauss–Seidel discretization of a Nesterov-type flow. For strongly convex quadratics, we determine the exact range of stable step sizes for fixed damping and the step size that minimizes the worst-case spectral radius. When the damping equals the strong-convexity parameter, the method recovers the accelerated asymptotic factor $(\sqrtκ-1)/(\sqrtκ+1)$, where $κ$ is the condition number. We also prove that the quadratic iteration converges linearly with dynamic damping when its limiting fixed-damping iteration is stable. For general strongly convex, possibly nonsmooth objectives, we establish an $O(1/k)$ bound on the expected objective gap of the last iterate, using a conditionally unbiased stochastic first-order oracle with bounded second moment. The result holds with dynamic damping from any $γ_0>0$. We show that NAG-GS can also be written as a Heavy-Ball iteration with explicitly determined step sizes and momentum coefficients. We test the stochastic result on nonsmooth hinge-loss classification problems, and further evaluate NAG-GS on quadratic problems, logistic regression, convolutional neural networks, RoBERTa, and Vision Transformers. The experiments show that NAG-GS can remain effective at learning rates where standard optimizers become unstable or lose accuracy.

comments
30 pages, 13 figures. Revised title and exposition; stochastic convergence with dynamic damping, Heavy-Ball equivalence, and nonsmooth numerical experiments
fields
Optimization and Control, Numerical Analysis
pdf ↗ page arXiv:2209.14937 ↗

Robust optimization models that allow decisions to adapt to newly revealed information remain computationally challenging and are often limited to two decision stages or are addressed heuristically or via approximation schemes. This challenge is further amplified when uncertainty sets depend on earlier decisions. Conventional approaches to such problems often require problem-specific reformulations and tailored algorithms, highlighting the need for an exact model-and-run approach. We address this gap by presenting an algorithmic framework for solving multistage robust discrete linear optimization problems. This exact approach adapts and integrates ideas from game-tree search, quantified conflict-driven clause learning, and integer linear programming to explore the minimax tree and is implemented in the open-source solver Yasol. The solver supports discrete decision variables across multiple stages, discrete uncertainty sets with polyhedral or decision-dependent structure, and mixed-integer recourse in the final stage. It can be used as a standalone executable or through a C++ library interface, allowing it to serve as a reference solver for exact methods in multistage robust discrete linear optimization. In an extensive computational study, we demonstrate the capabilities of our solver. The results show competitive performance against a reference approach from the literature for a protection-interdiction problem, as well as strong performance on bilevel benchmark instances. For a multistage cutting-stock problem, we demonstrate the substantial loss in solution quality that can arise from neglecting the multistage structure and show that our solver significantly outperforms decision-rule approximations in terms of computational efficiency while certifying the true multistage optimum.

fields
Optimization and Control
pdf ↗ page arXiv:2210.11132 ↗

Identifying the effects of causes and causes of effects is vital in virtually every scientific field. Often, however, the needed probabilities may not be fully identifiable from the available data sources. This paper shows how approximate identifiability is still possible for several probabilities of causation. We term this epsilon-identifiability and demonstrate its usefulness in cases where the behavior of certain subpopulations can be restricted within sufficiently narrow bounds. In particular, we show how unidentifiable causal effects and counterfactual probabilities can be epsilon-identified when such allowances are made. Often, these allowances are easily measured and reasonably assumed. Finally, epsilon-identifiability is applied to the unit selection problem.

fields
Statistics Theory
pdf ↗ page arXiv:2301.12022 ↗

We define and study the subspace of cuspidal functions for $G$-bundles (and its ramified analogs) on a class of nilpotent extensions of curves over a finite field. In the case of length $2$ extensions and of $G=\mathrm{GL}_2$, we prove that the space of cuspidal functions (for bundles with a fixed determinant) is finite-dimensional and provide bounds on its dimension. In this case we also relate cuspidality to Hecke finiteness, defined in terms of the action of some natural Hecke operators. Finally, we construct some Hecke eigenfunctions using the relation to Higgs bundles over the corresponding reduced curve.

comments
v1: 62 pages, v2: 72 pages, corrected version. Added local calculations of spherical vectors and some computations of characters of the automorphism groups of Higgs bundles associated with a square-zero extension of the curve over a finite field
fields
Number Theory, Algebraic Geometry, Representation Theory
pdf ↗ page arXiv:2303.16259 ↗

A Lorenz skeleton and Lorenz hull are the range and the convex hull of the range of an $\mathbb{R}^n$-valued measure, respectively. A Lorenz hull is also known as a zonoid. We extend the tensor product of centered zonoids introduced by Aubrun and Lancien and developed by Breiding et al. to Lorenz hulls without a centering restriction, and further to Lorenz skeletons. We show that this product admits a direct interpretation in terms of the elementary notion of a product measure. Inclusion preservation and Hausdorff continuity, shown by Aubrun and Lancien and Breiding et al., respectively, for centered zonoids, hold in this more general setting as well.

fields
Dynamical Systems, Probability
pdf ↗ page arXiv:2309.09790 ↗

We construct explicit affine uniformly Lipschitz actions on $\ell^1$ for certain groups with hyperbolic features. For acylindrically hyperbolic groups, our actions have unbounded orbits, while for residually finite hyperbolic groups and for mapping class groups, the actions have proper orbits, with the induced $\ell^1$-metric quasi-isometric (respectively, almost quasi-isometric, up to a logarithmic factor) to the word metric. This allows us to construct proper continuous affine uniformly Lipschitz actions on $L^1$, with undistorted orbits, for all simple groups of real rank one.

comments
38 pages
fields
Group Theory, Functional Analysis
pdf ↗ page arXiv:2309.12915 ↗

A central property of (Chern) topological insulators is the presence of robust asymmetric transport along interfaces separating two-dimensional insulating materials in different topological phases. A Topological Anderson Insulator is an insulator whose topological phase is induced by spatial fluctuations. This paper proposes a mathematical model of perturbed Dirac equations and shows that for sufficiently large and highly oscillatory perturbations, the systems is in a different topological phase than the unperturbed model. In particular, a robust asymmetric transport indeed appears at an interface separating perturbed and unperturbed phases. The theoretical results are based on careful estimates of resolvent operators in the homogenization theory of Dirac equations and on the characterization of topological phases by the index of an appropriate Fredholm operator.

comments
Corrects the norm-resolvent estimate (Theorem 2.6), index formulas (Sect. 2), and the gap-closing remark from the published version [Comm. Partial Differential Equations 49 (2024), 989-1010], alongside minor misprints. An official corrigendum has been submitted to the publisher
fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2311.15678 ↗

Quantum singular value transformation (QSVT) provides a unifying framework for quantum algorithms by applying the polynomials of quantum signal processing (QSP) to block-encoded matrices. We connect this framework with spectral analysis, design and topological protection in condensed matter physics. Extending QSVT beyond matrices to possibly infinite-dimensional operators (operator QSVT), we establish the following. First, it transforms every block of the operator by the QSP polynomials; no singular value decomposition is needed. Second, a spectral mapping theorem determines the full spectrum of the QSVT operator, including continuous spectra, from the embedded operator through the QSP polynomial, allowing spectral analysis without translation invariance. Third, the spectral mapping turns QSP into spectral design: the QSP phases can stretch, fold and separate the bands and place the eigenvalues. Finally, the kernel sectors that conventional QSVT usually sets aside carry a topological index, invariant under the choice of QSP phases, local perturbations and gap-preserving deformations, which guarantees topologically protected states. These results unify two quantum walks: Szegedy's walk from quantum algorithms and the split-step walk from condensed matter physics. Their kernel sectors encode the graph's Euler characteristic and the difference of bulk winding numbers, respectively. For the split-step walk, QSP breaks chiral symmetry but not the topological protection, and can move the protected state into a continuous band to form a bound state in the continuum (BIC).

comments
30 pages, 5 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2401.09007 ↗

We study a $q$-version of the chromatic polynomial of a given graph $G=(V,E)$, namely, \[ χ_G^λ(q,n) \ := \sum_{\substack{\text{proper colorings}\\ c\,:\,V\to[n]}} q^{ \sum_{ v \in V } λ_v c(v) }, \] where $λ\in \mathbb{Z}_{>0}^V$ is a fixed linear form. Via work of Chapoton (2016) on $q$-Ehrhart polynomials, $χ_G^λ(q,n)$ turns out to be a polynomial in the $q$-integer $[n]_q$, with coefficients that are rational functions in $q$. Additionally, we prove structural results for $χ_G^λ(q,n)$ and exhibit connections to neighboring concepts, e.g., chromatic symmetric functions and the arithmetic of order polytopes. We offer a strengthened version of Stanley's conjecture that the chromatic symmetric function distinguishes trees, which leads to an analogue of $P$-partitions for graphs.

comments
20 pages, 2 tables
fields
Combinatorics
pdf ↗ page arXiv:2403.19573 ↗

In [6], the authors construct a cochain complex $C\!H^*(\mathcal{Z}_{\mathcal{K}})$ on the cohomology of a moment-angle complex $\mathcal{Z}_{\mathcal{K}}$ and call the resulting cohomology the double cohomology, $H\!H^*(\mathcal{Z}_{\mathcal{K}})$. In this paper, we study the change of rank in double cohomology after gluing an $n$-simplex to a simplicial complex $\mathcal{K}$ in certain conditions. As an application, we give a positive answer to an open problem in [6]: For any even integer $r$, there always exists a simlicial complex $\mathcal{K}$ such that $\operatorname{rank} H\!H^*(\mathcal{Z}_{\mathcal{K}})=r$.

comments
12 pages
fields
Algebraic Topology
pdf ↗ page arXiv:2404.06763 ↗

In the early 2000's, R. Langlands proposed a strategy called Beyond Endoscopy to attack the principle of functoriality, which is one of the central questions of present day mathematics. A first step was achieved by A. Altug who worked with the group $GL(2,\mathbb{Q})$ in a series of three papers. We generalize the first part of this result to a class of totally real number fields. In particular, we cancel the contribution of the trivial and special representations to the trace formula using an additive Poisson summation on the regular elliptic terms.

comments
Comments welcome! This is the accepted version
fields
Number Theory, Representation Theory
pdf ↗ page arXiv:2404.10139 ↗

This paper develops an exact solution framework for the choice-based time slot management problem under mixed logit demand in attended home delivery systems. The problem jointly optimizes delivery slot offerings, price discounts, and routing decisions, with customer choices endogenously modeled through a simulation-based mixed logit formulation embedded via sample average approximation, resulting in a large-scale stochastic mixed-integer program. To address this complexity, we propose a logic-based Benders decomposition (LBBD) that separates strategic assortment and pricing decisions, together with customer choice, from scenario-specific vehicle routing subproblems. We derive problem-specific optimality cuts that exploit the routing structure to provide stronger bounds than generic cuts, and establish their validity. To enhance computational performance, we introduce and systematically evaluate several strengthening strategies, including relaxation-based cut generation and capacity- and flow-based valid inequalities. We further compare two equivalent encodings of the embedded choice model and show that the relative value of each strengthening strategy depends on the strength of the master formulation it complements. Computational experiments on benchmark instances show that the proposed framework significantly extends the range of solvable instances compared to direct MILP approaches. The framework is exact: it converges to a proven optimal solution given sufficient time, and the bounds it produces at intermediate iterations are valid. Within a 12-hour limit, proven optimality is reached for instances with up to 10 customers, and valid bounds are obtained up to 40 customers, including sizes at which the direct MILP formulation returns no feasible solution.

fields
Optimization and Control
pdf ↗ page arXiv:2407.02339 ↗

We study Hölder continuity, $p^\mathrm{th}$-variation function and Riesz variation of Weierstrass-type functions along the sequence of b-adic partitions, where b>1 is an integer. By a Weierstrass-type function, we mean that in the definition of the well-known Weierstrass function, the power function is replaced by a submultiplicative function, and the Lipschitz continuous cosine and sine functions are replaced by a general periodic Hölder continuous function.

comments
39 pages
fields
Classical Analysis and ODEs, Probability
pdf ↗ page arXiv:2407.09229 ↗

This paper shows that the wave trace of a bounded and strictly convex planar domain may be arbitrarily smooth in a neighborhood of some point in the length spectrum. In other words, the Poisson relation, which asserts that the singular support of the wave trace is contained in the closure of $\pm$ the length spectrum, can almost be made into a strict inclusion. To do so, we construct large families of domains for which there exist multiple periodic billiard orbits having the same length but different Maslov indices. Using the microlocal Balian-Bloch-Zelditch parametrix for wave invariants developed in our previous paper, we solve a large system of equations for the boundary curvature jets, which leads to the required cancellations. We call such periodic orbits silent, since they are undetectable from the ostensibly audible wave trace. Such cancellations show that there are potential limitations in using the wave trace for inverse spectral problems and, more fundamentally, that the Laplace spectrum and the length spectrum are inherently different mathematical objects, at least insofar as the wave trace is concerned.

comments
74 pages, 9 figures, to appear in Analysis & PDE
fields
Spectral Theory, Mathematical Physics, Analysis of PDEs, Differential Geometry, Dynamical Systems
pdf ↗ page arXiv:2408.09238 ↗

Given a field $K$, one may ask which finite groups are Galois groups of field extensions $L/K$ such that $L$ is a maximal subfield of a division algebra with center $K$. Such a group is called admissible over $K$. This connection between inverse Galois theory and division algebras was first explored by Schacher in the 1960s. In this manuscript we consider this problem when $K$ is a number field. For the case when $L/K$ is tamely ramified, we give a complete classification of number fields for which every solvable Sylow-metacyclic group is admissible, extending J. Sonn's result for $K = \mathbb{Q}$. For the case when $L/K$ is allowed to be wildly ramified, we give a characterization of admissible groups over several classes of number fields, and partial results in other cases.

comments
23 pages; exposition improved based on referee comments
journal
J. Algebra 690 (2026), 832-859
fields
Number Theory, Rings and Algebras
pdf ↗ page arXiv:2409.02333 ↗

You may have seen the words "topological recursion" mentioned in papers on matrix models, Hurwitz theory, Gromov-Witten theory, topological string theory, knot theory, topological field theory, JT gravity, cohomological field theory, free probability theory, gauge theories, to name a few. The goal of these lecture notes is certainly not to explain all these applications of the topological recursion framework. Rather, the intention is to provide a down-to-earth (and hopefully accessible) introduction to topological recursion itself, so that when you see these words mentioned, you can understand what it is all about. These lecture notes accompanied a series of lectures at the Les Houches school "Quantum Geometry (Mathematical Methods for Gravity, Gauge Theories and Non-Perturbative Physics)" in Summer 2024.

comments
54 pages. Comments, suggestions and corrections are welcome and encouraged! v2: minor revisions
journal
SciPost Phys. Lect. Notes 115 (2026)
fields
Mathematical Physics, Algebraic Geometry
pdf ↗ page arXiv:2409.06657 ↗

We study the \emph{Polynomial Avoidance Problem}: given a finite collection $F \subseteq Z[x_1,\dots,x_n]$ of nonzero polynomials, find a vector $x \in Z^n$ of minimum $\ell_p$-norm such that $f(x)\neq0$ for every $f \in F$. Each polynomial is given sparsely as a list of its nonzero terms. The affine special case is the \emph{Hyperplane Avoidance Problem}. This special case arises in Barvinok-type algorithms for counting integer points in polyhedra, when certain rational generating functions must be evaluated at a removable singularity. Our main results are as follows: 1) Let $D:=\sum_{f\in F} \text{deg} f$. Every instance admits a feasible vector $x$ satisfying \[ \|x\|_1 \leq \min\left\{D,\frac{D+n}{2}\right\}. \] This bound is tight. Such a vector can be found by a deterministic algorithm whose bit complexity is polynomial in $D$ and the sparse encoding length of the input. Our proof uses a new algorithmic variant of the \emph{Combinatorial Nullstellensatz}; 2) In the affine case, with $m$ input polynomials, such a vector can be constructed by a fast deterministic polynomial-time algorithm using $O(nm)$ arithmetic operations. For the homogeneous case, Barvinok's original approach provides only the weaker guarantee $\|x\|_1 = O((nm)^{n-1})$; 3) The problem is NP-hard for every $p\in R_{\geq 1}\cup\{\infty\}$, even for homogeneous linear forms; 4) As an application, we improve the complexity bound for counting integer points in general rational polytopes. We also improve the bound for polyhedra of bounded codimension; this includes the counting version of the \emph{Unbounded Subset-Sum} problem as a special case.

comments
This version is an improved version of the previous variant of the work, in which the case of a system consisting of polynomials is also considered. The overview part of the work has also been substantially expanded
fields
Combinatorics
pdf ↗ page arXiv:2411.07030 ↗

Temporal difference (TD) and Monte Carlo (MC) are two fundamental principles for value function estimation in reinforcement learning. TD is asymptotically efficient for value estimation, whereas MC requires only adequate target visitation. We show that these advantages are automatically combined by ordinary TD in the tabular setting. We identify the subgraph Bellman variance, arising from bootstrapping within a subgraph and rollouts outside it, as a fundamental quantity connecting the MC and efficient TD benchmarks. With independent trajectories, we prove a non-asymptotic oracle inequality: the mean squared error of full TD is bounded by this variance divided by the sample size for every sufficiently covered subgraph containing the target, up to a higher-order remainder whose occupancy dependence is confined to that subgraph. No subgraph is supplied to the algorithm. TD thus adapts without selecting a subgraph or requiring global coverage. A finite-sample minimax lower bound shows that the asymptotic variance of TD need not be attainable before rare states are adequately observed. Experimental studies examine the transition between these variance scales and its dependence on visitation structure.

fields
Statistics Theory
pdf ↗ page arXiv:2411.09731 ↗

We study the even and odd Jacobi heat kernels defined in the context of the multidimensional double cone and its surface, the multidimensional hyperboloid and its surface, and the multidimensional paraboloid and its surface. By integrating the framework of Jacobi polynomials on these domains, as analyzed by Xu, with contemporary methods developed by Nowak, Sjögren, and Szarek for obtaining sharp estimates of the spherical heat kernel, we establish genuinely sharp estimates for the even and odd Jacobi heat kernel in the double conic settings, and for the even Jacobi heat kernel on hyperbolic settings. We also discuss the limitations encountered in extending these results to the remaining settings.

comments
Revised version. Corrected the sharp heat kernel estimates near the antipodal regime and made several related consistency corrections
fields
Analysis of PDEs
pdf ↗ page arXiv:2411.15793 ↗

Many operations run on a budget that must rarely run out, such as a wear allowance. A wrapper keeps the remaining budget in the state, subtracts a barrier that grows as the budget shrinks and clips the combined reward from below at -m, and a standard policy-gradient learner trains unchanged. The clip level m is a price of failure risk, since the trained value splits exactly into utility before failure, the cost of a margin and the price times the risk, so whatever policy training returns earns the most such utility net of that cost among policies with no greater risk, up to its optimization error. Its risk exceeds a comparison policy's by at most order 1/m when charged costs stay bounded, and value or validation data certify it. The main theoretical result is exponential convergence of the population training flow under any parameter regularizer convex along the motions of the network's units and strong enough against the price, biased toward the prior policy. This extends mean-field convergence of policy gradients from relative entropy to weight decay, which covers continuous-time training of finite networks. Relative entropy adds noise that lets learning leave flat regions, weight decay pulls the parameters toward a reference, and action regularization bounds the risk change by the distance to a trusted rule. Experiments on an exact maintenance benchmark, a fuel-tank controller, a solvable flow and a benchmark-suite task measure the bounds' slack. We state when another method is better, without claiming this one outperforms them.

comments
38 pages, 7 figures, 2 tables; Online Resource 1 (44 pages) as an ancillary file. Revised and corrected version, replaces v2 (title changed). Code: https://github.com/pma-papers/bpg-replication
fields
Optimization and Control, Probability
pdf ↗ page arXiv:2411.19193 ↗

We extend the scope of analytic combinatorics to classes containing objects that have irrational sizes. The generating function for such a class is a power series that admits irrational exponents (which we call a Ribenboim series). A transformation then yields a generalised Dirichlet series from which the asymptotics of the coefficients can be extracted by singularity analysis using an appropriate Tauberian theorem. In practice, the asymptotics can often be determined directly from the original generating function. We illustrate the technique with a variety of applications, including tilings with tiles of irrational area, ordered integer factorizations, lattice walks enumerated by Euclidean length, and plane trees with vertices of irrational size. We also explore phase transitions in the asymptotics of families of irrational combinatorial classes. This revision contains corrections to the statement and proof of Theorem 1, to the statement of Conjecture 6, and to the proof of Proposition 13 that appear in the published version.

comments
31 pages
journal
Electron. J. Combin. 33(3): P3.15, 2026
fields
Combinatorics
pdf ↗ page arXiv:2412.14682 ↗

We introduce an $h$-Shuhan matrix as a real square matrix $H=(h_{ij})$ with a common diagonal entry $h\ge0$, nonpositive integral off-diagonal entries, and the following Cartan-type condition on each unordered pair $\{i,j\}$: either $h_{ij}=h_{ji}$, or $\max\{h_{ij},h_{ji}\}=-1$ and $\min\{h_{ij},h_{ji}\}\le -2$. These conditions imply that the scalar shift $H+(2-h)I_n$ is a generalized Cartan matrix; not every generalized Cartan matrix arises from an $h$-Shuhan matrix in this way. We classify, for $0\le h<2$, all indecomposable $h$-Shuhan matrices that are positive semidefinite, generalized positive semidefinite, or virtual positive semidefinite (the last notion means that all principal minors are nonnegative). For $h<2$, virtual positive semidefiniteness forces $H+(2-h)I_n$ to be of finite Cartan type, the positive-type case of Vinberg's trichotomy (1971). At $h=2$, we describe the boundary case in terms of finite and affine Cartan matrices. The associated determinant recurrences hold for general $h$. We study congruences for integral $h$ and asymptotic and Fibonacci-type identities for $h>2$.

comments
29 pages
fields
Rings and Algebras
pdf ↗ page arXiv:2501.01955 ↗

We characterize support $τ$-tilting modules as classical tilting modules over quotient algebras satisfying a tensor-vanishing condition, and characterize classical tilting modules as $τ$-tilting modules satisfying a ${\rm Tor}^1$-vanishing condition. We use delooping levels to study \emph{Self-orthogonal $τ$-tilting Conjecture}: any self-orthogonal $τ$-tilting module is classical tilting. We confirm the conjecture when the endomorphism algebra of the module has finite global delooping level.

comments
11 Pages,Typos and errors have been corrected
fields
Representation Theory, Rings and Algebras
pdf ↗ page arXiv:2501.02466 ↗

This paper introduces a fractal $ψ$-Fueter operator in the quaternionic context inspired by the concepts of proportional derivative and Hausdorff derivative ({fractal derivative}) of a function with respect to a fractal measure. Moreover, we establish the corresponding Stokes and Borel-Pompeiu formulas associated to this generalized fractal $ψ$-Fueter operator with respect to a vector valued function induced by truncated exponential functions.

comments
25 pages, 0 figures
fields
Complex Variables
pdf ↗ page arXiv:2502.15518 ↗

After the discovery of Khovanov homology, which categorifies the Jones polynomial, an analogous categorification of the chromatic polynomial, known as chromatic homology, was introduced. Its graded Euler characteristic recovers the chromatic polynomial. In this paper, we present a spanning tree model for the chromatic complex, i.e., we describe a chain complex generated by certain spanning trees of the graph that is chain homotopy equivalent to the chromatic complex. We employ the spanning tree model over $\mathcal{A}_m:= \frac{\mathbb{Z}[x]}{<x^m>}$ algebra to answer two open questions. First, we establish the conjecture posed by Sazdanovic and Scofield regarding the homological span of chromatic homology over $\mathcal{A}_m$ algebra, demonstrating that for any graph $G$ with $v$ vertices and $b$ blocks, the homological span is $v - b$. Additionally, we prove a conjecture of Helme-Guizon, Przytycki, and Rong concerning the existence of torsion of order dividing $m$ in chromatic homology over $\mathcal{A}_m$ algebra.

comments
Improved exposition and minor corrections
fields
Combinatorics, Quantum Algebra
pdf ↗ page arXiv:2504.00834 ↗

Let $(G,+)$ be a countable abelian group such that the subgroup $\{g+g\colon g\in G\}$ has finite index and the doubling map $g\mapsto g+g$ has finite kernel. We establish lower bounds on the upper density of a set $A\subset G$ with respect to an appropriate Følner sequence, so that $A$ contains a sumset of the form $\{t+b_1+b_2\colon b_1,b_2\in B\}$ or $\{b_1+b_2\colon b_1,b_2\in B\}$, for some infinite $B\subset G$ and some $t\in G$. Both assumptions on $G$ are necessary for our results to be true. We also characterize the Følner sequences for which this is possible. Finally, we show that our lower bounds are optimal in a strong sense.

comments
34 pages. Revised in response to the referee's comments. Appeared in International Mathematics Research Notices
fields
Dynamical Systems, Combinatorics
pdf ↗ page arXiv:2504.08649 ↗

We establish a geometric construction of Kashiwara crystals on the irreducible components of the varieties of multiparameter persistence modules. Our approach differs from the seminal work of Kashiwara and Saito, as well as subsequent related works, by emphasizing commutative relations rather than preprojective relations for a given quiver. Furthermore, we provide explicit descriptions of the Kashiwara operators in the fundamental cases of 1- and 2-parameter persistence modules, offering concrete insights into the crystal structure in these settings.

comments
Published version
fields
Representation Theory, Algebraic Geometry, Rings and Algebras
pdf ↗ page arXiv:2504.14844 ↗

We answer a question of Freedman and Krushkal, producing filling links in any closed, orientable 3-manifold. The links we construct are hyperbolic, and have large essential systole, contrasting earlier geometric constraints on hyperbolic links in 3-manifolds due to Adams-Reid and Lakeland-Leininger.

comments
Paper is more clear now thanks to referee comments; the proofs are largely unchanged. No AI use. To appear in Comment. Math. Helv
fields
Geometric Topology, Group Theory
pdf ↗ page arXiv:2505.00896 ↗

When synthesizing multi-source high-dimensional data, a key objective is to extract low-dimensional representations that effectively approximate the original features across different sources. Such representations facilitate the discovery of transferable structures and help mitigate systematic biases such as batch effects. We introduce Stable Principal Component Analysis (StablePCA), a distributionally robust framework for constructing stable latent representations by maximizing the worst-case explained variance over multiple sources. A primary challenge in extending classical PCA to the multi-source setting lies in the nonconvex rank constraint, which renders the StablePCA formulation a nonconvex optimization problem. To overcome this challenge, we conduct a convex relaxation of StablePCA and develop an efficient Mirror-Prox algorithm to solve the relaxed problem, with global convergence guarantees. Since the relaxed problem generally differs from the original formulation, we further introduce a data-dependent certificate to assess how well the algorithm solves the original nonconvex problem and establish the condition under which the relaxation is tight. Finally, we explore alternative distributionally robust formulations of multi-source PCA based on different loss functions.

comments
72 pages. Revised presentation, proofs, and numerical illustrations
fields
Optimization and Control
pdf ↗ page arXiv:2505.00940 ↗

Our GBDT (generalised Bäcklund-Darboux transformation) approach is used to construct explicit solutions of the focusing nonlinear Schrödinger (NLS) equation in the case of the exponential seed $a \exp\{2 i (cx +dt)\}$. The corresponding Baker-Akhiezer functions and evolution of the Weyl functions are obtained as well. In particular, the solutions, which appear in the study of rogue waves, step-like solutions and $N$-modulation solutions of the NLS equation are considered. This work is an essential development of our joint work with Rien Kaashoek and Israel Gohberg, where the seed was trivial, as well as several other of our previous works.

comments
This paper essentially develops some ideas from our paper arXiv:2002.04975. Some references and Remark 2.2 are added in the second version
fields
Mathematical Physics, Analysis of PDEs, Classical Analysis and ODEs
pdf ↗ page arXiv:2506.07702 ↗

We introduce the notion of doubling and r-tupling for simplicial complexes, a notion reminiscent to that of matching complexes in graph theory. We prove a connectivity result for such complexes and relate r-tupling to stabilizing r times faster in homological stability.

comments
final version. To appear in Combinatorial Theory
fields
Combinatorics, Algebraic Topology
pdf ↗ page arXiv:2506.10436 ↗

Let $P$ be a polytope defined by the system $A x \leq b$, where $A \in R^{m \times n}$, $b \in R^m$, and $\text{rank}(A) = n$. We give a short geometric proof of the following tight upper bound on the number of vertices of $P$: $$ n! \cdot \fracΔ{Δ_{\text{average}}} \cdot \text{vol}(B_2) \sim \frac{1}{\sqrt{πn}} \cdot \left(\frac{2 π}{e}\right)^{n/2} \cdot n^{n/2} \cdot \fracΔ{Δ_{\text{average}}}, $$ where $Δ$ is the maximum absolute value of $n \times n$ subdeterminants of $A$, and $Δ_{\text{average}}$ is the average absolute value of subdeterminants of $A$ corresponding to a triangulation of $P$'s normal fan. Assuming that $A$ is integer, such polyhedra are called $Δ$-modular polyhedra. Note that in the integer case, the bound can be simplified via the inequality $Δ_{\text{average}} \geq Δ_{\min} \geq 1$, where $Δ_{\min}$ is the minimum absolute value of subdeterminants of $A$ corresponding to feasible bases of $A x \leq b$. For this, we prove and use a symmetric variant of Macbeath's theorem. Additionally, we give a direct argument based on prior results in the field, showing that the graph diameter of $P$ is bounded by $O\bigl(n^3 \cdot \fracΔ{Δ_{\min}} \cdot \ln (n \fracΔ{Δ_{\min}}) \bigr)$. Thus, both characteristic of $P$ are linear in $Δ/Δ_{\min}$. From an algorithmic perspective, we demonstrate that: Given $A \in Q^{m \times n}$, $b \in Q^m$, and an initial feasible solution to $A x \leq b$, the convex hull of $P$ can be constructed in $O(n)^{n/2} \cdot m^2 \cdot \fracΔ{Δ_{\text{average}}}$ operations. For simple polyhedra, the dependence on $m$ reduces to linear; Given $A \in Z^{m \times n}$ and $b \in Q^m$, the number $|P \cap Z^n|$ can be computed in $O(n)^n \cdot \frac{Δ^4}{Δ_{\text{average}}}$ arithmetic operations.

comments
This version of the work has been substantially extended in the following way: it has been proved that the bound on the diameter is tight with respect to the parameter. A section has also been added that provides an in-depth comparison of existing parameters for polyhedra
fields
Combinatorics
pdf ↗ page arXiv:2506.12774 ↗

We study a Cahn-Hilliard model for phase separation in composite materials with multiple periodic microstructures. These are modeled by considering a highly oscillating potential. The focus of this paper is in the case where the scales of the microstructures are smaller than that of phase separation. We provide a compactness result and prove that the Γ-limit of the energy is a multiple of the perimeter. In particular, using the recently introduced unfolding operator for multiple scales, we show that the taking the limit of all of the scales together is equivalent to taking one limit at the time, starting from the smaller scale and keeping the larger fixed.

fields
Analysis of PDEs
pdf ↗ page arXiv:2506.14462 ↗

We study the regularity and well-posedness of physical solutions to the supercooled Stefan problem. Assuming only that the initial temperature is integrable, we prove that the free boundary, known to have jump discontinuities as a function of the time variable, is $C^1$ as a function of the space variable, and is $C^{\infty}$ outside of a closed, countable set, which we describe explicitly. We also prove that, as conjectured in arXiv:1902.05174, the set of positive times when a jump occurs cannot have accumulation points. In addition, we prove that short-time uniqueness of physical solutions implies global uniqueness, which allows us to obtain uniqueness for very general initial data that fall outside the scope of the current well-posedness regime. In particular, we answer two questions left open in arXiv:1811.12356, arXiv:2302.13097, regarding the global uniqueness of solutions. We proceed by deriving a weighted obstacle problem satisfied by the solutions, which we exploit to establish regularity and non-degeneracy estimates and to classify the free boundary points. We also establish a backward propagation of oscillation property, which allows us to control the occurrence of future jumps in terms of the past oscillation of the solution.

comments
To appear in Probab. Theory Related Fields. 31 pages, 2 figures. v3: Extended the local finiteness result to all singular times, including non-jump singularities
fields
Analysis of PDEs, Probability
pdf ↗ page arXiv:2506.18741 ↗

Estimation error in machine learning prognostic models can propagate into poor maintenance decisions. We study a data-driven predictive maintenance problem coupling health-condition estimation with a joint ordering-replacement model. We show that the conventional Estimate-Then-Optimize (ETO) approach yields inconsistent, suboptimal decisions for a broad class of such problems. To bridge the gap, we propose an Integrated Estimate-Optimize (IEO) framework that directly targets maintenance cost as the prognostic learning objective, bypassing surrogate statistical accuracy metrics such as Kullback–Leibler divergence or mean squared error. Theoretically, we prove out-of-sample performance guarantees of the IEO framework under certain conditions, including i.i.d. and sliding-window data-generating processes. Computationally, we develop a training procedure combining Monte Carlo dropout, central-difference gradient approximation, and an ETO warm-start to handle discrete maintenance decisions and nonlinear costs. Experiments on the four CMAPSS subsets demonstrate that, under the nominal setting, IEO reduces mean maintenance cost by 14.8–22.7% relative to ETO and by 7.2–11.8% relative to a decision-oriented baseline, while also reducing failure rates and cost variation. These improvements are obtained despite higher prediction errors. Increasing the ratio of unplanned-downtime costs to planned-downtime costs makes this tradeoff more pronounced: IEO shifts predictions further toward RUL underestimation, accompanied by greater cost savings and fewer failures. These findings support aligning prognostic learning with maintenance objectives when imperfect predictions must guide decisions under costly unplanned downtime.

fields
Optimization and Control
pdf ↗ page arXiv:2506.19698 ↗

We consider Dirichlet and Neumann problems for the heat equation in a domain containing an interior hole. The domain has a fixed outer boundary and a variable inner boundary determined by a diffeomorphism $φ$. We analyze the maps that assign to the infinite-dimensional shape parameter $φ$ the corresponding solutions, and we prove that they are smooth. Motivated by applications to inverse problems, we then compute the differentials with respect to $φ$ of the Dirichlet-to-Neumann and Neumann-to-Dirichlet maps.

fields
Analysis of PDEs
pdf ↗ page arXiv:2506.21196 ↗

We investigate a family of polynomials over finite fields of characteristic $2$ related to quadratic forms, and present a general treatment to study permutation properties of those polynomials. By determining some character sums associated with quadratic forms, we explicitly describe several classes of permutation polynomials.

fields
Number Theory
pdf ↗ page arXiv:2506.24012 ↗

This article investigates the robustness of gradient descent algorithms under perturbations. The concept of small-disturbance input-to-state stability (ISS) for discrete-time nonlinear dynamical systems is introduced, along with its Lyapunov characterization. The conventional linear Polyak-Lojasiewicz (PL) condition is then extended to a nonlinear version, and it is shown that the gradient descent algorithm is small-disturbance ISS provided the objective function satisfies the generalized nonlinear PL condition. This small-disturbance ISS property guarantees that the gradient descent algorithm converges to a small neighborhood of the optimum under sufficiently small perturbations. As a direct application of the developed framework, we demonstrate that the LQR cost satisfies the generalized nonlinear PL condition, thereby establishing that the policy gradient algorithm for LQR is small-disturbance ISS. Additionally, other popular policy gradient algorithms, including natural policy gradient and Gauss-Newton method, are also proven to be small-disturbance ISS.

comments
16 pages
fields
Optimization and Control
pdf ↗ page arXiv:2507.02131 ↗

We establish a local Li–Yau type gradient estimate for a nonlinear parabolic equation with logarithmic reaction on a forward complete Finsler metric measure manifold. The coefficients may depend on both space and time. The estimate uses a weighted Ricci curvature lower bound and explicit cutoff bounds for the divergence-form linearized Laplacian. We explain the role of this localization assumption when the reference direction varies with the solution. Applications include a local Harnack inequality for the equation with a potential and an upper bound for positive stationary solutions under a suitable cutoff exhaustion. An appendix gives a variational existence proof for positive bounded initial data, with quantitative bounds on every finite time interval, which is new as far as the authors know.

fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2507.06237 ↗

In this paper, we relate the MacDonald index of a 4d $\mathcal{N}=2$ SCFT with the Hilbert series of the arc space of the Zhu's $C_2$-algebra of the corresponding Schur VOA. Using this, we conjecture a simple formula for the MacDonald index of $(A_1,D_{2n+1})$ Argyres-Douglas theory. We perform checks of the formula against the known Schur limits and RG flows. To match the Schur limit, we prove new $q$-series identities.

comments
67 Pages, included discussion on classical freeness, corrected statement of main theorem
fields
Number Theory, Quantum Algebra
pdf ↗ page arXiv:2507.06294 ↗

We prove that, for every fixed $ε\in(0,1]$, the active-set method requires exponentially many iterations in the worst case to maximize a convex objective of the form $(x_1)^{1+ε}+\vec c^{\top}\vec x$ subject to linear constraints, regardless of the pivot rule used. Our lower bound therefore holds for objectives arbitrarily close to the linear case $ε=0$, where the active-set method coincides with the simplex method and a lower bound for all pivot rules would constitute a major breakthrough. The special case $ε=1$ yields a lower bound in convex quadratic maximization. This improves on the best previously known lower bound [IPCO 2025] requiring polynomial objectives of degree $ω(\log d)$ in dimension $d$ and firmly resolves the open question of whether a constant degree suffices. Our result is based on a novel extended formulation, recursively constructed using deformed products. Its key feature is that it projects onto a polygonal approximation of a parabola while preserving all of its exponentially many vertices. We define a convex objective with a single non-linear variable that forces the active-set method to follow the parabolic boundary of this projection, without allowing any shortcuts along chords corresponding to edges of its full-dimensional preimage.

fields
Combinatorics
pdf ↗ page arXiv:2507.16648 ↗

The main result of the present article is a proof that, over certain totally real number fields $K$, the Fermat equation $x^p+y^p+z^p=0$ has no non-trivial primitive solutions in $\mathcal{O}_K$ for sufficiently large prime exponents $p$ with $p \equiv 2 \pmod{3}$. A particular case of these fields are the maximal real subfields of the cyclotomic extensions $\mathbb{Q}(ζ_{3^n})$ for every $n$. Our strategy consists in combining the modular method with a generalization of an arithmetic result of Pomey to these fields.

comments
14 pages
fields
Number Theory
pdf ↗ page arXiv:2507.16883 ↗

We develop a geometric–analytic proof of the reverse isoperimetric inequality for compact AdS black hole horizons in Einstein gravity. At fixed thermodynamic volume, a York slice/Jacobi variation of the gravitational action shows that the spherical Schwarzschild–AdS solution is a strict local entropy maximum against generic non-spherical deformations. The finite global extension is formulated within the hypotheses of the Sherif–Dunsby (SD) rigidity theorem: compactness and spherical topology, the required intrinsic Ricci decomposition $\mathrm{Ric}_h=α\,\mathbf e\!\otimes\!\mathbf e+βN$ with $α\neqβ$, a proper conformal transformation of the hypersurface metric given by $\widetilde h_{ab}=e^{2\varphi}h_{ab}$ that preserves its intrinsic scalar curvature, $\widetilde R\equiv R[\widetilde h]=R[h]\equiv R>0$, and negative sheet expansion. Compactness and $S^3$ topology are built into the class considered here. In the static spherical sector such as Schwarzschild–AdS and RN–AdS, the SD Ricci form is naturally realized while it is an additional restriction for Kerr–AdS case. For the $S^3$ topology, a positive Yamabe type conformal class exists for which the Yamabe theorem provides a representative of positive and constant scalar-curvature. We further restrict to the scalar-curvature preserving conformal transformation required by the SD theorem. The negative sheet expansion is obtained using gravitational focusing. Within this admissible class, a non-round Kerr–AdS horizon cannot maximize the entropy, while fixed thermodynamic volume selects the corresponding round Schwarzschild–AdS comparator. The analysis thus establishes a generic local entropy maximum and a finite global result within the stated geometric rigidity class, identifying the role of gravity in reversing the ordinary isoperimetric problem.

comments
13 pages; revised manuscript with Supplemental Material included; accepted for publication in Physics Letters B
fields
Mathematical Physics, Differential Geometry
pdf ↗ page arXiv:2508.13235 ↗

We discuss the notion of submanifolds with boundary with intrinsic $C^1$ regularity in sub-Riemannian Heisenberg groups and we provide some examples. Eventually, we present a Stokes' Theorem for such submanifolds involving the integration of Rumin's differential forms in Heisenberg groups.

comments
9 pages, 6 figures; to appear in the series "Actes et Congrès" of the "Société Mathématique de France" (Proceedings of the "Winter School in Geometric Measure Theory", held in Hangzhou, China, 2025)
fields
Differential Geometry, Classical Analysis and ODEs, Metric Geometry
pdf ↗ page arXiv:2508.15634 ↗

Counting lattice points and triangulating polytopes is a prominent subject in discrete geometry, yet proving Ehrhart positivity or existence of unimodular triangulations remain of utmost difficulty in general, even for ``easy'' simplices. We study these problems for classes of s-lecture hall simplices. Inspired by a question of Olsen, we present a new natural class of sequences s for which the s-lecture hall simplices are not Ehrhart positive, by explicitly estimating a negative coefficient. Meanwhile, motivated by a conjecture of Hibi, Olsen and Tsuchiya, we extend the previously known classes of sequences s for which the s-lecture hall simplex admits a flag, regular and unimodular triangulation. The triangulations we construct are explicit.

comments
18 pages, 3 figures
fields
Combinatorics
pdf ↗ page arXiv:2508.18890 ↗

Transformers often transfer across related tasks with little fine-tuning; we study this question in noisy nonparametric regression for continuous maps $f,g:[0,1]^d\to\mathbb{R}^D$, representing the source and shifted tasks, using a single-query, single-head transformer $\mathcal{T}(\cdot)=V\operatorname{Softmax}(-α(\cdot,K))$ with $\ell^{\infty}$-based alignment scores $α$. Rather than assume that gradient-based training reaches an empirical risk minimizer, we give a model-specific two-stage procedure \textbf{certifying} source-task recovery and out-of-distribution (OOD) adaptation. With high probability, the resulting attention head recovers $f$ to arbitrary accuracy $\varepsilon>0$ in every fixed $L^p([0,1]^d)$ norm, $1\le p<\infty$, from noisy observations. For Lipschitz targets, its rate is minimax-optimal up to logarithmic factors without assuming oracle access to an asymptotic ERM. If $D=1$ and $f$ is Pfaffian than the values matrix factors into $AC$ with $A$ scaling at-most linearly $\mathcal{O}(\varepsilon^{-1})$ in the reciprocal approximation error $\varepsilon>0$. Importantly, our training procedure encodes all our identified (potentially) shared geometric structure from $f$ only into the values matrix factor $C$ and the keys matrix $K$ allowing for rapid access, and all $f$-specific structure only into $A$. Our main result concerns the OOD phase; where we identify geometric structures, which if shared by the source $f$ and OOD $g$ task; then only only the values matrix factor $A$ needs to be fine-tuned to guarantee an $L^p$-recovery of $g$. Critically, fine-tuning only $A$ is a convex problem which involving an \textit{at-most linear} number of parameters $\mathcal{O}(\varepsilon^{-1})$; dramatically cutting down on the worst-case scaling of $\mathcal{O}(\varepsilon^{-\mathbf{d}})$ unavoidable in the source-training phase.

comments
28 Pages + Proofs
fields
Numerical Analysis, Probability
pdf ↗ page arXiv:2509.00924 ↗

In this paper, we study two compactifications of general hypersurfaces defined by the vanishing of linear combinations of $d+2$ monomials in $d$-dimensional algebraic tori. We prove that the number of their $\mathbb{F}_q$-points is given by finite hypergeometric sums under certain general conditions. In the process, we introduce the notion of a gamma triple, which allows us to extend the classical definition of finite hypergeometric sums to prime powers corresponding to the cyclotomic field of definition of the associated monodromy representation. As a special case of our main results, we study the Dwork family and obtain a formula for the number of its $\mathbb{F}_q$-points. Our results generalise the work of Beukers, Cohen, and Mellit for finite hypergeometric sums defined over $\mathbb{Q}$.

comments
43 pages, 2 figures. v2: fixed a sign error in Theorems 1.4 and 1.8 (non-primitive case) and added a hypothesis to Proposition 3.14; revised proofs; references updated. The results of this paper also appear, in an earlier form, in the PhD thesis of the author. Comments are very welcome!
fields
Number Theory, Algebraic Geometry
pdf ↗ page arXiv:2509.17624 ↗

We study affine super Yangians of type $\widehat{\mathfrak{sl}}(m|n)$ over $\mathbb C[\hbar]$ for arbitrary cyclic parity sequences, with $m,n\geq2$ and $m\neq n$. We prove the equivalence of the full current and degree-zero-and-one presentations by reconstructing all current relations. A finite-type RTT calculation and a cut-diagram argument give explicit normalized quantum odd reflections, including the affine vertex. Together with even braid lifts and diagonal sign automorphisms, these maps satisfy the defining relations of an extended reflection groupoid. We compare the positive-root tensors under odd reflection and identify their difference with the reduced coproduct of the quadratic Cartan correction. Starting from the distinguished coproduct, this establishes the prescribed coassociative coproduct for every cyclic parity sequence. We specify the compatible degreewise completions needed for coproduct comparison at the affine vertex.

fields
Quantum Algebra
pdf ↗ page arXiv:2510.04221 ↗

Quantum error correction (QEC) is essential for scalable quantum computing, yet decoding errors via conventional algorithms result in limited accuracy (i.e., suppression of logical errors) and high overheads, both of which can be alleviated by inference-based decoders. To date, such machine-learning (ML) decoders lack two key properties crucial for practical fault tolerance: reliable uncertainty quantification and robust generalization to previously unseen QEC codes. To address this gap, we propose a Quantum Bayesian graph Attention decoder \textbf{(QuBA)} that enables expressive error-pattern recognition alongside calibrated uncertainty estimates. Building on QuBA, we further develop a multi-phase training framework with enhanced cross-domain robustness enabling decoding beyond the training set called Sequential Aggregate Generalization under Uncertainty \textbf{(SAGU)}. Experiments on bivariate bicycle (BB) codes and their coprime variants demonstrate that (i)QuBA consistently outperforms the classical baseline belief propagation (BP), achieving up to a \emph{two orders of magnitude} reduction in logical error rate (LER) under confident-decision bounds on the coprime BB code $[[154,6,16]]$; (ii) SAGU achieves decoding performance comparable to or even outperforming QuBA's domain-specific training approach.

fields
Information Theory
pdf ↗ page arXiv:2510.06257 ↗

We classify semisimple left module categories over the representation category of a type A quantum group whose fusion rules arise from Levi subalgebras. The classification is connected to equivariant Poisson structures on compact flag manifolds in the operator-algebraic setting, and on semisimple coadjoint orbits in the algebraic setting. We also provide an explicit construction based on the BGG categories of deformed quantum enveloping algebras, whose unitarizability corresponds to being of quotient type. Finally, we present a brief discussion of the non-quantum group case

comments
53 pages
fields
Quantum Algebra, Operator Algebras, Representation Theory
pdf ↗ page arXiv:2510.12057 ↗

We study the nonrelativistic limit of nonlinear Dirac ground states on finite noncompact metric graphs with Kirchhoff-type conditions. For a nondecreasing nonlinearity of arbitrary finite superquadratic growth, a sufficiently long pendant path guarantees ground states whose nonrelativistic limits are Schrödinger ground states. Every choice of these states is relatively compact in $H^1$ and satisfies a uniform exponential decay estimate. On general graphs, we prove a compactness alternative: loss of compactness requires at least the action of a full-line ground state. Translated line profiles show that the loss constants are sharp. The proof combines an $H^1$ estimate in terms of the action with convergence of Dirac fiber maxima to Schrödinger ray maxima. For power nonlinearities, we give an integral criterion in terms of the dimensionless pendant length and obtain the same conclusions for arbitrary edge lengths on four graph topologies.

comments
We have revised and rewritten the manuscript based on the reviewers' comments. 35 pages, comments are welcome
fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2510.20658 ↗

Using distinguished pseudo-spherical representations of covering tori constructed in a companion work and Dat's parahoric induction, we construct a distinguished family of representations for an Iwahori subgroup $\bar{I}$ in $\bar{G}$, an $n$-fold Brylinski-Deligne (BD) cover of a split connected reductive $p$-adic group $G$, under explicit hypotheses on the BD datum. No tameness assumption $\mathrm{gcd}(p,n)=1$ is imposed. The construction suggests a conjectural theory of Iwahori-types for BD covers beyond the tame case. We verify this for $2$-fold covers of almost simple simply-laced simply-connected Chevalley groups of rank at least $2$ over $2$-adic fields. In particular, for these non-tame $2$-fold covers we prove: 1) The pair $(\bar{I},σ)$ is a Bushnell-Kutzko type for the Bernstein block analogous to the block of unramified principal series in a linear reductive group. This is a non-tame analogue of the Borel-Casselman theorem. 2) The Hecke algebra $\mathcal{H}(\bar{G},\bar{I},σ^{\vee})$ has an Iwahori-Matsumoto presentation up to normalization. In types $A_{r},D_{2r+1},E_{6},E_{7}$ we obtain the complete presentation and show that $\mathcal{H}(\bar{G},\bar{I},σ^{\vee})$ is isomorphic to the Iwahori-Hecke algebra of an explicit linear group. This determines an equivalence of Bernstein blocks producing a local Shimura correspondence. 3) The representation $σ$ extends to a hyperspecial maximal compact subgroup $\bar{K}$. Among the genuine $\bar{K}$-representations, $σ$ has minimal depth with respect to the principal congruence filtration. It also yields a finite Shimura correspondence, generalizing a result of Savin.

comments
Revised to incorporate results from arXiv:2608.29548
fields
Representation Theory, Number Theory
pdf ↗ page arXiv:2510.23265 ↗

We prove that a 'pseudo-Hochster dual' of the noncommutative Balmer spectrum of a monoidal triangulated category classifies its semiprime thick tensor ideals. This extends Balmer's classification of radical thick tensor ideals to noncommutative tensor triangular geometry. To achieve this, we utilize the notion of support data for lattices and frames, under which the classification follows via Stone duality. We also give a characterization for when the noncommutative Balmer spectrum behaves as it does in tensor triangular geometry, that is, when it is a spectral space with quasi-compact opens given by complements of supports. Finally, we show that rigid centrally generated monoidal triangulated categories satisfy this property, and we answer a question posed by Negron–Pevtsova regarding classification of one-sided tensor ideals via cohomological support.

comments
33 pages. v3: Final version, to appear in AiM
fields
Category Theory, Representation Theory
pdf ↗ page arXiv:2510.23767 ↗

Much inspired by J. A. Wiśniewski's nef-value function method, we prove that in a smooth projective family over the unit disk, if the adjoint bundle of the canonical line bundle with a relatively semiample line bundle is nef on one fiber, then it remains nef on all fibers. We further extend this result to the semiampleness of the adjoint canonical line bundles. Using these, we prove the deformation invariance of any generalized plurigenera by assuming that only one fiber admits the semiample canonical line bundle and improve the first author–Xiao-Lei Liu's recent deformation rigidity of projective manifolds with semiample canonical line bundles. In particular, also by E. Viehweg–K. Zuo's result on the minimal number of singular fibers in a family and the first author–X. Liu's isotriviality result, if a projective family over $\mathbb{P}^1$ or an elliptic curve has one fiber with the big and nef (or more generally semiample) canonical line bundle, then all fibers are isomorphic to this fiber. Next, much inspired by M. Andreatta–T. Peternell's deformation theoretical approach, we prove that, in a smooth Kähler family of threefolds, if the canonical line bundle of one fiber is not nef, then none of its small deformations admits a nef canonical line bundle either. This partially confirms a problem posed by F. Campana–T. Peternell and the global stability of semiampleness of canonical line bundles of threefolds under a Kähler smooth deformation.

comments
21 pages, all comments and remarks are welcome
fields
Algebraic Geometry
pdf ↗ page arXiv:2510.23967 ↗

We aim to give a self-contained and detailed yet simplified account of the foundations of the theory of double operator integrals, in order to provide an accessible entry point to the theory. We make two new contributions to these foundations: (1) a new proof of the existence of the product of two projection-valued measures, which allows for the definition of the double operator integral for Hilbert-Schmidt operators, and (2) a variant approach to the integral projective tensor product on arbitrary (not necessarily separable) Hilbert spaces using a somewhat more explicit norm than has previously been given. We prove the Daletskii-Krein formula for strongly differentiable perturbations of a densely-defined self-adjoint operator and conclude by reviewing an application of the theory to quantum statistical mechanics.

comments
52 pages, typos corrected, introduction expanded, comments welcome
journal
J. Math. Phys. 67, 091101 (2026)
fields
Mathematical Physics, Functional Analysis, Operator Algebras
pdf ↗ page arXiv:2510.25789 ↗

We establish Lipschitz stability for both the potential and the initial conditions from a single boundary measurement in the context of a hyperbolic boundary initial value problem. In our setting, the initial conditions are allowed to differ for different potentials. Compared to the traditional B-K method, our approach does not require the time reflection step. This advantage opens a route towards applying our method to the fixed angle inverse scattering problem, which remains unresolved for the single incident wave case. To achieve our result, we impose a pointwise comparison estimate on the Laplacian of the differences of the initial conditions, together with a positivity condition on one of the initial positions. The assumption generalizes previous stability results that usually assume the difference to be zero. We propose the initial-potential problem and establish a Lipschitz stability, then propose an iterative reconstruction scheme and show its geometric convergence. The initial-potential problem serves as an attempt to relate initial boundary value problem to the scattering problem, and to explore the possibility of relaxing the positivity requirement on the initial data. We also establish a new pointwise Carleman estimate, whose proof is significantly shorter and the reasoning is much clearer than existing proofs.

comments
26 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2511.01688 ↗

Coastal distribution networks (DNs) are becoming key infrastructure for the low-carbon energy transition as electric vehicle (EV) charging, photovoltaic (PV) generation, energy storage, and marine renewable procurement are increasingly coupled at the planning stage. Existing planning models, however, usually treat network expansion, PV-storage-EV station (PSES) deployment, low-carbon operation, and coastal asset degradation separately. This paper proposes a data-driven distributionally robust planning framework for coastal DNs and PSESs. The framework jointly optimizes radial network topology and PSES siting while incorporating salt-spray degradation costs, carbon-emission-flow (CEF)-based nodal carbon accounting, and tidal-generation-driven procurement costs. To represent heterogeneous uncertainty, the model combines continuous box uncertainty for load and PV availability with probability ambiguity over EV-charging scenarios generated from historical charging profiles. The resulting two-stage distributionally robust optimization problem is reformulated through exact relaxation of ESS mode binaries and KKT-based decomposition, and is solved by an inexact column-and-constraint generation (i-C&CG) algorithm. Case studies on a realistic 47-node coastal feeder show that salt-spray exposure can alter the selected topology and PSES siting, while the proposed planning scheme improves voltage quality, reduces annualized total cost, and reveals the cost–carbon trade-off of tidal procurement. The tailored i-C&CG algorithm obtains the same optimal design as conventional C&CG with about an order-of-magnitude reduction in solution time.

fields
Optimization and Control
pdf ↗ page arXiv:2511.09321 ↗

The Low Order-Value Optimization (LOVO) problem involves minimizing the minimum among a finite number of function values within a feasible set. LOVO has several practical applications such as robust parameter estimation, protein alignment, portfolio optimization, among others. In this work, we are interested in the constrained nonlinear optimization LOVO problem of minimizing the minimum between a finite number of function values subject to a nonempty closed convex set where each function is a black-box and continuously differentiable, but the derivatives are not available. We develop the first derivative-free trust-region algorithm for constrained LOVO problems with convergence to weakly critical points. Under suitable conditions, we establish the global convergence of the algorithm and also its worst-case iteration complexity analysis. An initial open-source implementation using only linear interpolation models is developed. Extensive numerical experiments and comparison with existing alternatives show the properties and the efficiency of the proposed approach when solving LOVO problems.

fields
Optimization and Control
pdf ↗ page arXiv:2511.20783 ↗

Let $X$ be a locally finite simplicial tree, and let $Γ=\langleγ_1,γ_2\rangle\leq\mathrm{Aut}(X)$ be a discrete subgroup generated by hyperbolic automorphisms acting without inversions, whose axes have finite edge intersection. We locate explicitly, using the Euclidean algorithm, the first stopping state of rank-two Nielsen reduction and determine the exact characteristic-set geometry of its terminal Nielsen pair. Write $m_1\geq m_2>0$ for the normalized translation lengths, $l$ for the axis-overlap length, and $d=\gcd(m_1,m_2)$. If $l=0$, the initial pair satisfies the Schottky criterion. If $l>0$, Euclidean quotients and remainders partition $0<l\leq m_1+m_2-d$ into intervals locating the first stopping state and its ordered Nielsen generators. Interior points yield terminal Schottky pairs. At nonfinal right endpoints, an auxiliary translated-axis overlap $l_0$ distinguishes Schottky terminal pairs from pairs containing a nontrivial elliptic generator. At the final endpoint and for larger overlaps, a nontrivial elliptic terminal generator occurs. We compute the terminal translation lengths and the relative geometry of axes or fixed subtrees in every case. Each terminal pair is Nielsen-equivalent to the initial pair and generates $Γ$. If both terminal generators are hyperbolic, they form a Schottky pair; otherwise exactly one is nontrivial elliptic. By local finiteness and discreteness, this element has finite order, so $Γ$ contains nontrivial torsion and is not free of rank two.

comments
v2: Substantially revised, expanded, and retitled; abstract and introduction rewritten, with improved exposition, a running example, and fuller discussion of related work. The first stopping state and terminal characteristic-set geometry, including equality cases, are treated in greater detail. Weighted-tree material (Section 4 of v1) is reserved for subsequent work. 39 pages, 4 figures
fields
Group Theory
pdf ↗ page arXiv:2512.00957 ↗

We investigate a graph-theoretic problem motivated by questions in quantum computing concerning the propagation of information in quantum circuits. A graph $G$ is said to be a bounded extension of its subgraph $L$ if they share the same vertex set, and the graph distance $d_L(u, v)$ is uniformly bounded for edges $uv\in G$. Given vertices $u, v$ in $G$ and an integer $k$, the geodesic slice $S(u, v, k)$ denotes the subset of vertices $w$ lying on a geodesic in $G$ between $u$ and $v$ with $d_G(u, w) = k$. We say that $G$ has bounded geodesic slices if $|S(u, v, k)|$ is uniformly bounded over all $u, v, k$. We call a graph $L$ geodesically directable if it has a bounded extension $G$ with bounded geodesic slices. Contrary to previous expectations, we prove that $\mathbb{Z}^2$ is geodesically directable. Physically, this provides a setting in which one could devise exactly-solvable chaotic local quantum circuits with non-trivial correlation patterns on 2D Euclidean lattices. In fact, we show that any bounded extension of $\mathbb{Z}^2$ is geodesically directable. This further implies that all two-dimensional regular tilings are geodesically directable.

comments
23 pages, 5 figures; v2: minor corrections
fields
Combinatorics, Metric Geometry
pdf ↗ page arXiv:2512.03029 ↗

We prove the Lefschetz standard conjectures for 4d-dimensional projective varieties of generalized Kummer deformation type. Along the way, we prove a generalization of a theorem of Buchweitz and Flenner in the analytic setting

comments
47 pages. The contingency condition on the main theorem has been dropped. Latter sections have been reworked in light of new results. Comments welcome!
fields
Algebraic Geometry
pdf ↗ page arXiv:2512.04114 ↗

In this paper, we will characterize the spectrum of two $U(1)$--gauge Laplacians, which do not commute with each other, on the upper sheet of a two–sheet quantum hyperboloid equipped with a concrete differential calculus; while we provide a thorough and comprehensive study of the geometric structures on the related quantum spaces.

fields
Quantum Algebra
pdf ↗ page arXiv:2512.08041 ↗

We prove that stability conditions on the derived category of a product of curves of positive genus are uniquely determined by their central charge and the phase of skyscraper sheaves. As an application, we construct stability conditions on Hilbert schemes of points on certain surfaces, including some K3 surfaces of Kummer type.

comments
25 pages. Final version
journal
Geom. Topol. 30 (2026), 3485-3512
fields
Algebraic Geometry
pdf ↗ page arXiv:2512.14207 ↗

We study the distribution kernel of a Toeplitz operator associated with a classical pseudodifferential operator on a compact, embeddable, strictly pseudoconvex CR manifold. The main result consists of a formula for the values at the diagonal of the second coefficient in the expansion of the symbol of the kernel. We also establish asymptotic expansions for Toeplitz operators on the positive part of a compact not necessary strictly pseudoconvex CR orbifold under certain natural assumptions.

comments
61 pages. Revised, corrected version
fields
Complex Variables, Analysis of PDEs, Differential Geometry
pdf ↗ page arXiv:2512.16506 ↗

We give a combinatorial proof that the set of points associated to a self-dual code with no proportional columns is arithmetically Gorenstein if and only if the code is indecomposable. This answers a question asked by Tohăneanu. We do so by providing a combinatorial way to compute the dimension of the Schur square of a self-dual code through a zero-one symmetrization of a systematic generator matrix. Our approach also allows us to compute the Gorenstein defect. As a consequence, we obtain a combinatorial characterization of arithmetically Gorenstein self-associated sets of points over an algebraically closed field. We also show that, for any finite field, almost all self-dual codes of large length are indecomposable, and hence their associated sets of points are arithmetically Gorenstein.

comments
15 pages
fields
Combinatorics, Information Theory
pdf ↗ page arXiv:2512.16766 ↗

In this paper, we prove that the Euclidean distance between two independent random vectors uniformly distributed on $l_p^n$-balls ($1\leq p\leq\infty$) or on their boundaries according to the cone measure satisfies a central limit theorem as $n\to\infty$. Also, we give a compact proof of the case of the sphere, which was proved by Hammersley. Furthermore, we complement our central limit theorems by providing large deviation principles for points in the balls and on their boundaries when $1\leq p\leq\infty$. The speed is $n$ for $p\geq2$, whereas the regime $1\leq p<2$ has speed $n^{p/2}$ and an explicit rate function.

fields
Probability
pdf ↗ page arXiv:2512.24367 ↗

It is well known that limits can be computed by restricting along an initial functor, and that this often simplifies limit computation. We systematically study the algorithmic implications of this idea for diagrams indexed by a poset. We say an initial functor $F\colon C\to D$ with $C$ small is minimal if the sets of objects and morphisms of $C$ each have minimum cardinality, among the sources of all initial functors with target $D$. For $Q$ a finite poset or $Q\subseteq \mathbb N^d$ an interval (i.e., a convex, connected subposet), we describe all minimal initial functors $F\colon P\to Q$. In particular, we show that such $F$ is always a subposet inclusion, up to isomorphism. We give efficient algorithms to compute a choice of minimal initial functor. In the case that $Q\subseteq \mathbb N^d$ is an interval, we give asymptotically optimal bounds on $|\hom P|$, the number of relations in $P$ (including identities), in terms of the number $n$ of minima of $Q$: We show that $|\hom P|=Θ(n)$ for $d\leq 3$, and $|\hom P|=Θ(n^2)$ for $d>3$. We apply these results to give new bounds on the cost of computing $\lim G$ for a functor $G \colon Q\to \mathbf{Vec}$ valued in vector spaces. For $Q$ connected, we also give new bounds on the cost of computing the generalized rank of $G$ (i.e., the rank of the induced map $\lim G\to \operatorname{colim} G$), which is of interest in topological data analysis.

comments
Many expository improvements and minor fixes. 53 pages
fields
Algebraic Topology, Category Theory
pdf ↗ page arXiv:2601.00209 ↗

We study the problem of maximizing R{é}nyi entropy of order $2$ (equivalently, minimizing the index of coincidence) over the set of joint distributions with prescribed marginals. A closed-form optimizer is known under a feasibility condition on the marginals; we show that this condition is highly restrictive. We then provide an explicit construction of an optimal coupling for arbitrary marginals. Our approach characterizes the optimizer's structure and yields an iterative algorithm that terminates in finite time, returning an exact solution after at most $p-1$ updates, where $p$ is the number of rows.

fields
Information Theory, Statistics Theory
pdf ↗ page arXiv:2601.09347 ↗

We present a differential geometric framework for the motion of a non-Brownian particle in the presence of fixed obstacles in a quiescent fluid, in the deterministic Stokesian regime. While one may think (owing to the Helmholtz Minimum Dissipation Theorem) that the hydrodynamic resistance tensor could be the natural candidate for a Riemannian metric shaping the particle trajectories, we demonstrate that the trajectories driven by constant external forces are \emph{not} geodesics of this bare resistance metric. Instead, they experience a geometric drift due to the curvature induced by the presence of obstacles. By rigorously incorporating translational-rotational coupling through a modified effective resistance tensor $\tilde{\mathbf{R}} = \mathbf{R}_{\mathbf{FU}} - \mathbf{R}_{\mathbf{Fω}}\mathbf{R}_{\mathbf{Tω}}^{-1}\mathbf{R}_{\mathbf{Fω}}^T$, which accounts for the hidden dissipation savings afforded by particle rotation, we introduce a unified geometric formalism. We prove that the physical particle trajectories are exact geodesics of a conformally scaled metric, $\tilde{g}_{ij} = \mathcal{D}(\mathbf{x})\tilde{R}_{ij}$, where $\mathcal{D}$ is the instantaneous power dissipation. This framework establishes that the affine parameter along the trajectory corresponds strictly to the cumulative energy dissipated. We apply this theory to the scattering of a spherical particle by a fixed spherical obstacle, showing that the exact analytical trajectory derived previously by the author is recovered as a geodesic of the dissipation-scaled manifold.

fields
Mathematical Physics
pdf ↗ page arXiv:2601.11377 ↗

We show that if the curvature of a Cartan-Hadamard $n$-manifold is constant near a convex hypersurface $Γ$, then the total Gauss-Kronecker curvature $\mathcal{G}(Γ)$ is not less than that of any convex hypersurface nested inside $Γ$. This extends Borbély's monotonicity theorem in hyperbolic space. It follows that $\mathcal{G}(Γ)$ is bounded below by the volume of the unit sphere in Euclidean space $\mathbf{R}^n$. If $n=3$ or $Γ$ is $\mathcal{C}^{1,1}$, it suffices that the curvature be constant on $Γ$.

comments
8 pages; Improvements to the main result
fields
Differential Geometry
pdf ↗ page arXiv:2601.13280 ↗

We initiate a systematic study of unimodular random infinite ideal hyperbolic polyhedra in three-dimensional hyperbolic space, developing a geometric and probabilistic theory of their global behavior. At the heart of this theory is an intrinsic geometric characteristic, determined entirely by the dihedral angles surrounding a polyhedral face. We establish a unimodular Gauss-Bonnet identity and show that the expectation of this local geometric invariant completely determines the global conformal type: under suitable assumptions, a random ideal polyhedron is almost surely parabolic precisely when the expected characteristic vanishes, and hyperbolic precisely when it is negative. This provides a sharp geometric classification of random infinite ideal polyhedra. Building on the correspondence with ideal circle patterns, we further develop a refined Ring Lemma that yields quantitative geometric control without bounded-degree assumptions. Combining these geometric tools with unimodular probability methods, we identify the Poisson boundary of the face random walk and establish positive linear hyperbolic speed. Together, these results establish a local-to-global framework linking intrinsic dihedral geometry, conformal uniformization, and stochastic behavior at infinity for unimodular random infinite ideal hyperbolic polyhedra.

comments
We have revised the title, abstract, and introduction, and clarified several statements
fields
Probability, Combinatorics, Complex Variables, Differential Geometry, Geometric Topology
pdf ↗ page arXiv:2601.14909 ↗

We consider saddle connections on a translation surface in a hyperelliptic connected component of a stratum that do not intersect the interior of a distinguished saddle connection. For this restricted set of saddle connections, we show that it satisfies an $L (\log L)^{d-2}$ growth rate, where $d$ is the complex dimension of the hyperelliptic stratum. The upper bound holds for all translation surfaces in the hyperelliptic stratum while the lower bound holds for almost every surface in the hyperelliptic stratum. The proof of the lower bound uses horocycle renormalization.

comments
Revisions following referee report. 76 pages, 7 figures
fields
Dynamical Systems, Geometric Topology
pdf ↗ page arXiv:2601.15993 ↗

The distinguishability quantified by information measures after being processed by a private mechanism has been a useful tool in studying various statistical and operational tasks while ensuring privacy. To this end, standard data-processing inequalities and strong data-processing inequalities (SDPI) are employed. Most of the previously known and even tight characterizations of contraction of information measures, including total variation distance, hockey-stick divergences, and $f$-divergences, are applicable for $(\varepsilon,0)$-local differential private (LDP) mechanisms. In this work, we derive both linear and non-linear strong data-processing inequalities for hockey-stick divergence and $f$-divergences that are valid for all $(\varepsilon,δ)$-LDP mechanisms even when $δ\neq 0$. Our results either generalize or improve the previously known bounds on the contraction of these distinguishability measures.

comments
6 pages, 3 figures; classical results from the paper arXiv:2512.16778 [quant-ph] that studies related quantum results; v2: Accepted to Information Theory Workshops (ITW-2026)
fields
Information Theory
pdf ↗ page arXiv:2601.16845 ↗

We present the interpolating conformal algebra between the instant form dynamics (IFD) and the light-front dynamics (LFD) in $(1+1)$ dimensions, along with a $4\times4$ interpolating projective spacetime matrix representation. While there are six generators in the $(1+1)$ dimensional conformal algebra, the number of kinematic and dynamic generators dramatically changes in LFD, maximizing (minimizing) the number of kinematic (dynamic) generators to four (two) with respect to two (four) kinematic (dynamic) generators in IFD, as well as in any other forms of dynamics between IFD and LFD. It confirms and signifies the utility of LFD, saving substantial dynamical efforts in solving the $(1+1)$ dimensional quantum field theories. We also present $2\times2$ Pauli matrix representation of $(1+0)$ and $(0+1)$ conformal groups, and creation/annihilation operators of quantum simple harmonic oscillator representations of $(1+0)$ dimensional conformal groups.

comments
14 pages, 9 tables
journal
Phys. Rev. D 113, 096018 (2026)
fields
Mathematical Physics
pdf ↗ page arXiv:2601.19251 ↗

We analyze the finite-size corrections to the crosscap overlap in the two-dimensional classical Ising model along its self-dual critical line. Using a fermionic formulation, we express the lattice crosscap overlap in terms of Bogoliubov angles and develop a contour-integral approach by analytically continuing the lattice momentum to the complex plane. This leads to a remarkably simple expression for the crosscap overlap, which demonstrates that the finite-size corrections decay exponentially with system size. We further derive an exact analytical formula for the corresponding decay constant and show that it is determined by the complex singularity structure of the Bogoliubov angle.

comments
6+5 pages, 3+1 figures, published version
journal
Phys. Rev. E 114, L042110 (2026)
fields
Mathematical Physics
pdf ↗ page arXiv:2601.21502 ↗

For Euclidean domains whose boundaries are sufficiently close to a sphere in $C^2$, we bound the deficit of the Escobar constant from below by the squared $H^{3/2}$ norm of the boundary graph, after fixing its constant and first spherical harmonics. This holds in every dimension $n\ge3$ and implies a bound in Fraenkel asymmetry with optimal quadratic power. Under fixed $C^{6,α}$ and size bounds, the asymmetry estimate also holds for star-shaped domains in dimensions $n\ge6$. For compact Riemannian manifolds with minimal boundary and uniform geometric bounds, we prove $S_\ast-C^*_{\text{Esc}}(M,g)\ge ca^4/\log(Λ/a)$ in dimension six and $S_\ast-C^*_{\text{Esc}}(M,g)\ge ca^4$ in higher dimensions, where $a=\|\mathring{\mathrm{II}}\|_{L^\infty}>0$. A family satisfying the same bounds has deficit comparable to $a^{8/3}$, excluding uniform lower bounds with smaller powers. We also compute the cubic coefficient of the standard boundary bubble quotient and give uniform remainder estimates.

comments
52 pages. Split from arXiv:2601.22665v2 (Part 1/3), with new results on the geometric stability of the Escobar functional. Other parts forthcoming. Comments welcome!
fields
Differential Geometry, Analysis of PDEs, Classical Analysis and ODEs
pdf ↗ page arXiv:2601.22665 ↗

Learning a common policy from multiple source environments requires accounting for cross-source heterogeneity, and the offline setting introduces an additional challenge of limited data coverage. We formulate this problem as a distributionally robust Markov decision process under a shared linear MDP representation with source-specific transition and reward components. The uncertainty set is constructed through feature-wise mixtures of these components, yielding a $d$-rectangular structure that captures cross-source heterogeneity and admits a robust Bellman recursion. Building on this formulation, we develop Group-Robust Pessimistic Value Iteration (GR-PEVI), an offline policy-learning algorithm that combines feature-wise worst-case aggregation with a data-dependent pessimism penalty. Under a robust partial coverage condition, we establish finite-sample suboptimality guarantees for GR-PEVI and extend the analysis to clustered sources, characterizing the trade-off between the statistical gain from pooling similar sources and the approximation error induced by within-cluster heterogeneity. Simulation studies demonstrate the effectiveness of GR-PEVI, with empirical behavior consistent with the theoretical analysis. We apply GR-PEVI to a multi-source electronic health record study of multiple sclerosis treatment decisions to illustrate its use in a clinical setting.

fields
Optimization and Control
pdf ↗ page arXiv:2602.01825 ↗

Efron et al. (2004) introduced least angle regression (LAR) as an algorithm for linear predictions, intended as an alternative to forward selection with connections to penalized regression. However, LAR has remained somewhat of a "black box," where some basic behavioral properties of LAR output are not well understood, including an appropriate termination point for the algorithm. We provide a novel framework for inference with LAR, which also allows LAR to be understood from new perspectives with several newly developed mathematical properties. The LAR algorithm at a data level can viewed as estimating a population counterpart "path" that organizes a response mean along regressor variables which are ordered according to a decreasing series of population "correlation" parameters; such parameters are shown to have meaningful interpretations for explaining variable contributions whereby zero correlations denote unimportant variables. In the output of LAR, estimates of all non-zero population correlations turn out to have independent normal distributions for use in inference, while estimates of zero-valued population correlations have a certain non-normal joint distribution. These properties help to provide a formal rule for stopping the LAR algorithm. While the standard bootstrap for regression can fail for LAR, a modified bootstrap provides a practical and formally justified tool for interpreting the entrance of variables and quantifying uncertainty in estimation. The LAR inference method is studied through simulation and illustrated with data examples.

comments
57 pages, 9 figures
fields
Statistics Theory
pdf ↗ page arXiv:2602.02491 ↗

We study the time evolution of Bose–Einstein condensates with three-body interactions in the Gross–Pitaevskii regime. We show that Bose–Einstein condensation is preserved under many-body evolution and that the condensate wavefunction evolves according to the quintic Gross–Pitaevskii equation in $\mathbb{R}^3$, which is energy critical. In particular, we show that the effective coupling constant is universal and depends only on a three-body scattering hypervolume.

comments
56 pages, 2 figures, comments welcome!
fields
Mathematical Physics
pdf ↗ page arXiv:2602.06481 ↗

These are notes from the lectures I gave at the Oberwolfach seminar `Tensor Triangular Geometry and Interactions' which was held in October 2025. The aim of these notes is to give an introduction to tensor triangular geometry, for both small and large categories, through the lens of lattice theory. We do not try to be exhaustive and this is reflected in both the content and the bibliography. For instance we are quite light on triangulated preliminaries, especially for compactly generated categories. The first three sections treat the essentially small case and conclude with a tensor triangular proof of Thomason's theorem computing the spectrum of the perfect complexes on a quasi-compact and quasi-separated scheme. The last section treats the compactly generated case. This final section is somewhat experimental and contains some new thoughts.

comments
This is a preview of notes to appear in the Oberwolfach Seminars series published by Birkhäuser
fields
Category Theory, Algebraic Geometry, Algebraic Topology, Representation Theory
pdf ↗ page arXiv:2602.08480 ↗

We propose an optimization-informed deep neural network approach, named iUzawa-Net, aiming for the first solver that enables real-time solutions for a class of nonsmooth optimal control problems of linear partial differential equations (PDEs). The iUzawa-Net unrolls an inexact Uzawa method for saddle point problems, replacing classical preconditioners and PDE solvers with specifically designed learnable neural networks. We prove universal approximation properties and establish the asymptotic $\varepsilon$-optimality for the iUzawa-Net, and validate its promising numerical efficiency through nonsmooth elliptic and parabolic optimal control problems. Our techniques offer a versatile framework for designing and analyzing various optimization-informed deep learning approaches to optimal control and other PDE-constrained optimization problems. The proposed learning-to-control approach synergizes model-based optimization algorithms and data-driven deep learning techniques, inheriting the merits of both methodologies.

comments
Strengthened theoretical results under weaker regularity assumptions via a refined analysis framework; expanded numerical experiments
fields
Optimization and Control, Numerical Analysis
pdf ↗ page arXiv:2602.12273 ↗

We introduce new families of quandles associated with closed orientable surfaces that can distinguish surfaces up to homeomorphism. These quandles generalize the classical Dehn quandle and are defined, respectively, on isotopy classes of unoriented essential closed curves and on integral weighted multicurves. We establish their fundamental algebraic properties and construct a natural quandle covering relating the two families. We further investigate their large-scale geometry and prove that they are unbounded with respect to the quandle metric. We determine their second bounded quandle cohomology and show that its dimension equals the cardinality of the continuum. We also establish a quandle-theoretic analogue of the Gromov Mapping Theorem, showing that the natural projection from an abelian quandle extension onto its base quandle induces an injection on bounded quandle cohomology in every degree. Finally, motivated by recent developments in quandle algebras, we study idempotents in the integral quandle algebras associated with the classical Dehn quandle. In particular, we prove that the integral quandle algebra of the Dehn quandle of the torus admits no non-trivial idempotents.

comments
31 pages, 8 figures, comments are welcome
fields
Geometric Topology, Group Theory, Quantum Algebra
pdf ↗ page arXiv:2602.17661 ↗

Molecular communication (MC) enables information exchange in nanoscale sensor networks operating in biological environments, yet privacy remains largely unaddressed. We integrate local differential privacy (LDP) into diffusion-based MC by privatizing each user's measurement at the transmitter and conveying the resulting randomized report over the MC channel. To our knowledge, this is the first systematic LDP implementation for diffusion-based MC, enabling privacy-preserving aggregate data analysis for in-body health monitoring and other population-scale sensing applications. We benchmark major LDP mechanisms under a realistic channel model. Simulation results show that Optimized Local Hashing (OLH) achieves the lowest average $\ell_1$ distribution-estimation error under the MC channel when channel resources are sufficient and the number of possible user values (alphabet size) $k$ is moderate to large, whereas k-ary Randomized Response (KRR) and, for privacy budget $ε=1$, Binary Local Hashing (BLH), which have shorter reports, are more robust as the MC transmission quality deteriorates. We further propose RLIM-LDP, which combines run-length-limited ISI-mitigation (RLIM) coding with LDP coding. Extensive simulation results demonstrate that RLIM-LDP improves end-to-end reliability and reduces the final distribution-estimation error when time and molecule resources are limited.

comments
8 pages, 4 figures
fields
Information Theory
pdf ↗ page arXiv:2603.00690 ↗

We propose neural network operator inference (NN-OpInf): a structure-preserving, composable, and minimally restrictive operator inference framework for the non-intrusive reduced-order modeling of dynamical systems. The approach learns latent dynamics from snapshot data, enforcing local operator structure such as skew-symmetry, (semi-)positive definiteness, and gradient preservation, while also reflecting complex dynamics by supporting additive compositions of heterogeneous operators. We present practical training strategies and analyze computational costs relative to linear and quadratic polynomial OpInf (P-OpInf). Numerical experiments across several nonlinear and parametric problems demonstrate improved accuracy, stability, and robustness over P-OpInf and prior NN-ROM formulations, particularly when the dynamics are not well represented by polynomial models. These results suggest that NN-OpInf can serve as an effective drop-in replacement for P-OpInf when the dynamics to be modeled contain non-polynomial nonlinearities, offering potential gains in accuracy and out-of-distribution performance at the expense of higher training computational costs and a more difficult, non-convex learning problem.

fields
Dynamical Systems
pdf ↗ page arXiv:2603.08488 ↗

It remains unknown if every prismatoid has a nonoverlapping edge-unfolding, a special case of the long-unsolved "Dürer's problem." Recently nested prismatoids have been settled [Rad24] by mixing (in some sense) the two natural unfoldings, petal-unfolding and band-unfolding. Band-unfolding fails due to a specific counterexample [O'R13b]. The main contribution of this paper is a characterization when a band-unfolding of a nested prismatoid does in fact result in a nonoverlapping unfolding. In particular, we show that the mentioned counterexample is in a sense the only possible counterexample. Although this result does not expand the class of shapes known to have an edge-unfolding, its proof expands our understanding in several ways, developing tools that may help resolve the non-nested case.

comments
Sadegh Merati found a counterexample to Lemma 8. Until I revise the argument, it is safest to withdraw
fields
Metric Geometry
pdf ↗ page arXiv:2603.09813 ↗

We develop a mixed-integer nonlinear programming (MINLP) approach for the Heilbronn triangle problem on the unit square. The formulation exploits the boundary structure of optimal configurations, which we establish rigorously, to fix coordinates and orientation variables a priori. With this formulation, a general-purpose global solver solves the case $n=9$ to global optimality within its numerical tolerances in about eight minutes on a laptop and without any precomputation, compared with about 23 hours of total computation for the best previous approach. The same model also solves $n=10$ for the first time, in under seven hours, and confirms the configuration of Comellas and Yebra (2002). The numerical solutions then guide an exact symbolic reconstruction, and the minimum triangle area of every reconstructed configuration is verified in exact arithmetic. This yields exact coordinates for all $n\le 10$, including the configurations of Comellas and Yebra (2002) for $n=9$ and $n=10$, together with rigorous lower bounds. Finally, we derive structural properties of optimal configurations, formulate a conjecture on the number of critical triangles, and pose an open question. All code and data are publicly available.

comments
31 pages, 9 figures, 16 tables. Revised version, submitted to Mathematical Programming Computation. v3: new title, n=10 solved, revised status of results. Code and data: https://github.com/spiralulam/heilbronn
fields
Optimization and Control, Combinatorics
pdf ↗ page arXiv:2603.11107 ↗

The Cauchy problem for the two-dimensional incompressible Euler equation is globally well-posed for smooth initial data. We prove that, for all $p>2$ and for a dense $G_δ$ set of initial vorticities in $W^{1,p}$, the solutions lose regularity in infinite time, in the sense that the $L^p$ norm of the gradient of the vorticity is unbounded. This confirms a long-standing conjecture of Yudovich.

comments
This new version extends the previous result: we now prove the full Yudovich conjecture. It avoids some issues in Sections 8 and 10, simplifies the proof, and significantly shortens the paper
fields
Analysis of PDEs
pdf ↗ page arXiv:2603.13079 ↗

Given random walk on a graph, the corresponding discrete-time quantum walk can be constructed using the method proposed by Szegedy. On the other hand, given a partition of the set of states of a Markov chain, one can study the corresponding aggregated process. We extend the aggregation technique to the level of quantum Markov chains. We provide conditions under which application of these two operations - Szegedy's quantization and aggregation - give the same result. These consist of the strong lumpability of the underlying Markov chain supplemented by nonlinear relations, generalizing Kolmogorov's cycle condition, that account for the coefficients of both the original and the aggregated transition matrices. In particular, we show that the conditions are satisfied in the case of the random walk on graphs equipped with equitable partitions. We present several examples, which include the classical/quantum walks on Platonic solids. We discuss also relation of discrete-time classical/quantum walks on $N$-dimensional hypercube and the Ehrenfests urn model with $N$ particles. We apply our technique for of discrete-time walks on Cayley graphs of free groups. We also compare our results with those obtained using Cantero-Moral-Velazquez uniformization of unitary matrices.

comments
22 pages, 10 figures; v2 presentation improved; v3 formulation of Proposition 3.1 corrected, while the main idea of the work remains unchanged
fields
Mathematical Physics, Combinatorics, Dynamical Systems
pdf ↗ page arXiv:2603.14269 ↗

This note documents the specification of normal forms in cubical type theory. The definition is already present in the proof of normalization for cubical type theory, but we present it in a more traditional style explicitly for reference.

comments
22 pages; corrected minor errors
fields
Logic
pdf ↗ page arXiv:2603.24923 ↗

First, we consider general Brylinski–Deligne covers of the $p$-adic general linear groups, and discuss the theory of Bernstein–Zelevinsky derivatives. We also recall the Zelevinsky-type classification of the irreducible genuine spectrum for the Kazhdan–Patterson and Savin covers. Following this, for these two special families of covers, we determine the wavefront sets of their irreducible genuine representations, expressed in terms of the iterated degrees of the highest Bernstein–Zelevinsky derivatives. Finally, for Kazhdan–Patterson covers, we reinterpret this result on the wavefront set using a version of the local Langlands correspondence and the covering Barbasch–Vogan duality.

comments
32 pages. Final version accepted by IMRN
fields
Representation Theory
pdf ↗ page arXiv:2604.01585 ↗

We present Conditional Wasserstein Autoencoders (CWAEs), a framework for conditional simulation that exploits low-dimensional structure in both the conditioned and the conditioning variables. The key idea is to modify a Wasserstein autoencoder to use a (block-) triangular decoder and impose an appropriate independence assumption on the latent variables. We show that the resulting model gives an autoencoder that can exploit low-dimensional structure while simultaneously the decoder can be used for conditional simulation. We explore various theoretical properties of CWAEs, including their connections to conditional optimal transport (OT) problems. We also present alternative formulations that lead to three architectural variants forming the foundation of our algorithms. We present a series of numerical experiments that demonstrate that our different CWAE variants achieve substantial reductions in approximation error relative to the low-rank ensemble Kalman filter (LREnKF), particularly in problems where the support of the conditional measures is truly low-dimensional.

comments
Accepted to the 65th IEEE Conference on Decision and Control (CDC 2026). 8 pages, 5 figures
fields
Optimization and Control
pdf ↗ page arXiv:2604.02644 ↗

We study $\GA$-Lie foliations on compact five-manifolds. We establish a structural classification of the holonomy group $Γ$ into five types according to its algebraic structure inside $\GA$. For three of these types (I, II, IIIa) we prove homogeneity via Meigniez's theorem and the Matsumoto–Tsuchiya theorem. For Type~IIIb we prove non-homogeneity by exhibiting a non-polycyclic holonomy group, extending the Dathe–Ndiaye construction \cite{DN11}. We introduce a cohomological class $[F]\in H^2(Γ^{\mathrm{ab}},[Γ,Γ])$ as a \emph{necessary} obstruction to homogeneity: if $\mathcal{F}$ is homogeneous and $Γ$ is polycyclic then $[F]=0$. We show that Type IIIb foliations satisfy $[F]\neq 0$. For Type IIIc, we prove that certain foliations are non-homogeneous by a direct argument when the generating translations are $\Q(λ)$-linearly independent; we exhibit explicit homogeneous examples showing that this type is \emph{not} universally non-homogeneous. Finally, we construct the first known non-homogeneous $\mathrm{Sol}$-Lie foliation in dimension five by a verified lattice deformation.

fields
Dynamical Systems
pdf ↗ page arXiv:2604.04965 ↗

We study one-parameter conifold degenerations whose central fiber \(X_0\) has finitely many ordinary double points and construct a mixed-Hodge-module refinement of the corrected perverse variation-cone object associated with the degeneration. At each node \(p_k\), the local \(\can^H/\Var^H/N\) geometry determines a rank-one point-supported interface block \(W_k^H\simeq i_{k*}\Q^H_{\{p_k\}}(-1)\). These blocks assemble canonically into \(Q_Σ^H=\bigoplus_{k=1}^r W_k^H\), through which the node-supported monodromy factors. We prove that the perverse-normalized cone \(\Cone(\Var^H)\) is concentrated in the mixed-Hodge-module heart, thereby defining \(\mathcal P^H\in MHM(X_0)\), and that \(\rat(\mathcal P^H)\simeq\mathcal P\). In the forward Yoneda orientation, the point-supported quotient is instead \(R_Σ^H\simeq\bigoplus_{k=1}^r i_{k*}\Q^H_{\{p_k\}}(-2)\), exhibiting a Hodge-theoretic distinction invisible after rational realization. We further prove a canonical nodewise decomposition of the rational perverse extension space and construct a morphism from the formal local Hodge sector to the independently defined limiting mixed Hodge structure. Its image records the node contributions that survive in global limiting cohomology, allowing nontrivial global relations among vanishing cycles.

comments
Strengthened the local and global mixed-Hodge-module constructions, clarified Tate normalization and Yoneda orientations, expanded the nodewise Ext and LMHS comparisons, added technical appendices, and revised the exposition, citations, notation, and figures throughout
fields
General Mathematics
pdf ↗ page arXiv:2604.05367 ↗

In this paper, we propose a unified compression algorithm for distributed nonconvex opitmization with both the locally- and globally-bounded communication compressors, including 1-bit compressors, saturating quantizers, and the globally-bounded compressors with both relative and absolute compression errors, as well as additional arbitrary bounded noise. We provide a rigorous convergence analysis in nonconvex settings and establish linear convergence under the Polyak-Lojasiewicz (P-L) condition. Notably, we establish an $\mathcal{O}(1/\sqrt{T})$ convergence rate for the locally-bounded class in the distributed nonconvex setting, matching that achieved by the centralized algorithms with 1-bit compressors, where $T$ denotes the total number of iterations. Moreover, one initial uncompressed communication round further yields an order-wise improvement to $\mathcal{O}(1/T^{2/3})$. For the P-L setting and the globally-bounded class, we recover state-of-the-art convergence rates.

fields
Optimization and Control
pdf ↗ page arXiv:2604.10615 ↗

We study dimensions of sets projected to an $(n-2)$-dimensional family of hyperplanes in $\mathbb{R}^n$ under curvature conditions. Let $n\ge 4$, $Σ\subset S^{n-1}$ be an $(n-2)$-dimensional $C^2$ manifold such that $Σ$ has sectional curvature $>1$ and $Z \subset \mathbb{R}^{n}$ be analytic. If $\dimπ_{T_yS^{n-1}}(Z) \le n-2$ for some $y \in S^{n-1}$, then $\dim π_{T_xS^{n-1}}(Z) = \dim Z$ for $\mathcal{H}^{n-2}$-a.e. $x \in Σ$.

comments
27 pages, 2 figures
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2604.14662 ↗

Invariant manifolds and foliations associated to invariant pseudo-hyperbolic reference geometries have been well-studied in the context of dynamics implemented by a diffeomorphism. In this paper, we provide sufficient conditions for the existence of maximally smooth, \emph{locally} invariant manifolds and foliations containing a given \emph{locally} invariant, pseudo-hyperbolic reference geometry, in the context of dynamics implemented by a smooth map which need not be invertible. The theory is motivated by applications to the developing theory of large-step optimisation.

comments
Entirely rewritten since previous version. New version includes more results and more efficient proofs
fields
Dynamical Systems
pdf ↗ page arXiv:2604.18202 ↗

For an integer $n\geq 1$, consider the $n$-th metallic number $φ_n=\frac{n+\sqrt{n^2+4}}{2}$ (e.g. $φ_1$ is the golden number) and denote by $[φ_n]_q$ its $q$-deformation in the sense of S. Morier-Genoud and V. Ovsienko. This is an algebraic continued fraction which admits an expansion into a power series $[φ_n]_q =\sum_{l=0}^{+\infty} κ_l(φ_n) q^l$ around $q=0$, with integral coefficients. By using techniques from analytic combinatorics, we establish several properties of the sequence $( κ_l(φ_n))_{l\geq 0}$ of Taylor coefficients: characterisation by recurrences or by differential equations, closed-form expressions when $n=1,2,3$, and asymptotics. We also present some remarkable identities induced by the action of the modular group $PSL(2,Z)$ and address, mainly through computer experimentations, the question of the logarithmic behaviour of the sequence $( κ_l(φ_n))_{l\geq 0}$. A particular accent is put on the comparison between the $q$-deformation $[φ_1]_q$ of the golden ratio and RNA secondary structures, the former being actually a signed version of the latter. By doing so, we would be pleased to bring the interest of combinatoricians to the newly discovered world of $q$-numbers.

comments
34 pages. Second and final version (minor corrections) to appear in Arnold Math. J
fields
Combinatorics
pdf ↗ page arXiv:2604.19898 ↗

Learning the underlying potential energy of stochastic gradient systems from partial and noisy observations is a fundamental problem arising in physics, chemistry, and data-driven modeling. Classical approaches often rely on direct regression of governing equations or velocity fields, which can be sensitive to noise and external perturbations and may fail when observations are incomplete. In this work, we propose a structure-aware, energy-based learning framework for inferring unknown potential functions in generalized diffusion processes, grounded in the energetic variational approach. Starting from the energy-dissipation law associated with the Fokker-Planck equation, we construct loss functions based on the De Giorgi dissipation functional, which consistently couple the free energy and the dissipation mechanism of the system. This formulation avoids explicit enforcement of the governing partial differential equation and preserves the underlying variational structure of the dynamics. Through numerical experiments in one, two, and three dimensions, we demonstrate that the proposed energy-based loss exhibits enhanced robustness with respect to observation time, noise level, and the diversity and amount of available training data. These results highlight the effectiveness of energy-dissipation principles as a reliable foundation for learning stochastic diffusion dynamics from data.

fields
Dynamical Systems
pdf ↗ page arXiv:2604.20188 ↗

We address a short-wave asymptotic for one class of quasi-linear second-order PDE systems involving the cross-diffusion described by the so-called Patlak-Keller-Segel law. It is common to employ these equations for modeling the predator-prey community with the prey-taxis, which means the interactions of two species of particles or cells or anything else through which the species called "predators" is capable of moving directionally while searching for the other species called "prey." However, we suppose the predators to be sensitive not to the prey density but to a cue produced by the prey. Additionally, the production of the cue is assumed to be sensitive to the intensity of an external field, which is independent of the community state. This is what we call the external signal. It can be due to the spatiotemporal inhomogeneity of the environment arising from natural or artificial reasons. We assume that the external signal takes a general short-wave form and construct a complete formal asymptotic expansion for the short-wave solutions (modulo solving a chain of the so-called slow systems) with no restrictions on the spatial dimension or kinetics of inter- or intra-specific reactions. Further, we apply the short-wave asymptotic to studying the stability or instability induced by the external signal following Kapitza's theory for the upside-down pendulum. Applying the general results to some special classes of external signals, we get examples of suppressing the taxical transport, examples of the robustness of the species equilibrium to the signal up to a very strong stabilization or, oppositely, destabilization, and something like blurring the borderline in the parametric space between the areas of stability and instability of this equilibrium.

comments
34 pages, 2 figures
fields
Analysis of PDEs
pdf ↗ page arXiv:2604.20469 ↗

We prove that any closed, convex hypersurface in an $(n+1)$-dimensional Riemannian manifold with $\lceil \frac{n}{2} \rceil$-positive curvature operator is a rational homology sphere with finite fundamental group. The same conclusion holds for any $\lceil \frac{n}{2} \rceil$-convex hypersurface, provided that the mean curvature satisfies a sharp pinching condition. Both results follow from more general vanishing and estimation theorems for the Betti numbers of closed $q$-convex immersed hypersurfaces in $(n+1)$-dimensional Riemannian manifolds, under a lower bound on the average of the smallest $(n-p)$ eigenvalues of the curvature operator.

fields
Differential Geometry
pdf ↗ page arXiv:2604.20695 ↗

A parametric theory of statistical inference is developed for the moderate deviation probability zone. The new approach to the proofs is based on the Taylor series expansion of the logarithm of the likelihood ratio based on the Hellinger distance. The Large Deviation Principle in the moderate deviation probability zone is proven for Bayesian estimators and maximum likelihood estimators. A uniform approximation of the logarithm of the likelihood ratio and Theorem on concentration of the posterior Bayesian measure are also established for the zone of moderate deviation probabilities.

comments
22 pages
fields
Statistics Theory
pdf ↗ page arXiv:2604.24736 ↗

We study the problem of minimizing a multivariate polynomial function over the unit hypercube. Exploiting sparsity in the interaction graph or hypergraph, we identify variables that can be restricted to binary values and eliminate the remaining continuous variables component-wise, reducing the problem to structured binary polynomial optimization. For quadratic objectives, we obtain polynomial-time solvability in the standard Turing model under conditions on treewidth and a nonconvexity measure of the continuous components. For higher-degree objectives, we obtain a polynomial-time algorithm in the unit-cost algebraic RAM model under logarithmic incidence treewidth and small, weakly coupled continuous components.

fields
Optimization and Control
pdf ↗ page arXiv:2604.25033 ↗

Kullback-Leibler (KL) regularization is widely used in offline decision-making and offers several benefits, motivating recent work on the sample complexity of offline learning with respect to KL-regularized performance metrics. Nevertheless, the exact sample complexity of KL-regularized offline learning remains largely from fully characterized. In this paper, we study this question in the setting of multi-armed bandits (MABs). We provide a sharp analysis of KL-PCB (Zhao et al., 2026), showing that it achieves a sample complexity of $\tilde{O}(ηSAC^{π^*}/ε)$ under large regularization $η= \tilde{O}(ε^{-1})$, and a sample complexity of $\tildeΩ(SAC^{π^*}/ε^2)$ under small regularization $η= \tildeΩ(ε^{-1})$, where $η$ is the regularization parameter, $S$ is the number of contexts, $A$ is the number of arms, $C^{π^*}$ policy coverage coefficient at the optimal policy $π^*$, $ε$ is the desired sub-optimality, and $\tilde{O}$ and $\tildeΩ$ hide all poly-logarithmic factors. We further provide a pair of sharper sample complexity lower bounds, which matches the upper bounds over the entire range of regularization strengths. Overall, our results provide a nearly complete characterization of offline multi-armed bandits with KL regularization.

fields
Statistics Theory
pdf ↗ page arXiv:2605.02141 ↗

In this article, we prove that viscosity subsolutions to nonlocal mean curvature-type equations satisfy universal volumetric estimates at all scales provided their boundary has zero Lebesgue measure. Our results hold for a general class of symmetric kernels including kernels that are comparable to the fractional Laplacian. Furthermore, we prove that subsolutions with low density (with respect to a universal constant) necessarily have `fat boundary', that is, have topological boundary with positive Lebesgue measure.

fields
Analysis of PDEs
pdf ↗ page arXiv:2605.02176 ↗

In training a neural network with gradient descent (GD), each iteration induces a linear operator that governs the first-order updates to a model's internal state variables. We call this operator the Global Empirical Neural Tangent Kernel (GeNTK). In finite-width networks, GeNTK is typically intractable to form, leading prior work to focus on restrictive settings, such as tracking only outputs or infinite-width limits. Here, we bridge this gap by studying the structure of GeNTK across many models. Formulating the model state as the solution to a global implicit constraint, we derive GeNTK as a product of two operators: K, accounting for immediate parameter-to-state interactions, and P, describing internal state-to-state dependencies. For a broad class of "weight-based models", such as RNNs and transformers, we prove a universal Kronecker-core theorem showing that K admits an exact, forward-pass-computable form given by the Gram matrix of weight-site variables. This core gives a measurable state-dependent restriction on Euclidean GD, termed "self-referential bias": current activity filters which adjoint directions can drive a first-order update. For recurrent models (GRUs and RNNs), we examine the spectrum of GeNTK and formalize notions of low-rank, biased learning in terms of reduced spatial or temporal ranks of the full operator. Practically, this explains why some RNN initializations yield better learning, while some plateau, by biasing the GeNTK and thus restricting learning to a small domain of learnable tasks. To demonstrate computational viability, we validate the main theorems in a self-attention transformer and a pretrained Pythia transformer language model with 410 million parameters. Overall, our results expose tractable local rank and mode-selective biases in the finite-width GD geometry and open the door to principled study of the GeNTK, a bridge between theory and practice.

comments
Published in Transactions on Machine Learning Research (TMLR), 10/2026
fields
Dynamical Systems, Optimization and Control
pdf ↗ page arXiv:2605.08746 ↗

Muon and related spectral optimizers have recently demonstrated strong empirical performance as scalable stochastic methods, often outperforming Adam. Yet their behavior remains poorly understood. We analyze stochastic Muon, which orthogonalizes an exponential moving average of minibatch gradients, on a high-dimensional matrix-valued least squares problem, and we compare it with stochastic SignSGD, a simple baseline comparable to Adam. We derive explicit deterministic dynamics that provide a tractable framework for studying learning behavior. We show that Muon has non-linear sensitivity to the effective batch $B_{\rm eff}=B(1+β)/(1-β)$. With exact whitening and small $B_{\rm eff}$ Muon fails to resolve smaller eigenmodes, slowing down convergence. Momentum and finite Newton–Shulz iterations can act as a ridge which lessen this effect. This behavior is in contrast with SignSGD which has no transition, leading to different optimal learning rates, but with matching convergence rate up to a constant factor in the isotropic setting. For anisotropic data, large batch setting Muon performs a square-root preconditioning with respect to the data covariance spectrum while SignSGD provides no preconditioning for generic covariances. We analyze time to target loss in a power law covariance model and derive a phase diagram in the data exponent $α$ and target exponent $ζ$. Muon is favored for steep covariance spectra with rich targets ($α>2$, $ζ<α+1$), SignSGD is favored for flat spectra ($α<1$), and in other regions the winner depends on how the target loss scales with dimension. Our results shed light on why replacing Adam with Muon is not trivial even from a pure optimization perspective.

fields
Optimization and Control
pdf ↗ page arXiv:2605.09552 ↗

We show how to use divisors on the projectivized Hodge bundle to construct special vector-valued modular forms and then apply invariant theory to construct all vector-valued Siegel modular forms of level two and degree two. Thus we construct all modular forms in terms of certain basic modular forms that are intimately connected to the moduli of curves of genus two.

comments
22 pages; section 7 revised after AI found a mistake
fields
Algebraic Geometry, Number Theory
pdf ↗ page arXiv:2605.13300 ↗

Already in $\mathbb{R}^4$, there are many minimal hypersurfaces but few structural results. The half-space theorem fails there: higher-dimensional catenoids are among the many examples contained in a slab. Nevertheless, we show that in any dimension and codimension, the height of a minimal submanifold controls its volume growth: confinement to a narrow cone forces polynomial growth, and sublinear height forces Euclidean growth, the slowest possible. Thus volume bounds, which regularity theory assumes, follow from height alone. The key is a volume doubling theorem that converts geometric confinement into quantitative rigidity: if an $n$-dimensional stationary integral varifold lies in a thin slab at a given scale, then its volume doubling ratio at that scale is bounded by a universal constant. Iterating this gives the polynomial growth bounds above. Sublinear height is exactly the borderline for Euclidean growth: examples grow at least like $r^{n+4ε^2/5}$ in cones of aperture $ε$. From Euclidean volume growth, we obtain an optimal Bernstein theorem, in all dimensions, for stable minimal hypersurfaces with sublinear height growth, generalizing results of Moser, Bombieri–De Giorgi–Miranda, Trudinger, Caffarelli–Nirenberg–Spruck, and Ecker–Huisken.

comments
Minor changes
fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2605.15031 ↗

The Korteweg-De Vries (KdV) equation is a paradigmatic model of integrable classical fields, with solitonic solutions where localised profiles interact by elastic, factorised scattering. When solitons interact, their shapes are modified, and individual solitons cannot be identified. This is a key problem in soliton gas where one needs a density of solitons per unit distance: can we locate asymptotic objects, quasiparticles, even at finite times in the macroscopic limit? Quasiparticles' positions should satisfy locality properties, the field in mesoscopic regions being determined by quasiparticles lying there. We solve this problem for the KdV equation. We define solitons' positions as functions of spectral and impact parameters via an expansion bijection generalising that of hard rods. These ``KdV rods'' solve the semiclassical Bethe equations, implementing the collision rate ansatz, and satisfy a repulsion condition generalising the rods' lengths to the KdV scattering shifts and related to the condensate condition. This gives an explicit characterisation of multi-soliton fields in terms of solitons' positions instead of impact parameters. We construct fluid-cell projections, that project out solitons lying outside mesoscopic regions: the resulting field is unchanged in this region and supported there up to small corrections. We show that in the weak limit, conserved densities are given by the empirical density for these solitons' positions. Thus we have associated a kinetic theory of quasiparticles to the KdV equation. The results hold under natural assumptions on spectral and impact parameters, and no randomness is required. Our proof is based on novel solitonic tau functions. This is a crucial step towards rigorously deriving the kinetic equation of the KdV soliton gas, first proposed by Gennady El in 2003. We believe the methods are generalisable to other solitonic models.

comments
v1: 85 pp, 3 figs. v2: 109 pp, 3 figs. Exact solution to SBE. Significantly simplified assumptions, clear physical meaning, substantially strengthen results. Analysis based on SBE instead of magnifying-glass positions. v3: 113 pp, 4 figs, typos corrected, assumptions further simplified."KdV rods" with clear separation conditions, proof that SBE is bijection between this and impact parameters
fields
Mathematical Physics
pdf ↗ page arXiv:2605.18093 ↗

We prove a sharp regularity threshold for uniqueness in two anisotropic Calderón-type inverse problems in dimension $n\ge 3$. The main setting is the Riemannian Schrödinger problem with fixed scalar potential: for a prescribed nonconstant analytic function $V$, we study whether the Dirichlet-to-Neumann map of $-Δ_g+V$ on a domain $Ω\subset\mathbb{R}^n$ determines the unknown metric $g$. The natural gauge is the group of boundary-fixing diffeomorphisms preserving $V$. We show that, while analytic metrics are uniquely determined modulo this gauge by a minor adaptation of the Lassas–Uhlmann reconstruction theorem, uniqueness fails densely in every non-analytic Gevrey class $G^σ$, $σ>1$. In fact, our counterexamples are not isometric in the sense that they are not connected by the pushforward of any diffeomorphism of $\overlineΩ$. We also prove the analogous sharp threshold for the anisotropic Calderón problem at fixed nonzero frequency, thereby upgrading the previously known finite-regularity counterexamples to Gevrey and $C^\infty$ regularity. The two constructions use different scalar mechanisms: for fixed potentials, the nonconstant potential itself provides a local coordinate, while at nonzero frequency one uses a compactly supported prescribed-Jacobian lemma in Gevrey spaces. Thus analyticity is the exact threshold for uniqueness in both problems.

comments
Nonuniqueness results (in the smooth and Gevray category) are correct. Uniqueness in the analytic category is not. Details will be provided elsewhere
fields
Analysis of PDEs, Differential Geometry
pdf ↗ page arXiv:2605.21705 ↗

Condensable algebras in 2+1d non-chiral topological orders characterize gapped boundary conditions and interfaces. Applied to the Symmetry Topological Field Theory, they allow classification of symmetric gapped phases and impose sharp constraints on possible phase transitions. A condensable algebra is specified not only by its underlying set of anyons, which end on the boundary or interface, but also by its algebra structure. We introduce the concept of twin condensable algebras, which have the same anyon decomposition, but inequivalent algebra structure. We revisit the classification of condensable algebras in $\mathcal{Z}(\text{Vec}_G^ω)$, i.e. in group-theoretical topological orders for finite groups $G$ with anomaly $ω$. In this context we are able to identify twin algebras that arise from different mechanisms, such as subgroup data, SPT cocycles, and symmetry actions. In particular, we construct infinite families of examples of twins from so-called Gassmann triples, and exhibit cases in which the reduced topological orders are inequivalent despite having identical anyon content. Physically, twin algebras describe distinct symmetric phases that have isomorphic spaces of ground states, but inequivalent order parameters. Such twin phases never exhibit relative spontaneous symmetry breaking, and can be used to construct phase transitions without hidden symmetry breaking, which are intrinsically beyond Landau transitions.

comments
v2: revised equations and added examples; 40 pages, 2 ancillary files
fields
Quantum Algebra
pdf ↗ page arXiv:2605.31602 ↗

We establish a complete structural classification for cw$_F$-hyperbolic surface homeomorphisms. Specifically, we prove that a surface homeomorphism is cw$_F$-hyperbolic if, and only if, it is a pseudo-Anosov homeomorphism whose singularities consist exclusively of spines (1-prongs). Furthermore, we classify these systems up to topological conjugacy, showing that every such homeomorphism is conjugate to either an Anosov automorphism on the torus $\mathbb{T}^2$ or to a standard hyperelliptic quotient on $\mathbb{S}^2$ of a hyperbolic affine toral map. As a rigid consequence of this classification, we show that such dynamics are strictly obstructed on surfaces of genus greater than one, the Klein bottle, and the projective plane.

fields
Dynamical Systems, Differential Geometry, General Topology
pdf ↗ page arXiv:2606.11943 ↗

Neural operator models trained on simulation data often lose accuracy when applied to experimental measurements due to the sim-to-real gap. Standard fine-tuning with limited real data can reduce this gap, but it may also damage the core physics-relevant representations learned during pretraining. Although knowledge-preserving adaptation has been widely investigated in vision or language tasks, it remains unclear whether these methods are suitable for neural operators whose architectures and protected knowledge are fundamentally different. Neural operators need to preserve core-scale physical structures rather than semantic or visual features. We propose PhysGuard, a physics-preserving framework for accurate sim-to-real adaptation of neural operators. Specifically, PhysGuard uses the empirical Fisher Information Matrix computed on simulation data to identify physics-critical parameter directions, then restricts fine-tuning updates to directions that do not interfere with them. A layer-wise Gram-matrix formulation makes this efficient for models with millions of parameters, while an adaptive threshold automatically determines the protected subspace size. A spectral probe experiment shows that the dominant Fisher directions are strongly associated with low-frequency output structures. Experiments on benchmark across four neural operator architectures and different physical systems show that PhysGuard performs strongly on most evaluation metrics compared to baselines. The benefits are most evident under severe domain shift, where it reduces low-frequency error by up to 32% compared to standard fine-tuning while maintaining adaptability. Our code is available at https://github.com/Data-Driven-Computational-Geotechnics/PhysGuard.

comments
Accepted at NeurIPS 2026
fields
Numerical Analysis
pdf ↗ page arXiv:2606.16602 ↗

We establish uniqueness of blow-ups, with sharp quantitative convergence, for several classes of singular minimizing cones in the Alt-Phillips problem, in the range $γ\in(0,2)$. As a consequence, we obtain uniqueness at every free boundary point in dimensions $d=2,3,4$ for $γ\in(1,2)$, and in dimensions $d\geq 5$ for $γ\in\left(1,\frac32\right)$. The proof of uniqueness is based on three new logarithmic epiperimetric inequalities. The sharp distinction between polynomial and logarithmic convergence is governed by a finite-dimensional integrability condition (sub-integrability) for the spherical linearized problem. We prove this sharpness for the radial cone and its cylindrical extensions through an explicit integrability and bifurcation analysis, showing in particular that logarithmic convergence may be sharp even in dimension two. In contrast, we show that the one-dimensional cone is exceptional: although the integrability condition fails, the convergence is polynomial. Finally, we characterize the minimality of the radial cone in terms of $d$ and $γ$ by means of a one-dimensional calibration argument, exhibiting in dimension $d\geq6$ a nontrivial regime in which the radial cone is stable but not minimizing.

fields
Analysis of PDEs
pdf ↗ page arXiv:2606.17015 ↗

Let $X$ be a log del Pezzo surface with only $\frac13(1,1)$ singularities. We compare the derived categories of its minimal resolution, canonical stack, and singular surface. For every family in the Corti–Heuberger classification, Orlov's blow-up formula and the Ishii–Ueda correspondence give explicit full exceptional collections on the canonical stacks, of length $12-K_X^2+\frac{4k}{3}$ when $X$ has $k$ singular points. In the one-point index-one cascade, the canonical stacks are blow-ups of the weighted projective stack $\mathbb P(1,1,3)$ at distinct smooth points and admit explicit full strong exceptional collections of line bundles. For quasismooth hypersurfaces $X_{10}\subset\mathbb P(1,2,3,5)$, we construct a full strong exceptional collection of $13$ objects with an explicitly computed Euler matrix. The exceptional lattice determines the Brauer group: $\mathrm{Br}(X)=0$ for $k\leq5$, while $\mathrm{Br}(X)\cong\mathbb Z/3$ for $k=6$. We construct semiorthogonal decompositions of $D^b(\mathrm{coh}\,X)$ for the one- and three-point cascades and of a Brauer-twisted derived category for the six-point cascade.

comments
64 pages, 7 figures. Major revision with a new title. Added explicit collections on the rigid cascade bases, full strong line-bundle collections for the one-point cascade, and descent results for the three-point and Brauer-twisted six-point cases. Expanded Euler-matrix comparisons, reorganized proofs, and clarified geometric attributions
fields
Algebraic Geometry
pdf ↗ page arXiv:2606.18238 ↗

In this paper we study the dynamics of relativistic detonation waves theoretically and numerically. The reaction is physically accounted for by an extra term in the definition of the total energy density and by an additional equation for the evolution of the mass fraction of the reactant, while leaving formally unmodified the equations of mass and energy-momentum conservation. In this way, the Rankine-Hugoniot relations maintain the same formal structure of the inert version. For the numerical solution we use a finite volume ALE scheme with a second order TVD space-time reconstruction for the homogeneous part of the system, where the mesh velocity is chosen equal to the shock speed, and we combine it through operator splitting with a locally implicit algorithm for the potentially stiff reaction source terms that arise in the equation of the reactant. We furthermore propose a particularly efficient algorithm for the conversion from the conserved to the primitive variables, which for the relativistic Euler equations is known to be nontrivial. Following this approach, we can successfully solve the Zel'dovich-von Neumann-Doering profile of a relativistic detonation wave, up to Lorentz factors of the shock front $γ_s\sim 7$. Our analysis allowed us to highlight a new special relativistic effect, which has remained unnoticed so far. While in Newtonian detonations the Zel'dovich pressure jump decreases monotonically with the mass flux through the shock front, in the relativistic case it shows a minimum and then rises monotonically as a function of the mass flux. This may have interesting physical implications on the amount of energy that can be extracted from a relativistic detonation wave.

comments
27 pages 7 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2606.27113 ↗

The blowing-up of the projective plane at a finite set of points yields a del Pezzo surface if and only if the points lie in general position. In this note, we generalize this result to Brauer–Severi surfaces over arbitrary ground fields. Using Galois descent, intersection theory and combinatorial arguments, we provide explicit arithmetic and geometric conditions on the centre of the blowing-up.

comments
9 pages, 3 figures. To appear in Bulletin of the Polish Academy of Sciences - Mathematics
fields
Algebraic Geometry
pdf ↗ page arXiv:2606.27940 ↗

A complete theory of null Cartan normal helices in Minkowski space-time $\mathbb{E}^4_1$ is developed. Two algebraic conditions, obtained by successive differentiation of the helix invariant along a unit $C$-constant normal field, fully characterize null Cartan helices; the quadratic condition yields two mutually orthogonal helix axes in the Lorentzian metric. Special field types are analyzed and null Cartan cubics are shown to be normal helices. On a timelike hypersurface, a Darboux frame with six curvature functions is constructed from first principles, the normal isophotic condition is shown to reduce to a linear first-order ODE, and the existence of normal silhouettes in $\mathbb{E}^4_1$ is established.

fields
General Mathematics
pdf ↗ page arXiv:2607.01262 ↗

We investigate the probabilistic counterpart of Yau's conjecture on the nodal volume of Laplace eigenfunctions on compact manifolds, by studying the high-frequency variance asymptotics of Riemannian random waves. Theorem A gives quantitative bounds on the fluctuations of their nodal volume for all regimes of spectral windows, including the monochromatic one, of spectral size 1. For every non-monochromatic window the variance has the exact order $1/N$, where $N$ is the dimension of the underlying space of eigenfunctions, on any compact manifold. In the monochromatic regime, on manifolds without conjugate points and in particular on negatively curved ones, we prove that the variance is $o(1/N)$: this improves the previous best bound, cf. Canzani and Hanin 2020, to its square divided by a logarithmic factor, and exhibits a form of Berry's cancellation, relative to the natural scale $1/N$, for monochromatic random waves on such chaotic manifolds. Our proofs rely on a local and global analysis combining the Kac-Rice formula, the Wiener–Itô chaos decomposition recently obtained by the authors Stecconi and Todino 2025, and a sharp analysis of the error in the pointwise Weyl law associated to an arbitrary spectral window (Theorem B). We also introduce a general machinery (Theorem C) that yields variance decay for the nodal volume of an arbitrary Gaussian field under broad geometric conditions, subject to correlation decay assumptions.

fields
Probability, Mathematical Physics, Differential Geometry
pdf ↗ page arXiv:2607.09946 ↗

We develop a sketch-and-restart framework for computing the action of a matrix function on a vector, $f(A) b$, where $A$ is large, sparse, and non-Hermitian. The framework combines quadrature-based restarting with Arnoldi-like decompositions generated by sketched or truncated Arnoldi processes. Within this framework, we develop two classes of restarted algorithms. The first uses a fixed Krylov subspace dimension and is based either on the sketched Arnoldi process or on a new sketched harmonic Arnoldi process proposed in this work. The second class chooses the Krylov subspace dimension adaptively by running the truncated Arnoldi process until the condition number of the generated basis, estimated from its sketch, exceeds a prescribed threshold. We also establish the convergence of the restarted sketched harmonic Arnoldi method for Stieltjes functions under the assumption that $A$ is positive real. Numerical experiments demonstrate the effectiveness of the proposed framework, including the computational savings achieved through sketching, the storage reduction enabled by adaptive truncation, and the acceleration obtained from thick restarting.

comments
30 pages, 6 figures; revised version
fields
Numerical Analysis
pdf ↗ page arXiv:2607.10354 ↗

We study the periodicity and $p$-adic convergence of sequences of values of the translated division polynomials $(Ψ_n(P))_{n\geq 0}$, $(Φ_n(P))_{n\geq 0}$, and $(Ω_n(P))_{n\geq 0}$ evaluated at a pair of points on an elliptic curve. We prove that these sequences are purely periodic over finite fields. We then use their periodicity properties to prove that certain subsequences are $\mathbb{Z}_p$-Cauchy. Applying this result to Somos 4 and Somos 5 sequences yields analogous convergence results. We also establish pure periodicity modulo prime powers for the even- and odd-indexed subsequences of Somos 5 sequences under suitable hypotheses.

comments
Revised version; proofs and hypotheses corrected and clarified, and exposition improved
fields
Number Theory
pdf ↗ page arXiv:2607.11261 ↗

We study elliptic superlevel geometry and return dynamics for the class $\mathcal{O}_{C}$. Let $C \subset \mathbb{R}^{N}$ be a compact strictly convex core, and $Ω\in \mathcal{O}_{C}$ a bounded domain whose boundary is a global normal graph over $\partial C$ with inward normals returning to $\operatorname{int} C$. For $-Δu = f$ in $Ω$ ($u=0$ on $\partialΩ$), with $f \geq 0$, $f \not\equiv 0$, $\operatorname{supp} f \Subset \operatorname{int} C$, we analyze the superlevel sets $Ω_{t} = \{u > t\}$ for $t \downarrow 0$. We prove that for small $t > 0$, $\partialΩ_{t}$ is a global normal graph given by $\partialΩ_{t} = \{c + d_{t}(c)ν(c) : c \in \partial C\}$, satisfying $\|d_{t} - d_{0}\|_{C^{1}} = O(t)$ and $d_{t} = d_{0} - t b_{0} + o_{C^{1}}(t)$, where $b_{0}$ depends on the Hopf derivative and transversality. Using normal-graph formulas, we show $Ω_{t} \in \mathcal{O}_{C}$. The return maps $F_{t} : \partial C \to \partial C$ are well-defined, satisfying $\|F_{t} - F_{0}\|_{L^{\infty}} = O(t)$ and $F_{t} = F_{0} + t \mathcal{A}_{0} + o_{L^{\infty}}(t)$. For any sequence $t_{n} \downarrow 0$, the non-autonomous system $x_{n+1} = F_{t_{n}}(x_{n})$ is asymptotically autonomous with background map $F_{0}$. These results bridge elliptic boundary-layer deformation and return-map dynamics relative to a convex core.

fields
Analysis of PDEs
pdf ↗ page arXiv:2607.16912 ↗

In this paper, we construct countable uniformly locally finite metric spaces $X$ and $Y$ such that $C_u^*(X)$ is isomorphic to a hereditary $C^*$-subalgebra of $C_u^*(Y)$, but neither space coarsely embeds into the other. This gives a negative answer to the embedding rigidity problem for uniformly locally finite coarse spaces. On the positive side, we prove that a hereditary embedding $C_u^*(X)\to C_u^*(Y)$ implies an injective coarse embedding $X\to Y$ whenever $C_u^*(Y)$ contains no block-rank-one ghost projection. In particular, this strengthens the coarse embedding theorem of Braga, Farah, and Vignati by obtaining injectivity under the same hypothesis that every sparse subspace of $Y$ yields only compact ghost projections.

comments
v2, 17 pages; v1, 23 pages
fields
Operator Algebras
pdf ↗ page arXiv:2607.16949 ↗

This is an expository article. The curve complex of a surface is deliberately forgetful, yet in most cases its automorphism group recovers the mapping class group. We explain what happens when the same question is asked for the essential-surface complex $\mathcal{ES}(E)$ of a knot exterior $E=E(K)$, equivalently Schultens's initial surface complex $\mathcal{S}_0(E)$. The discussion is organized around three questions: which combinatorial symmetries are geometric (the image), which homeomorphisms are invisible (the kernel), and how much of $E$ can be read off from the complex (reconstruction). The basic objects are first drawn in classical examples, and the general framework is introduced only afterwards. We collect the elementary structural facts that organize the subject -- the flag property, the layering by boundary slope, finite dimension, unbounded dimension for iterated cables, and Kakimizu's complexes as full subcomplexes -- and compare with handlebodies and connected sums of $S^2\times S^1$, where image and kernel are known completely. Along the way we record two short consequences that we have not found stated in this form: for a slope-separated knot every orientation-preserving mapping class acts trivially, and the image is $\mathbb{Z}/2$ exactly when the knot is amphichiral and a nonzero boundary slope occurs; and for a connected sum of nonfibered prime knots the decomposing-annulus twist translates Banks's winding coordinate. A table separates what is classical from what is recorded here. The article closes with a three-task formulation of the rigidity problem, a list of problems with their current status, a worked two-bridge recipe, and a guide to further reading.

comments
41 pages, 17 figures, 1 table. v3: rewritten as an expository article; added a status table of classical vs. new results, a comparison with handlebodies and #S^2xS^1, a problem list with status, and further references; corrected the two-bridge continued-fraction convention
fields
Geometric Topology
pdf ↗ page arXiv:2607.20869 ↗

A tree cover of an $n$-point metric space is a collection of $k$ trees such that every pairwise distance is approximately preserved in at least one tree. For fixed $k$, the best known distortion upper bound is $\tilde{O}(n^{1/k})$ and the best known lower bound is $Ω_k(n^{1/2^{k-1}})$. In this paper, we prove a lower bound of $Ω_k(n^{1/(k-1)^2})$ for all $k\ge 3$, significantly narrowing the gap towards the upper bound. Our approach makes use of a novel algebraic topological notion that we call the \emph{agreement index}: the equivariant topological index of a simplicial complex encoding pairwise agreement among tree labels. This notion helps characterize the power and limit of algebraic and topological methods for analyzing tree covers, and we believe it will also find applications to other problems.

fields
Combinatorics
pdf ↗ page arXiv:2607.27264 ↗

We survey a circle of recent results showing that a minimal submanifold obeys a dichotomy: either it fills up space, spreading out like a space-filling curve, or it is confined, and confinement forces quantitative restrictions. On the confined side these restrictions are both geometric and function-theoretic: Euclidean volume growth, an optimal rate of convergence of the density, complex-curve rigidity for stable surfaces in $\bf{R}^4$, and a Liouville theorem forcing slowly growing harmonic functions on minimal disks to be constant. A prototypical restriction is Euclidean volume growth, which in these results is forced by the geometry rather than assumed. The mechanism is a volume doubling theorem that converts geometric confinement into quantitative rigidity: a stationary integral varifold trapped in a thin slab at a given scale cannot double its volume by more than a universal factor. We explain how this principle, and the height-excess bounds behind it, produce Euclidean volume growth for submanifolds of sublinearly growing height in every dimension and codimension, the optimal density rate in a slab, the complex-curve and Liouville rigidity above, a higher-codimension Bernstein theorem for disks, one-sided volume bounds, and an optimal stable Bernstein theorem in all dimensions generalizing Moser, Bombieri-De Giorgi-Miranda, Caffarelli-Nirenberg-Spruck and Ecker-Huisken. We describe three regimes with sharp borderlines. In narrow cones the volume grows polynomially with exponent $n+O(\varepsilon^2)$, and the order $\varepsilon^2$ is attained; in wide cones the growth can be arbitrarily fast. For holomorphic curves in $\mathbb{C}^2$, there are also three regimes and the borderline is determined by the Kähler angle.

fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2607.29502 ↗

Recently, rectified flow has emerged as a fundamental framework for large-scale image generation, powering state-of-the-art systems such as FLUX.1 and Stable Diffusion 3. Despite its remarkable empirical success, the computational and statistical guarantees of iterative rectified flow have remained largely unexplored. We address this problem by studying c-rectified flow, a cost-aware class of rectified flow that projects velocity fields onto a gradient class while preserving endpoint marginals. The ordinary rectified flow can fail to recover the optimal transport coupling: in a Gaussian case study, the iteration converges to the optimal coupling if and only if the source and target covariance matrices commute. More generally, for a Gaussian to log-concave coupling, the ordinary rectified flow can reach a fixed point after one step from independent initialization. In contrast, under suitable regularity and uniform-integrability assumptions, iterative c-rectified flow always converges to the optimal transport coupling. We further establish quantitative one-step contraction and exponential convergence guarantees under projection-stability assumptions for both quadratic and strongly convex displacement costs. Finally, under a Hölder ball assumption, we develop new minimax-optimal score estimation rates and show that, when combined with iterative c-rectified flow, they yield a rate-optimal estimator of the optimal transport for the dimension \(d \ge 3\) and a nearly parametric rate for \(d=1,2\).

fields
Optimization and Control, Probability, Statistics Theory
pdf ↗ page arXiv:2608.02487 ↗

We prove that a Noetherian local ring $R$ with canonical module $ω_R$ is quasi-Gorenstein if and only if $\Fitt_1^R(ω_R) \cong ω_R$. This settles the expectation of Eisenbud, Ficarra, Herzog, and Moradi on canonical ideals.

comments
2 pages, I simplified the proof
fields
Commutative Algebra
pdf ↗ page arXiv:2608.03766 ↗

We construct noncommutative mirrors from the bulk-deformed Floer theory of compact immersed Lagrangians equipped with higher rank flat bundles, whose Floer theory can have non-zero curvature terms due to contributions of stable constant discs. For a nodal union of Lagrangian spheres, codimension-two bulk cycles meeting the compact core produce the additive deformed preprojective relations and the corresponding framed Maurer-Cartan spaces at nonzero complex moment-map levels. In particular, our localized mirror functors produce noncommutative ADHM and Nakajima monadic complexes. We obtain local charts and gluing maps of the noncommutative mirror varieties via explicit computations of Lagrangian deformation spaces and Fukaya isomorphisms between them. These constructions recover noncommutative spaces studied by Kapustin-Kuznetsov-Orlov, Baranovsky-Ginzburg-Kuznetsov, and Kawamata, with further examples in dimension three. We also establish the categorical framework in the unobstructed setting. For a compact $\mathbb Z$-graded unobstructed immersed Lagrangian $\mathbb{L}$ without nonpositive-degree immersed generators, Koszul duality identifies the derived Fukaya subcategory generated by $\mathbb{L}$ with the derived category of modules with finite-dimensional total cohomology over its extended localized mirror $\tilde{\mathcal{A}}_{\mathbb{L}}$. This construction starts from the Floer algebra of the compact reference brane and does not require a global wrapped generator. Applications to plumbings recover known split-generation results for compact objects.

comments
57 pages. Improved exposition and added a discussion of anticanonical compactifications, including an explicit example. Comments are welcome!
fields
Symplectic Geometry, Algebraic Geometry
pdf ↗ page arXiv:2608.05764 ↗

Gödel's ontological argument is modal by design. Possibility and necessity are among its central concerns, and connected interpretation decisions -- extensions or intensions for the positive properties, constant or varying quantifier domains, the choice of modal logic, etc. -- subtly shape the argument and, in particular, whether the modal collapse is implied or not. It is tempting to abstract the modality away and to read the positive properties as merely forming an ultrafilter -- which, however, may replace Gödel's argument with a perhaps more manageable alternative that may have little to do with Gödel's original intention. In this spirit, Odifreddi and Gomes argue that modal collapse is an intrinsic structural feature of any theory that characterises positive properties as an ultrafilter and identifies God as its principal generator. This comment shows that the structural claim is false. There is an explicit S5 model, with constant domains, in which positivity is at every world a modal ultrafilter principally generated by a unique and necessarily existing Godlike individual, and in which the modal collapse nevertheless fails -- already for an atomic formula. What drives the collapse is not the ultrafilter structure but the rigidity of positivity -- itself a thoroughly modal condition: given only that positivity is a modal ultrafilter, modal collapse and rigidity are equivalent. Rigidity is thus exactly what the collapse adds to the ultrafilter structure, and the disagreement is whether it is to be counted part of that structure at all. Two further claims of Odifreddi and Gomes are clarified: one on the derivability of Gödel's Theorem IV, one on the extensionality of positivity. All results are machine-checked in Isabelle/HOL, which distinguishes kernel-checked proofs from model-finder output.

comments
21 pages. Accepted for publication in Logica Universalis (revised version). Companion Isabelle/HOL session included as ancillary files (directory anc/GoedelUltrafilterNote); builds with Isabelle2025-2 in about 15 s. Previous title: "A Comment on Modal Collapse and Ultrafilters in Gödel's Ontological Argument"
fields
Logic
pdf ↗ page arXiv:2608.07578 ↗

We present a systematic study of finite, non-split 2-group symmetries with a non-trivial Postnikov class, focusing on the simplest example with a $\mathbb{Z}_2$ 0-form symmetry and a $\mathbb{Z}_2$ 1-form symmetry, intertwined together by the non-trivial Postnikov class in $H^3(B\mathbb{Z}_2;\mathbb{Z}_2)\cong\mathbb{Z}_2$, denoted by $\mathcal{G}$. We classify the anomalies of this 2-group symmetry for physical theories in $d$-dimensional spacetime, by computing the oriented bordism groups $Ω_{d+1}^{\rm SO}(B\mathcal{G})$ and the spin bordism groups $Ω_{d+1}^{\rm Spin}(B\mathcal{G})$ for $d\leq 5$. For the case of $d=3$, we investigate the Symmetry TFT/TO of the 2-group symmetry $\mathcal{G}$ using the language of fusion 2-categories, as well as (3+1)D TQFT actions, for the cases without or with the 2-group anomaly classified by $\operatorname{Hom}\left(\widetildeΩ_4^{\rm SO}(B\mathcal{G}),U(1)\right) \cong H^4(B\mathcal{G};U(1))\cong\mathbb{Z}_2$. We classify the minimal topological and physical boundary conditions of the Symmetry TFTs, and carry out the categorical Landau paradigm for such a non-split finite 2-group.

comments
43 pages + appendices; v4
fields
Mathematical Physics, Algebraic Topology
pdf ↗ page arXiv:2608.08766 ↗

We study landscapes for complex-parameterized networks. Our approach is motivated with an information-theoretic manifold perspective of the parameter and via classical optimization guarantees although of complex geometric variety such as through Dolbeault asymptotics. The descent path admits a Kähler information metric under a cross-entropy via the Wirtinger Hessian on the log-likelihood potential. We restrict attention to a descent update rule with natural gradient descent via a differentiated loss scaled by the inverse metric, so the descent path remains in the holomorphic tangent bundle. We emphasize Calabi-Yau information manifolds which profane theoretical guarantees via an ill-curvature-conditioned, or lack thereof, landscape. We focus on Calabi-Yau metrics specifically in a non-compact setting with a global potential, so defined geometrically rather than invoking the topological requirements of the Calabi conjecture. In non-compact settings, we can write the metric determinant with respect to a background in terms of a function, pluriharmonic or real. Using Yau's Liouville theorem, we get a partial eigenvalue blow-up effect. In an empirical setting, a Ricci-flat metric will not form, but a blow-up can be a local condition and can hold empirically on open sets even partially in generalized Kähler cases due to lower bounds on the Ricci potential. Moreover, it has been discovered that negative curvature subverts the loss landscape, specifically sectional curvature, so we expand on this and draw interconnections to negative-definite Ricci curvature. Our arguments primarily exist via geometric modality, although we establish roots in deep learning theory such as through asymptotics at initialization and connections through failure modes of neural network guarantees under vanishing and negative Ricci curvature.

fields
Differential Geometry
pdf ↗ page arXiv:2608.19584 ↗

Stochastic reaction networks are continuous-time Markov chain models for interacting populations, with applications in biochemistry, epidemiology, ecology, and related areas. Coupling multiple stochastic reaction networks effectively is a fundamental challenge in stochastic simulation. Existing variance-reducing coupling methods are typically designed for pairs of processes, though specialized multi-path constructions have been developed for particular applications. To fill this gap, we introduce the multi-path stacked coupling (MSC), a space-time Poisson construction that jointly simulates a finite collection of stochastic reaction networks with common reaction channels and reaction vectors, but possibly different intensity functions and initial conditions. Each pairwise marginal of MSC has the same law as the corresponding split-coupled pair, allowing existing variance bounds to transfer directly; in finite-state settings, we also obtain first-order expansions for the mean and second moment of finite-difference numerators. We demonstrate the framework through three sensitivity estimation tasks requiring several nearby parameterized paths: estimating many first derivatives simultaneously, estimating a single first derivative using a wider finite-difference stencil, and estimating higher-order derivatives. Numerical experiments on a processive phosphorylation network demonstrate strong performance in each of the three application areas considered, consistent with the theoretical advantages of MSC: across all three applications, MSC achieves the smallest root mean square error (RMSE) among the methods considered over the tested computational budgets.

comments
Substantially revised and retitled to emphasize the general multi-path coupling framework; numerical results unchanged. Now 28 pages
fields
Numerical Analysis, Probability
pdf ↗ page arXiv:2608.22627 ↗

For two digraph properties $P_1$ and $P_2$, a $(P_1,P_2)$-arc-decomposition of a digraph $D$ is a partition $A(D)=A_1\mathbin{\dot\cup}A_2$ such that the spanning subdigraphs $D[A_1]$ and $D[A_2]$ have properties $P_1$ and $P_2$, respectively. We prove that it is NP-complete to decide whether a digraph admits an arc-decomposition with properties $(P_1,P_2)$ where $(P_1,P_2)\in \{$(perfect matching, having no odd cycle), (perfect matching, strong), (perfect matching, having an out-branching), (cycle factor, having no odd cycle)$\}$. All four problems remain NP-complete even when restricted to strongly connected digraphs. We also prove that every strongly connected digraph with minimum semi-degree at least three contains two vertex-disjoint cycles whose simultaneous arc deletion leaves the digraph strongly connected. These results settle some open problems posed by Bang-Jensen, Bessy, Gonçalves, and Picasarri-Arrieta [Theoret. Comput. Sci. 928 (2022), 167–182].

comments
18 pages, 10 figures
fields
Combinatorics
pdf ↗ page arXiv:2608.24031 ↗

We introduce stable symmetric series which encode normalized conjugacy classes and their multiplication operators simultaneously for all symmetric groups. This gives a direct route from the Ivanov–Kerov algebra to shifted symmetric functions and to differential operators in $U(\mathcal W_{1+\infty})$. Using the Goulden–Jackson product, we extend the construction to Jack polynomials, recover shifted Jack eigenvalues and Pieri-type relations, and obtain explicit candidate operators in degrees three and four. Their real and quaternionic specializations to zonal polynomials are verified by Gaussian matrix integrals and exhaustive Wick enumeration.

comments
32 pages, 2 figures
fields
Combinatorics
pdf ↗ page arXiv:2608.25651 ↗

We study bubbling for sequences of solutions of the Hamiltonian-perturbed Fueter equation. By Walpuski's compactness theorem, a bounded-energy sequence converges weakly to a Fueter map, with energy loss recorded by a defect measure supported on a codimension 2 rectifiable set. Assuming that the limiting map has no non-removable singularities and the bubbles attach to it on a subset of the bubbling locus of positive measure, we show that the limiting Hamiltonian lies in an exceptional subset of infinite codimension. The proof combines quantitative control of rational sweepout loci in hyperkähler manifolds with a transversality argument. We expect the attachment assumption to be automatic and follow from a suitable no-neck result. This strongly suggests that, in the absence of non-removable singularities, bubbling of Hamiltonian-perturbed Fueter maps is highly nongeneric.

comments
Main theorem slightly strengthened, exposition improved and a reference added
fields
Differential Geometry
pdf ↗ page arXiv:2608.26803 ↗

The topological Halmos–von Neumann theorem is a fundamental result of topological dynamics that completely classifies continuous actions of abelian topological groups which are ergodic and have discrete spectrum. It rests on a correspondence between such actions (up to the choice of a distinguished point), compactifications of the acting group and subgroups of its dual group. We generalize both the classical statement and the underlying categorical equivalences to the framework of open group bundles over locally compact base spaces. As a byproduct, we establish a Pontryagin-type duality between étale Hausdorff abelian group bundles and open proper Hausdorff abelian group bundles, extending the classical duality between discrete and compact abelian groups.

fields
Dynamical Systems, General Topology, Group Theory
pdf ↗ page arXiv:2608.28335 ↗

This is the first publication, in which an inverse problem of the simultaneous reconstruction of an unknown (x,t) - dependent coefficient and an unknown initial condition in a general linear parabolic equation of the second order is considered. The input data depend on (n+1) variables and are, therefore, formally determined ones. Stability estimates and uniqueness theorem are obtained for this inverse problem. As a by-product, logarithmic stability estimates for initial conditions of parabolic equations and inequalities with time reversed data are obtained for the first time. All known publications about inverse problems for parabolic equations with formally determined input data and unknown coefficients assume that those coefficients depend either only on x or only on t. In addition, the initial condition is assumed to be known, except of two publications cited in the text. The inverse problem of this paper has two potential applications. The first one is in forecasting of public opinions in the framework of the Mean Field Games theory. The second one is in tracking spatiotemporal outbreaks of epidemics.

fields
Mathematical Physics
pdf ↗ page arXiv:2608.29377 ↗

We characterize exact nonuniqueness in the planar interior Radon problem for arbitrary pairs of open convex sets. There exists a nonzero smooth function compactly supported in the first set whose integral over every line meeting the second set vanishes if and only if an ellipse contains the closure of the second set and is compactly contained in the first set. The same criterion holds for a nonzero $L^1$ function whose essential support is compactly contained in the first set and whose Radon transform vanishes almost everywhere on those lines. For concentric open squares, the criterion gives the sharp threshold $1/\sqrt2$ for the inner-to-outer half-side ratio. This yields counterexamples to Conjecture 1.2 of Boman (2021) and to Theorem 40.1 of Boman (2025).

comments
16 pages, 2 figures
fields
Classical Analysis and ODEs, Functional Analysis
pdf ↗ page arXiv:2608.29442 ↗

We design low-complexity, fault-tolerant decoders for quantum low-density parity-check (QLDPC) codes with the goal of reducing decoding latency. We target two major bottlenecks of decoding under the \emph{circuit-level} noise model: (i) post-processing via order-statistics decoding (OSD), and (ii) the large number of auxiliary variable nodes commonly introduced to represent CNOT-induced correlations during syndrome extraction. Our key observation is that propagating CNOT faults (\emph{hook errors}) create \emph{stabilizer-induced} trapping sets (TSs) that are intrinsic to hypergraph-product (HGP) and lifted-product (LP) constructions. Therefore, instead of modeling each such fault with an explicit correlation node and relying on OSD to clean up the resulting failures, we design message-passing decoders that resolve the corresponding \emph{stabilizer-induced} TSs directly. We obtain these decoders by deriving QLDPC decoders from decoders for the parent classical LDPC codes and using them collectively to correct broad families of \emph{stabilizer-induced} TSs. For CNOT faults that manifest primarily as syndrome errors, we show that their effect is equivalent to a data error together with syndrome-bit measurement errors. Consequently, given repeated measurements and a decoding graph that already includes nodes representing syndrome-bit errors, no distinct variable node is needed for each CNOT fault. Using a \emph{phenomenological} Tanner graph with nodes representing only data errors and syndrome-bit errors, simulations on LP codes show a significant improvement in logical error rate relative to BP+OSD. Moreover, the improved error correction relies on low-complexity finite-state iterative decoders, targeted for fast and low-latency FPGA implementation.

fields
Information Theory
pdf ↗ page arXiv:2608.31040 ↗

For a row-finite weighted graph $(E,w)$, Preusser showed that the monoid $\mathcal{V}(L_k(E,w))$ of finitely generated projective modules over the weighted Leavitt path algebra $L_k(E,w)$ is isomorphic to a combinatorially defined weighted graph monoid $\mathcal{M}(E,w)$. We study two structural properties of $\mathcal{M}(E,w)$: confluence and cancellativity. We introduce a reduction system on the free commutative monoid presenting $\mathcal{M}(E,w)$, obtain sufficient conditions for non-confluence by constructing explicit non-confluent triples, and provide a complete confluence characterization for certain classes of weighted graphs. Turning to cancellativity, we work within Preusser's class of weighted graphs satisfying Condition (LPA), for which $L_k(E,w)$ is isomorphic to an unweighted Leavitt path algebra $L_k(F)$ via a two-step construction. We introduce an auxiliary graph associated to the intermediate step of this construction and use it to give a graph-theoretic characterization of when $\mathcal{M}(E,w)$ is cancellative. Finally, under Condition (LPA), we show that Preusser's construction upgrades to a graded isomorphism $L_k(E,w) \cong_{\operatorname{gr}} L_k(F)$ with respect to the standard $\mathbb{Z}^{λ(E,w)}$-grading of weighted Leavitt path algebras, yielding $\mathcal{V}^{\operatorname{gr}}(L_k(E,w)) \cong \mathcal{V}^{\operatorname{gr}}(L_k(F))$ as $\mathbb{Z}^{λ(E,w)}$-monoids.

comments
23 pages. v2: corrected typos and some errors in Sections 3 and 4. Comments are welcome
fields
Rings and Algebras, Combinatorics
pdf ↗ page arXiv:2608.31087 ↗

Let $G$ be a compact connected Lie group and $K$ its connected Lie subgroup. Using the $K$-theoretic version of equivariant formality developed in [F2] which involves analysis of vector bundles over $G/K$, we find two characterizations of equivariant formality of the isotropy action of $K$ on $G/K$. The first one equates equivariant formality with a "smoothness" condition of the restriction map of the representation ring of $G$ to that of $K$. The second characterization asserts that equivariant formality amounts to the equivariant $K$-theoretic index pairing of two certain special $K$-theory classes being nontrivial in some sense. By applying these characterizations, we are able to give a representation theoretic criterion for equivariant formality of the isotropy action by a circle subgroup, as well as an invariant theory criterion for the case where $K$ is a torus of dimension one less than the rank of $G$, culminating in a complete classification of equivariant formality where $G$ is further assumed to be simple.

comments
28 pages, 1 table. Computation of K-theoretic index in Example 5.3 corrected; proof of Theorem 1.4 expanded
fields
Algebraic Topology, K-Theory and Homology
pdf ↗ page arXiv:2609.00448 ↗

We study when and how momentum improves large-batch training in the one-pass regime, using power-law kernel regression as a tractable setting. We first characterize stochastic stability through the \emph{critical learning rate}, defined as the largest learning rate for stable training, and obtain $η_{\mathrm{SGD}}^{\mathrm{crit}}\eqsim 1$, $η_{\mathrm{Polyak}}^{\mathrm{crit}}\eqsim \min\{1,B(1-β)\}$, and $η_{\mathrm{Nesterov}}^{\mathrm{crit}}\eqsim \min\{1,B^q(1-β)\}$, where $B$ is the batch size, $β\in[0,1)$ is the momentum factor, and \(q>1\) is the capacity exponent governing spectral decay. Within the stable region, we derive scaling laws for the full risk dynamics, capturing the progression from an early transient through power-law decay to a noise floor. We then optimize the final-step risk under a fixed data budget, obtaining a three-regime phase diagram that reveals how the benefits of momentum change with batch size. Notably, Polyak enlarges the critical batch size, the largest batch size preserving the optimal small-batch data-scaling exponent, thereby allowing fewer sequential updates without sacrificing data efficiency. In contrast, Nesterov improves the optimal data-scaling exponent in the large-batch regime through look-ahead-induced suppression of noise amplification. Experiments on PLK regression quantitatively support the theoretical scaling predictions, while neural-network experiments provide qualitative evidence for the predicted batch-size dependence and stability behavior.

comments
63 pages, 10 figures, 1 table. Revised proofs; added neural-network experiments
fields
Optimization and Control
pdf ↗ page arXiv:2609.02728 ↗

Goal-oriented sensing, estimation, and control benefit from prompt, regular access to task-relevant updates. Although 1-persistent age-threshold slotted ALOHA (1-pTSA) can self-organize into a periodic collision-free schedule, its acquisition transient can dominate finite-horizon performance. We propose role-protected counter-threshold slotted ALOHA (RP–CTSA), which separates reservation memory from the age of information (AoI). A scheduled singleton retains its phase after colliding with active contenders, whereas a collision involving multiple scheduled nodes releases them immediately; AoI resets only after a decoded update. The protocol requires individual acknowledgments and a binary RELEASE/HOLD indication, but no centralized phase assignment or identification of colliders. Let $n$ denote the number of nodes and $Γ_n$ the counter threshold. Under inverse-population access scaling, the acquisition dynamics admit an exact pure-death representation. When $Γ_n=n$, the acquisition time is $O(n^2)$; when $Γ_n\sim(1+θ)n$, with $θ>0$, it is $O(n\log n)$. Simulations confirm both regimes and show substantial finite-horizon AoI gains over 1-pTSA while preserving every collision-free schedule.

comments
Submitted to Wireless Networks
fields
Information Theory
pdf ↗ page arXiv:2609.09934 ↗

Let $S=\kk[x_1,\ldots,x_n]$ with its standard $\mathbb N^n$-grading, and let $\mathfrak m=(x_1,\ldots,x_n)$. We study the essential images of the functors $M\mapsto H^i_{\mathfrak m}(M(-\mathbf1))$ on categories of positively determined $S$-modules, where $\mathbf1=(1,\ldots,1)$. For a fixed determining bound $\mathbf t\in\mathbb N^n$ and $0\le i<n$, we identify the image in degree $i$ with the category of finite-dimensional representations of the subposet of $[\mathbf0,\mathbf t]$ consisting of multidegrees with at least $n-i$ positive coordinates. These images are Serre subcategories of the category of negatively $\mathbf t$-determined modules. The top-degree image consists precisely of second cosyzygies in that category and is extension-closed, but is not abelian when $\mathbf t\ne\mathbf0$. For the non-top image categories, we determine attached primes from projective covers and compute the Bass numbers at $\mathfrak m$ and the injective dimensions over $S$ of projective and injective objects. In every non-top degree, we construct an exact functorial section selecting, for each nonzero target, a source of minimal projective dimension $n-i$ and maximal depth $i$ among all its sources. If $n\ge2$ and $\mathbf t\ge\mathbf1$, no fully faithful section exists in degrees $1$ and $n-1$, even without requiring exactness. In top degree, graded Matlis duality followed by $S$-duality gives a fully faithful functorial section. We also describe the global images obtained by varying the determining bound. The non-top global categories have enough injectives and no nonzero projective objects, while the top category corresponds to second cosyzygies in the category of finite-support representations of $\mathbb N^n$.

comments
34 pages comments are welcome
fields
Commutative Algebra
pdf ↗ page arXiv:2609.11219 ↗

For a nontrivial finite abelian group $G$, let $ν(G)$ be the smallest {nonnegative} integer $\ell$ such that every zero-sum free sequence $T$ over $G$ of length at least $\ell$ has the following property: all nonzero elements of $G$ that do not occur as a subsequence sum of $T$ lie in a proper coset of some subgroup of $G$. It is easy to check that $\mathsf d (G)-1 \le ν(G) \le \mathsf d (G)$, where $\mathsf d (G)$ is the small Davenport constant of $G$. A conjecture by Gao from the year 2000 stated that equality should always hold at the lower bound. This conjecture has since been confirmed for many families of groups (including all p-groups and groups of rank at most two). In this paper, we disprove the conjecture.

fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2609.17127 ↗

Let $p>3$ be a prime number. We prove that every principally polarized superspecial abelian surface over $\mathbb{F}_{p^{2}}$ with $p^{2}$-Frobenius $[-p]$ admits a separable polarized isogeny to its Frobenius conjugate with multiplier at most $(p^{3}/2)^{1/5}$. This generalizes a recent result of Aubry, Oyono, and Vincent (2026, arXiV:2607.14624) who proved that every supersingular elliptic curve defined over $\bar{\mathbb{F}}_{p}$ admits an isogeny to its Frobenius conjugate with degree at most $(p/2)^{1/3}$.

fields
Algebraic Geometry, Number Theory
pdf ↗ page arXiv:2609.17330 ↗

We study the asymptotic behavior of a sequence of Galton–Watson processes with inhomogeneous immigration when the limit of the means of the offspring distributions is less than $1$ or equal to $1$. Under growth conditions on the expected values of the immigration distributions and the variances of the offspring distributions, and assuming the weak convergence of properly scaled immigration processes towards a non-negative stochastic process $\mathcal Y$ with càdlàg or continuous sample paths, we establish functional limit theorems for the sequence of Galton–Watson processes with inhomogeneous immigration in question. The limit stochastic processes can be represented as a constant multiple or an integral functional of $\mathcal Y$ .

comments
50 pages
fields
Probability
pdf ↗ page arXiv:2609.18503 ↗

Rump asked in 2016 whether every finite quasi-linear cycle set is retractible. This was recently proved by Dietzel. In this paper, we reinterpret Rump's question in terms of automorphisms of certain pointed magmas, and propose a more general conjecture on the automorphisms of these magmas. We provide some theoretical and computational evidence in support of our conjecture.

comments
Updated to cite recent proof Rump's conjecture by C. Dietzel. Some typos fixed
fields
Quantum Algebra
pdf ↗ page arXiv:2609.20028 ↗

We introduce a unified framework for supervised and unsupervised learning through density estimation using reproducing kernel Hilbert spaces (RKHSs). To this end, we introduce the notion of a nested RKHS, that is, an increasing sequence of RKHSs whose union is dense in an ambient $L^2$-space. This construction allows us to exploit the computational structure of RKHS methods at each finite approximation level while preserving the approximation power of the entire $L^2$-space. We first study statistical learning in a fixed RKHS. In both unsupervised and supervised settings, we establish strong convergence results for empirical averages of random elements taking values in an RKHS continuously embedded in $L^2$. We then use the nested RKHS structure to combine these statistical convergence results with the density of the union of the RKHSs in the ambient $L^2$-space. We apply this construction to probability density estimation and binary Bayesian classification. For density estimation, we construct explicit empirical estimators of the orthogonal projections of the target density onto the RKHSs and establish their almost sure convergence. A diagonal selection procedure then yields a sequence of estimators converging almost surely to the target density in $L^2$. For binary Bayesian classification, we formulate the approximation of the Bayes target as the minimization of suitable population and empirical risk functionals. We prove consistency of the empirical minimizers and use the nested structure to obtain convergence towards the Bayes target in $L^2$. We further provide a geometric interpretation in terms of hyperplanes and weighted distances in the ambient $L^2$-space. Finally, we show that the population risk admits a direct probabilistic interpretation in terms of the probability of misclassification.

comments
Enhanced bibliography, a few updates, more detailed proofs
fields
Functional Analysis
pdf ↗ page arXiv:2609.20792 ↗

An APN mapping $F:\mathbb{F}_{2^n}\to \mathbb{F}_{2^n}$ is a polynomial characterized by the non-vanishing property on 2-flats. In this work, we analyze notions that are closely related to this property. To understand which $k$-flats of $\mathbb{F}_{2^n}$ remain flats under $F$, we study the $k$-breaking. The function $x^{-1}$ has been studied in the past in this context—we extend this study to general mappings and characterize the 2-breaking of APN functions. Recently, two generalizations of the APN property have been introduced: $k$-strongly non-normality and $k$-th-order sum-freedom. Sum-freedom generalizes the non-vanishing property of APN functions to higher dimensional flats. We provide in-depth observations of the relations between the breaking property, strongly non-normality and sum-freedom. We introduce a fourth concept called $k$-strongly breaking, which implies the breaking property. We derive several structural results for both notions and give a characterization of a subclass of APN functions in terms of the 2-strongly breaking property. We propose a different perspective of the non-vanishing property via a natural character transformation, which is closely related to the sum-of-square indicator of the components of $F$. We derive a precise value for the total sum of the sum-of-square indicators of $F$. With this approach, we provide a simple answer to Open Problem 4 in IEEE Trans. Inf. Theory 52(9): 4160-4170, 2006. Moreover, it allows us to explore balancedness properties of polynomials, one of which characterizes component-wise APNness, for odd $n$, and provides a natural extension to any dimension. We show that Dillon's APN permutation and the Gold functions satisfy a related property, termed $k$-balanced, which is presented under our framework.

comments
30 pages, 2 figures, 3 tables. Example 3.2 has been added. As a consequence, Conjecture 3.1 was modified and renamed as Open Problem 1. The discussion leading to it has been revised. The abstract, introduction, and conclusion are rewritten accordingly
fields
Information Theory, Combinatorics
pdf ↗ page arXiv:2609.22394 ↗

We study signal prediction and smoothing in hidden Markov models whose latent signal is a multi-type Wright-Fisher diffusion and whose discrete-time observations are categorical. We consider a repeated-sampling regime with finitely many inspection times and large within-time samples. Under neutrality and parent-independent mutation, the Bayesian predictive and smoothing laws admit exact representations as mixtures of Dirichlet distributions. We show that the one-step predictive and local smoothing laws converge in total variation to the corresponding Wright-Fisher transition and bridge laws determined by the neighboring population frequencies, thereby justifying plug-in approximations based on empirical proportions. For prediction, we derive explicit finite-sample bounds for this approximation and for bounded predictive functionals such as diversity, together with a finite-sample concentration bound for the marginal smoother. The joint posterior at the inspection times concentrates at the target frequencies and satisfies a Bernstein-von Mises limit with distinct interior and boundary regimes, with Gaussian and Gamma limits on different scales. The results allow arbitrary dependence across inspection times. A key technical ingredient is a fixed-interval tail bound for the coalescent-based dual death process.

comments
To appear on Bayesian Analysis
fields
Statistics Theory, Probability
pdf ↗ page arXiv:2609.22444 ↗

Intelligent reflecting surface (IRS) has emerged as a promising wireless technology for enhancing legitimate communications. In this paper, we investigate an IRS-assisted multiuser MISO downlink network in which an attacker reconfigures the IRS to degrade legitimate data transmission. Specifically, we consider a black-box attack setting in which the attacker has access to neither channel information nor symbol-level observations and can observe only the BER reported by each user. The objective of the attacker is to identify an IRS configuration that maximizes the minimum BER among all users under a limited budget of non-repeated tests. To address this BER-only black-box attack problem, we propose PRIS, an online search method that learns effective phase transitions from previous trials and progressively narrows the search from large-block modifications to single-element refinement. Specifically, an annealed user-balancing reward and prioritized replay are incorporated to exploit noisy BER feedback. Numerical results demonstrate that PRIS achieves a higher minimum BER than existing benchmarks.

comments
5 pages, 4 figures
fields
Information Theory
pdf ↗ page arXiv:2609.23112 ↗

In his seminal work from 1973, Delsarte conjectured that the Johnson schemes admit no nontrivial perfect codes. We prove this fundamental conjecture which is central in combinatorial coding theory. Our arguments combine lower bounds on the design strength of a perfect code with divisibility conditions for the size of a metric ball. We show that the ball size divides the greatest common divisor of specified binomial coefficients. An exact formula for this gcd implies squarefreeness and localizes a power of each prime divisor of the ball size to a short interval. Lloyd polynomials, explicit prime estimates, and bounds for linear forms in logarithms then exclude all sufficiently large radii. Additional arithmetic arguments handle the remaining radii.

fields
Combinatorics
pdf ↗ page arXiv:2609.23128 ↗

We provide a complex-geometric approach to the Blaschke conjecture, i.e., that a manifold whose injectivity radius equals its diameter is isometric to a compact rank-one symmetric space (CROSS). In particular, we introduce the great quadric fibration, a complex analogue of the great sphere fibration. Requiring its total space to be a complex submanifold simultaneously explains many properties that a Blaschke manifold is expected to possess, including its Clifford structure, Hopf fibration, and diffeomorphism class. Furthermore, the great quadric bundle coincides with the variety of minimal rational tangents (VMRT) of the ambient Fano variety in the standard CROSS cases, and we explain this through a bend-and-break argument in the setting of the Blaschke conjecture under a suitable complexification assumption. Combining this with Hwang-Mok VMRT recognition, we prove the Blaschke conjecture for manifolds admitting an adapted complex structure on their entire tangent bundles.

comments
23 pages, minor errors corrected, comments very welcome
fields
Differential Geometry, Algebraic Geometry, Complex Variables
pdf ↗ page arXiv:2609.23822 ↗

We give a new elementary proof of Sullivan's no-wandering-domain theorem for rational maps of degree at least two using hyperbolic area. We then develop this method further to prove that a transcendental entire function has no wandering Fatou components if its set of singular values is compact and all accumulation points of that set lie in the Fatou set. Neither proof uses quasiconformal techniques.

comments
8 pages,1 figure,Revised exposition of the proofs;the discussion of the main ideas has been rewritten and incorporated into the Introduction.comments are welcome
fields
Dynamical Systems, Complex Variables
pdf ↗ page arXiv:2609.23834 ↗

We provide a clearer presentation of the Chair44 monotile, a new strongly aperiodic monotile in three-dimensional Euclidean space $E^3$.

comments
10 pages, 6 figures, one model to assemble
fields
History and Overview, Combinatorics
pdf ↗ page arXiv:2609.24779 ↗

A smooth scalar process with one fixed Gaussian covariance kernel determines any finite graded family of symmetric Hilbert tensors up to a common orthogonal isomorphism of their minimal supports. For two heat tensors, its law determines a closed positive-dimensional Riemannian manifold, a Euclidean bundle, its connection and potential, and the unknown heat time, up to isometry and gauge. The quadratic tensor recovers the scaled generator; the quartic tensor recovers the spatial norm and volume, which fix the time scale. For compact finite-dimensional RCD spaces, the same law determines the measured space with distance divided by the square root of the heat time.

fields
Probability, Differential Geometry, Spectral Theory
pdf ↗ page arXiv:2609.27967 ↗

We prove Shor's conjecture: an orthogonal measurement maximizes the information obtainable from any two quantum states in finite dimension. We then show that an error constraint can require three outcomes. For a class of full-rank qubit pairs, the best orthogonal basis switches abruptly as the error penalty increases. Near the switch, three outcomes yield more information at the same error than any orthogonal measurement or random mixture of them. The extra outcome distinguishes confidence levels in one guess; every outcome is used.

comments
33 pages, 6 figures, 4 tables. Substantially expanded to include information-error trade-offs, a three-outcome measurement advantage, and generalized losses. Title and author list updated. The original version was publicly available on Zenodo on 13 August 2026: https://doi.org/10.5281/zenodo.21911604
fields
Mathematical Physics
pdf ↗ page arXiv:2609.27992 ↗

In 1984, Frank Harary introduced the first convexity games in graphs, all of them based on the geodesic convexity, which is the graph convexity related to shortest paths. In 2024, Araújo et al. obtained the first PSPACE-hardness proofs on some of these geodesic games and generalized them to any graph convexity. In this paper, we investigate convexity games on several known induced path convexities: the monophonic $\mathrm{m}$-convexity and the $\ell_k$-convexities, based on induced paths and on induced paths of size at most $k$. We prove that the hull games $\mathrm{CHG}_{\mathrm{m}}$, $\mathrm{CHG}_{\ell_k}$ and their misère variants are PSPACE-complete for every $k\ge2$ even in graphs with diameter at most 3. We also use the Sprague-Grundy Theory to obtain a polynomial time algorithm to decide the winner of the games $\mathrm{CHG}_{\mathrm{m}}$ and $\mathrm{CHG}_{\ell_k}$ for any $k\ge2$ in disjoint unions of paths and cycles. For $k\ge3$ odd, we prove that Alice (1st player) wins $\mathrm{CHG}_{\ell_k}$ in the path $P_n$ if and only if $n$ is odd and she wins in the cycle $C_n$ if and only if $n=3$ or $n=α\cdot (k+1)-1$ with $α\ge2$. For $k\ge2$ even, the only periodic nimber sequences of $\mathrm{CHG}_{\ell_k}$ obtained through extensive computational testing occurred for $k=2^h-4$ with $h\ge3$, e.g, $k\in\{4,12,28,60,\ldots\}$. In this case ($k=2^h-4$ with $h\ge3$), we prove that the nimber sequences of $\mathrm{CHG}_{\ell_k}$ in $P_n$ and in $C_n$ are periodic and Alice loses (resp. wins) in $P_n$ (resp. $C_n$) with $n>k$ only when $n=3k+4$ (resp. $n\in\{2k+1,5k+3\}$). Finally, we show that, for $k=2$, the game $\mathrm{CHG}_{\ell_2}$ in paths $P_n$ is closely related to the classical game \emph{Couples-are-Forever} of J. H. Conway: it is still an open problem if the nimber sequence is periodic or not and Alice loses only for 12 values of $n$ up to $150$ million.

fields
Combinatorics
pdf ↗ page arXiv:2609.30302 ↗

We prove that, in every closed Riemannian three-manifold $(M^3,\bar g)$, there exists a constant $C=C(M,\bar g)>0$ such that every closed smoothly embedded minimal surface $Σ\subset M$ satisfies $b_1(Σ;\mathbb{Z}_2)\le C\bigl(\operatorname{Ind}(Σ)+\operatorname{Area}(Σ)\bigr)$, establishing Song's additive genus-index-area conjecture. For closed connected orientable two-sided minimal immersions, we prove the universal inequality $γ(Σ)\le8\operatorname{Ind}(Σ)$ in positive Ricci curvature and the estimate $γ(Σ)\le C(M,\bar g)\operatorname{Ind}(Σ)$ in positive scalar curvature. In the latter case, we also bound both area and total curvature by $C(M,\bar g)(1+\operatorname{Ind}(Σ))$.

comments
24 pages. Any comments are welcome
fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2609.32001 ↗

We relate critical diameters of finite configurations to the topology of Vietoris-Rips complexes. For a compact metric space and $0<r<s$, we prove that the canonical inclusion from scale $r$ to scale $s$ is a homotopy equivalence whenever $[r,s)$ contains no diameter of a finite labelled configuration at which the diameter function has zero weak slope. On Riemannian manifolds, weak-slope stationarity implies Clarke criticality. At positive diameter, when the distances realizing the diameter are smooth, both are equivalent to first-order stationarity: the absence of a direction decreasing all these distances to first order. For closed connected smooth manifolds, the Clarke critical diameter spectrum has Hausdorff dimension zero, even at the cut locus. In the real-analytic case, it is finite at each fixed number of labels and countable over all label numbers. On the unit round sphere $S^m$, $m\geq1$, positive nonantipodal weak-slope stationary configurations are characterized by nonzero nonnegative equilibrium stresses. The least positive critical diameter is $\arccos(-1/(m+1))$, and the first accumulation point is $\arccos(-1/m)$. We lift every such stress through a spherical stack construction adapted from Lovász, producing stationary diameters that approach the original value from below in the next dimension as the number of layers increases. We also give example in general metric space showing that weak-slope stationary diameters need not correspond to changes in the homotopy type of Vietoris-Rips complexes.

comments
71 pages, 6 figures
fields
Algebraic Topology
pdf ↗ page arXiv:2609.33009 ↗

We prove a formula for the total variation of SBV functions in sub-Riemannian Heisenberg groups arising as the limit of certain BMO-type functionals.

comments
23 pages
fields
Functional Analysis, Classical Analysis and ODEs, Metric Geometry
pdf ↗ page arXiv:2609.35577 ↗

We construct a smooth projective family of rational surfaces and an effective line bundle on its total space that give a negative answer to Păun's metric-extension question. The restricted bundle on the central fibre carries a semipositively curved singular Hermitian metric with analytic and minimal singularities, but this metric admits no semipositive extension to any neighbourhood of the whole fibre, even up to singularity type. The construction uses four-point blow-ups of the plane, with three centres becoming collinear on the central fibre. This change produces a nonzero integrable adjoint section on the central fibre, whereas the corresponding adjoint systems vanish on nearby noncentral fibres. If a metric extension existed, an $L^2$ extension theorem would extend that section, contradicting this vanishing.

fields
Algebraic Geometry, Complex Variables
pdf ↗ page arXiv:2609.36410 ↗

Let $K$ be an algebraically closed field of odd characteristic $p$, let $G= {\rm SL}_2(K)$, and let $V$ be the natural representation of $G$. Let $b_k$ denote the number of $G$-indecomposable factors of $V^{\otimes k}$, counted with multiplicity, and let $δ_p=1-\log_{p^2}\!\bigl(\tfrac{p+1}{2}\bigr)$. Then there exists a smooth, strictly positive, multiplicatively $p^2$-periodic function $ω(t)$ such that $b_k$ is asymptotic to $ω(k)k^{-δ_p}2^k$. We also show that $t^{-δ}ω(t)$ arises as the limiting density of renormalized convolutions of rescaled copies of a positive weight $3/2$ theta function, obtained from the boundary heat flux of the Dirichlet heat kernel on $(0,p)$.

comments
40 pages v2. Typos corrected. Sections 4 simplified
fields
Representation Theory
pdf ↗ page arXiv:2609.37706 ↗

Smoothed particle hydrodynamics (SPH) is well suited to large-deformation fluid problems, and the growing use of automatic differentiation in engineering applications has motivated interest in differentiable SPH solvers. In this study, a composable differentiable SPH framework, CraftSPH, has been developed. In CraftSPH, major computational operations are implemented as independent modules that users can combine to construct solvers for different purposes. This design supports high-accuracy discretization and both explicit and implicit computation within a common framework, while allowing individual components to be maintained and extended. Selected numerical operations also support automatic differentiation (AD) for physical-parameter estimation and potential coupling with deep learning models. These capabilities were demonstrated through forward and inverse analyses. Forward analyses using an implicit SPH scheme with high-accuracy discretization showed good agreement with analytical and reference solutions for Poiseuille flow and a rising bubble, while capturing the key experimental trends in two- and three-dimensional dam-break problems. In inverse analyses, explicit and implicit SPH schemes were used to estimate kinematic viscosity in Taylor-Green vortex flow and Reynolds number in lid-driven cavity flow. The results also showed that the choice of operations included in the computational graph affects the computed gradients. Together, these findings demonstrate that CraftSPH enables users to assemble problem-specific solvers for forward and inverse analyses within a common framework.

comments
The code is publicly available at https://github.com/GenMNL/CraftSPH 43 pages, 14 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2609.38208 ↗

Closed bosonic lattice systems do not generally admit operator norm Lieb-Robinson bounds uniformly in the initial state. The reason is that the information propagation velocity can grow with the local boson occupancy, which can be macroscopically large. Here, we show that local dissipation restores an operator-norm Lieb-Robinson bound. We consider the dissipative Bose-Hubbard model described by a Lindbladian operator with on-site $\ell$-photon loss, $\ell>2$. For our main result, we establish an almost-ballistic Lieb-Robinson bound. Dissipation rapidly depletes highly occupied sites, thus regularizing the state on the Sobolev-type scale of local particle moments. The resulting moment bounds diverge as time approaches zero but remain integrable near zero at sufficiently low orders, which is crucial for our argument. We extend these ideas to treat cat-code dissipation and, for initial states in the code space, we prove a local adiabatic approximation uniform in the total volume. Further consequences include local channel approximation, a thermodynamic limit, and efficient hybrid bosonic-qubit simulation. The point is that these applications are now available uniformly in the input state, as for quantum spin systems, but in contrast to closed Bose-Hubbard systems.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.01669 ↗

We present sparse dictionary differentiable predictive control (SD-DPC), a framework for learning sparse, interpretable feedback policies for nonlinear systems from data. A prediction model is first identified by rollout-based sparse identification of nonlinear dynamics (SINDy), building on gradient-based and multistep formulations. The policy is then parameterized as a sparse combination of dictionary functions and trained by differentiating a constrained finite-horizon predictive-control objective through this model, so that its terms are selected by closed-loop performance rather than by imitating a previously trained controller. The result is an explicit feedback law with only a handful of terms. Across three benchmark control problems, SD-DPC satisfies the constraints in almost all test scenarios, outperforms a policy distilled onto the same terms by up to an order of magnitude, and requires orders of magnitude less memory and online computation than an optimization benchmark, while admitting explicit sensitivity bounds.

comments
Revised the experimental setup details in Table I and the evaluation reporting, corrected the numerical results in Table II, and clarified the training and validation protocols. The main conclusions are unchanged
fields
Optimization and Control
pdf ↗ page arXiv:2610.02466 ↗

For a positive integer $k$, let $β_k(G)$ be the maximum cardinality of a vertex set inducing maximum degree less than $k$, and let $SF_k(G)$ be the maximum cardinality of a minimal $k$-star-forming set. The known inequality $β_k(G)\le SF_k(G)$ suggests asking when equality holds. We place this question in the framework of upper domination: at $k=1$, $β_1(G)=α(G)$ and $SF_1(G)=Γ(G)$, so the classical equality $Γ=α$ on bipartite graphs is exactly the first member of the proposed hierarchy. We prove the equality for complete bipartite graphs for every $k$, obtaining \[ β_k(K_{a,b})=SF_k(K_{a,b})=\max\{a,b,2k-2\}\qquad(a,b\ge k), \] and record the elementary low-degree case $Δ(G)<k$. For $k=2$ we derive certificate restrictions for minimal $2$-star-forming sets in bipartite graphs. For chain graphs we go further: we prove $β_2(G)=SF_2(G)$ and obtain an exact formula for their common value. The proof uses the nested-neighborhood structure together with a classification of witnesses to the indispensability of a high internal-degree vertex. We retain the equality problem for chain graphs as a conjecture only for $k\ge3$, and formulate the broader bipartite conjecture. We also determine the upper domination number of every rectangular grid and combine it with the known exact dissociation number to compare $Γ$, $β_2$, and $SF_2$. In particular, $Γ=β_2$ on every even-by-even rectangular grid, while $β_2\le SF_2$ always; this motivates a grid equality conjecture whose even-by-even case would yield a three-parameter identity.

comments
Invalid conjectures
fields
Combinatorics
pdf ↗ page arXiv:2610.03785 ↗

A normalized analytic function $f$ defined on the unit disk $\mathbb{D}$ is called $α$-spirallike if $f(\mathbb{D})$ is an $α$-spirallike domain, that is, for every $z\in f(\mathbb{D})$, the $α$-spiral $ze^{-e^{iα}t}$, $t\ge0$, is contained in $f(\mathbb{D})$, or equivalently, if $\operatorname{Re}\!\left(e^{iα} {zf'(z)}/{f(z)}\right)>0$ for all $z\in \mathbb{D}$. The 0-spirallike functions are the usual starlike functions. The function $f$ is parabolic starlike, if $\operatorname{Re}\!\left(\frac{zf'(z)}{f(z)}\right)>\left|\frac{zf'(z)}{f(z)}-1\right|$, and lemniscate starlike if $|(zf'(z)/f(z))^2-1|<1$ respectively. In this paper, we determine the radii of parabolic starlikeness and lemniscate starlikeness of $α$-spirallike functions. Our approach combines geometric characterizations of the underlying regions with the resultant method for eliminating auxiliary variables, leading to explicit equations for the sharp radii. Some known radius results are recovered as special cases.

comments
25 pages, 3 Figures
fields
Complex Variables
pdf ↗ page arXiv:2610.06175 ↗

We construct real-valued solutions of the defocusing energy-supercritical nonlinear wave equation $\partial_t^2u-Δu+u^p=0$ that blow up in finite time from smooth, compactly supported initial data. We consider three cases: for $p=3$ we construct blow-up solutions in all dimensions $d\geq 9$, for $p=5$ we treat all dimensions $d\geq 7$, while for $p=7$ we treat all dimensions $d\geq 6$. The solutions we construct here appear to be the first blow-up solutions of defocusing energy-supercritical nonlinear wave equations with smooth real-valued data. The main step is the construction of nonzero \emph{discretely self-similar profiles} that are periodic in self-similar time and real-analytic across the light cone. We find approximate profiles by numerical continuation from rotating self-similar solutions of the complex-valued equation and upgrade them to exact solutions by a computer-assisted Newton–Kantorovich argument in a Banach algebra of Fourier–Chebyshev coefficients. The main analytic ingredients are an exact description of the linear mode operators and uniform bounds for their inverses. The codes for the computer-assisted parts of the proofs, together with the approximate solutions, are available on GitHub \cite{code}.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.06678 ↗

We build on EndoWake, an endogenous wake model that links the wind speed at every grid cell to its upwind neighbors by linear constraints, so that wake field, siting decisions and project constraints share one mixed-integer model. Its four parameters are calibrated once on the single wake of a Reynolds-averaged Navier-Stokes (RANS) simulation, and a graphics processing unit (GPU) scores over a million layouts per second per wind direction. At Lillgrund, we compare searches guided by EndoWake and by PyWake under a RANS judge that no search calls: the fidelity of a wake model is not the quality of the layouts optimized under it. Two diagnostics separate a quadrature error between the sampled wind directions from a surrogate bias localized off each wake edge, not reported before to our knowledge; at random directions, five paired starts detect no difference between the pipelines. The main result puts the speed to use: EndoWake ranks the moves of a local search on the GPU, and PyWake confirms every accepted move. On twenty unseen starts, with one GPU and thirty CPU processes, screen-and-confirm reaches similar mean gains to a search on PyWake alone, 7.9 times faster at one wind speed and 14.9 times faster on the annual energy (3.0 and 4.7 with a proof of local optimality charged to the timed search); on a second site it keeps its speed, and its local optima fall slightly short of those of PyWake alone. Code and data are deposited on Zenodo.

comments
v2, revised (October 2026): formal terms replace the metaphors of v1; new comparisons with TOPFARM and on a second site (Horns Rev 1); sensitivity to the length of the candidate list; speedups with and without the final certificate. The replication package (doi:10.5281/zenodo.22017446) now includes the local searches, the sector quadrature and the fast evaluator. 39 pages, 5 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.07844 ↗

We establish an Alexandrov–Fenchel type inequality in which one of the reference bodies is replaced by a function whose Hessian satisfies a balancing condition on its eigenvalues, without being required to be positive semidefinite. The proof hinges on a new extension of Alexandrov's inequality for mixed discriminants. As applications, we derive a log-concavity principle for linear functionals of area measures, answering a question of Colesanti, Hug, and Saorín-Gómez, as well as Brunn–Minkowski type inequalities for intrinsic volumes of mean section bodies, answering a question of Schuster.

comments
29 pages
fields
Metric Geometry, Functional Analysis
pdf ↗ page arXiv:2610.08746 ↗

Let $X$ be a complex $\mathbb Q$-factorial Gorenstein Fano fourfold with $ρ(X)=1$. If $X$ has canonical singularities, we prove that $(-K_X)^4\le3528$, with equality if and only if $(X\simeq \mathbb P(1,1,12,28,42)$. In fact, $(-K_X)^4\ge3130$ already implies this isomorphism. If $X$ has terminal singularities, we prove the sharp bound $(-K_X)^4\le648$, with equality if and only if $X\simeq \mathbb P(1,1,1,1,2)$. In the canonical case, we also obtain the sharp bound $(-K_X)^4\le1024$ when a positive generator of $Cl(X)/ \mathrm{tors}$ is Cartier outside a finite set. The proofs use jet estimates involving all Harder–Narasimhan slopes of the tangent sheaf, the geometry of algebraic foliations, and the integrality of intersection numbers on a general leaf. The equality case in the canonical bound is determined by two computations of divisorial section rings.

comments
37 pages; v2; we also obtained the sharp bound in the canonical case
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.09419 ↗

A $λ$-self-expander $F: M^n\to \mathbb{R}^{n+1}$ is the solution of the isoperimetric problem of weight $e^{\frac{|x|^2}{4}}$. In this paper, for sufficiently large $λ$, we construct an immersed non-embedded spherical $λ$-self-expander in $\mathbb{R}^{n+1}$ with the mean curvature $H>0$. Moreover, we obtain the explicit solution for $n=2$.

comments
32 pages Version 2: Minor correction, fixed misspelling of Florian Johne's surname in the text and references
fields
Differential Geometry
pdf ↗ page arXiv:2610.10634 ↗

We show that $m$-linear homogeneous rough singular integral operators $\LL_Ω$ associated with $L^\infty$ functions $Ω$ on $\mathbb{S}^{mn-1}$ with mean value zero may not be bounded from $L^1(\mathbb R^n)\times\cdots\times L^1(\mathbb R^n)$ to $L^{\frac 1 m,\nf}(\mathbb R^n)$ except possibly for the case $n=1$ and $m=2$.

fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.10856 ↗

Let $n,k$ be positive integers and $q$ be a prime power. We prove that there exists a real constant $C_k$ depending only on $k$ such that if $ \# M_n(\mathbb{F}_q) > C_k$ then every element in $M_n(\mathbb{F}_q)$ is a sum of two $k$th powers, proving Larsen's conjecture.

fields
Group Theory
pdf ↗ page arXiv:2610.17698 ↗

Birkhoff pseudospectral methods replace the differentiation matrices of classical transcriptions with integration operators whose conditioning does not deteriorate with polynomial order. This paper presents theoretical and computational advancements toward their practical use in aerospace optimal control. The initial-point-anchored Birkhoff integration operator is constructed explicitly for flipped Legendre-Gauss-Radau grids and related exactly to the corresponding Lagrange differentiation matrix and, through a reflection, to the standard Radau Birkhoff operator, whose conditioning properties it inherits. Rate constraints on nonlinear functions of states and controls are enforced directly through the pseudospectral operators and, for control-rate limits, shown to be primal and dual equivalent to state augmentation. Non-affine time mappings concentrate the nodes around localized trajectory features, and composing affine scaling and equilibration transformations provides a systematic way to construct scaling strategies. The methods are implemented in the open-source MATLAB framework RAPTOR and tested across a range of challenging optimal control problems, including a dense-grid orbit-raising problem, a rate-limited landing scenario, a Space Shuttle entry problem, and a five-phase reusable-launch-vehicle mission.

fields
Optimization and Control, Dynamical Systems
pdf ↗ page arXiv:2610.17842 ↗

We introduce a class of extensions of unital operator systems satisfying a certain Toeplitz type property and show that these extensions induce corresponding extensions of the $C^*$-algebras they generate. We prove that an extension in this class is hyperrigid in the $C^*$-algebra it generates if and only if both operator systems from which it is constructed are hyperrigid in the $C^*$-algebras they generate. We then introduce the quantum double suspension of operator systems as a special case of this construction and show that its iterates yield the Hong-Szymański quantum double suspensions at the $C^*$-algebra level, including the Vaksman-Soibelman quantum spheres. In particular, we construct hyperrigid operator systems that generate the $C^*$-algebras of these quantum spheres. As a further application, we show that every counterexample to Arveson's hyperrigidity conjecture gives rise to a sequence of counterexamples. We identify the Šilov ideal and the $C^*$-envelope of these extensions and derive a $K$-theoretic condition under which each such sequence contains infinitely many pairwise non-isomorphic counterexamples. In particular, Scherer's counterexample gives rise to infinitely many pairwise non-isomorphic counterexamples. Finally, we compute the propagation number, an invariant of operator systems introduced by Connes and van Suijlekom, for this class of extensions.

comments
Minor modification due to the very recent commutative counterexample that appeared in arXiv:2610.26556. A few typos corrected, references added
fields
Operator Algebras
pdf ↗ page arXiv:2610.18129 ↗

The Sachdev–Ye–Kitaev (SYK) model is a quantum many-body model of Majorana fermions with all-to-all random interactions. It has attracted wide interest in condensed matter and high-energy physics as a solvable model of non-Fermi liquids and holography. Fix an even integer $q\geq4$, and let $Z_N(β)$ be the partition function of the Gaussian $q$-body SYK model. We prove that for all sufficiently small inverse temperatures $β> 0$, there exists $ρ=ρ(β)>0$ such that \[ N^{q/2 -1}(\log Z_N(β)-\mathbb{E}[\log Z_N(β)]) \Longrightarrow \mathcal{N}(0,ρ^2). \] In addition to the mathematical results, we discuss how the proofs were developed and refined through interaction between human researchers and AI, and we describe the subsequent formalization of the main arguments in Lean 4. This work thus serves as a case study of a research workflow that integrates human insight, AI-assisted exploration, iterative proof refinement, and formal verification.

code
github.com/TokyoProbabilityGroup/SYK_CLT ∎ Lean project (Blueprint/lakefile.toml, Blueprint/lean-toolchain) builds, no sorry at 89f791e, checked 2026-10-28 build details
fields
Probability, Mathematical Physics, History and Overview
pdf ↗ page arXiv:2610.18440 ↗

Motivated by recent advances in the synthesis of effective one-dimensional mechanical metamaterials, we develop a direct theory of strain-gradient special Cosserat rods based on the principle of virtual work. The internal virtual work is formulated intrinsically in terms of variations of the classical Cosserat strains and their spatial derivatives, without requiring a three-dimensional parent material or being restricted to elastic materials. Objectivity and insensitivity to rigid virtual displacements characterize the admissible internal-work density, while localization yields the dynamic field equations, higher-order contact interactions, natural boundary conditions, and balances of linear and angular momentum. For inextensible and unshearable rods, the constrained principle determines a unique multiplier functional, although its representation by reactive force and double-force fields possesses a gauge freedom absent from classical rod theory. We also formulate objective elastic, hyperelastic, and viscoelastic constitutive laws, extend transverse hemitropy, transverse isotropy, and flip symmetry to strain-gradient rods, and derive invariant representations of symmetric stored energies. As an application, we analyze a planar inextensible cantilever subjected to a compressive terminal force, with stored energy depending on the spatial derivative of curvature. We prove that the eigenvalues of the linearized problem are simple and coincide precisely with the loads at which bifurcation from the straight state occurs. In addition, each bifurcation point generates a connected global branch of nonlinear equilibria that is locally analytic near its origin.

comments
30 pages, 2 figures, 1 table; v2: added reference, minor corrections
fields
Mathematical Physics
pdf ↗ page arXiv:2610.19171 ↗

This paper studies an infinite-horizon equilibrium dividend problem for a firm whose surplus follows a diffusion with observable Markov regime switching and is absorbed at bankruptcy. Dividend payments are modeled as singular controls, and non-exponential discounting gives rise to time inconsistency. We formulate equilibrium control laws and establish a verification theorem for a general finite-state model with surplus-dependent coefficients. For a two-state model with constant positive drifts and volatilities and bi-exponential discounting, we prove the existence of a regime-dependent equilibrium barrier strategy for arbitrary positive switching intensities. The proof relies on uniform estimates for a system of coupled boundary-value problems and the Poincaré-Miranda theorem. We also verify that the resulting strategy satisfies the equilibrium variational inequalities. Numerical results show that greater dispersion in discount rates, with the initial average discount rate held fixed, raises both barriers. The effect of switching intensity can be nonmonotone and may reverse the ordering of the barriers.

fields
Optimization and Control
pdf ↗ page arXiv:2610.19655 ↗

Radical parametrizations describe algebraic varieties by means of nested radicals of polynomials, and their theory relies on three varieties associated with a parametrization: the incidence, tower and radical varieties. In this paper we revisit this framework from a computational point of view and show that it is governed by the radical tower rather than by the particular parametrization. We prove that the tower variety depends only on the tower and that the incidence variety is always irreducible, so that no prime decomposition is needed. We show how to replace a tower by an isomorphic one whose tower variety is defined by the equations of the tower, so that it can be computed without elimination. We introduce universal radical parametrizations and we use them to settle, under a natural hypothesis on the coefficients, an open case in the transformation of differential equations with radical coefficients into equations with rational coefficients. We show that radical subtowers form a lattice, which yields minimal radical parametrizations, and we give algorithms to compute them. We also prove that dominance between towers induces a unique dominant rational map between their tower varieties, through which unirationality is transferred. In addition, we study towers over subfields of $\mathbb{C}$ and, in the real case, relate the real points of the tower variety to the real domains of the associated parametrizations. Finally, we present computational experiments which show that our methods succeed in many cases where the previous approach does not finish within a reasonable time.

comments
46 pages, 3 figures. Scripts and data of the computational experiments: https://doi.org/10.5281/zenodo.23623629
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.20404 ↗

In this article, we provide a complete answer to one of the most classical conjectures in differential geometry (of foliations). More precisely, we present a complete positive answer to the classical conjecture, due to Meeks III, Pérez, and Ros (2008), that states that every foliation on $\mathbb{R}^{n+1}$ by (possibly varying) CMC hypersurfaces is a foliation by minimal hypersurfaces. Moreover, if $n\leq 6$, such a CMC foliation must consist of parallel hyperplanes.

comments
Some typos were fixed. 21 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.20406 ↗

In this article, we study the fractal refinement of the Carleson's problem: (1) We introduce an equivalence relation on linear dispersive PDEs on $\mathbb{R}^{n}$ based on the high-frequency oscillation of their Fourier multipliers, proving that equations in the same class share the same Hausdorff dimension of divergence sets. As applications, we estimate these dimensions for abstract dispersive families. Consequently, we recover known $1$-dimensional results for linearized Boussinesq and Beam propagations, while establishing new results for the both of them in higher dimensions and for Klein-Gordon propagations in all dimensions. (2) Next, we study the effect of radial symmetry by obtaining dimension estimates for fractional Schrödinger equations with concave phase and radial initial data. Applying a transference principle, dimension estimates are extended to families of abstract dispersive equations. As special cases, we recover the known result for the Beam propagation and also obtain new results for the linearized Boussinesq and the Klein-Gordon propagations in the radial setting. (3) We study analogues of (2) for the $3$-dimensional real hyperbolic space by obtaining dimension estimates for a wide class of dispersive propagations: the Schrödinger, fractional Schrödinger (with convex and concave phases), Beam, linearized Boussinesq, Wave and Klein-Gordon equations. This is the first result of its kind, outside the setting of flat Euclidean geometry. Our arguments are based on an inter-spatial transference principle of geometric-analytic flavour. In the process, we also obtain quantitative results on the singular sets of Sobolev functions, in the spirit of Žubrinić, which may be of independent interest.

comments
Fixed some minor typos
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.20480 ↗

In this paper, we show that complete steady gradient Yamabe solitons are either rotationally symmetric manifolds or scalar-flat manifolds. We also classify nontrivial complete shrinking gradient Yamabe solitons.

comments
23 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.21078 ↗

Permutation polynomials over the Galois ring $\mathrm{GR}(2^k,m)$ are nonlinear components for symmetric primitives that add keys in the ring, as in secure computation modulo $2^k$. Rønjom and Sandrib showed that their differential uniformity is at least $2q^{k-1}$, where $q=2^m$. For the class studied here it is attained on top-layer differences, while almost all differences are units. We ask how small the difference table can be on unit differences when the residue permutation is fixed and only higher coefficient digits of its lift are chosen. Let $m\ge5$ be odd, $k\ge3$, and let $p$ be a reduced quadratic almost perfect nonlinear (APN) permutation of $\mathbb{F}_q$ whose derivative is nowhere zero and two-to-one. Over all lifts of $p$, the least possible maximum on unit differences is $q$ or $2q$, decided by two invariants of $p$ and the level $k$. We describe every lift attaining it and show that the coefficients modulo $8$ decide optimality at every level. For every odd $m\ge5$, residue permutations with at most $16$ monomials attain $q$ after one $2$-adic digit of one coefficient of the natural lift is changed. For an explicit infinite family, one added monomial lowers the maximum on unit differences from $4q$, or from at least $q^2$, to $2q$ at every level without raising the degree. Under a uniform starting state and independent uniform round keys, valuation preservation yields layerwise bounds for characteristics and two-round differentials that distinguish lifts with the same global uniformity. Core results are formalized in Lean 4.

code
github.com/hirokifukui/residue-apn ∎ Lean project (supplement/lean/lakefile.toml, supplement/lean/lean-toolchain) builds, no sorry at 4702842, checked 2026-10-28 build details
comments
25 pages. Version 2 is a substantially revised and extended paper with a new title. Lean 4 formalization, blueprint and finite computations: https://doi.org/10.5281/zenodo.23683493 (version 2.0.0; https://github.com/hirokifukui/residue-apn)
fields
Number Theory, Information Theory
pdf ↗ page arXiv:2610.21178 ↗

We investigate the recovery of an unknown number of stationary point heat sources, together with their locations and constant positive strengths, from temperature histories recorded by a few sensors. The reconstruction combines candidate selection by orthogonal matching pursuit or least angle regression with continuous constrained fitting. A noise-based discrepancy rule selects the smallest accepted source count among those tested. The analysis distinguishes source-count selection, individual-parameter recovery, and aggregate estimation. For a close pair, the local single-source residual is quadratic in separation when the projected curvature is nonzero, whereas a compensating unit parameter direction produces a linearized response bounded by a constant times the cube of separation. This difference explains why a pair may be distinguishable from a single source while its individual parameters remain poorly determined. A conditional stability bound controls errors in total strength and the strength-weighted center, and a local Gaussian approximation relates single-source rejection to noise and separation. Numerical experiments assess sensor configuration, temporal sampling, noise, separation, and initialization, with additional tests using independently generated finite-plate data and a three-dimensional model. Correct source-count selection need not imply accurate individual parameters, whereas unresolved pairs can retain accurate aggregate estimates.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.21301 ↗

We study branching Brownian motion in a resource environment created by the particles themselves. Particles reproduce at rate $β$ times the resource nearby, resource is consumed at rate $c$ times the number of particles nearby, and each death returns resource mass $m$. Let $θ=βm/c$, which represents the number of births supported by resource from one corpse, assuming perfect efficiency. We prove that the population dies out almost surely if $θ\leq1$, even with perfect recycling $\left(θ=1\right)$. If $θ>1$ and the interaction kernel has finite range and is bounded below at short distances, the population survives with positive probability from every initial state with at least two particles, or with one particle and some resource. The threshold does not depend on the diffusivity, the death rate or the interaction kernel. We prove extinction for $θ\leq 1$ by showing the total mass in the system to be a supermartingale. The proof of survival rests on two estimates: a large colony grows if it consumes its corpses with little waste, and a dense colony recycles corpses efficiently. Together these imply that a large, dense colony grows with exponentially high probability, and a standard doubling and restart argument then yields survival.

comments
12 pages, no figures
fields
Probability
pdf ↗ page arXiv:2610.21476 ↗

We study over-the-air computation of functions, such as empirical variance, formed by combining two separately postprocessed sums of locally transformed observations. We consider continuous encoders subject to pointwise peak-power constraints over a single use of a complex multiple-access channel. With four or more transmitters, we identify conditions under which exact noiseless recovery forces every continuous encoder collection into a common affine structure that preserves both sums, reducing infinite-dimensional encoder design to selecting a finite set of coefficients. To quantify resilience against channel noise, we introduce a robustness score that measures the separation of noiseless received aggregates relative to variations in the desired function value. Within this characterized encoder family, we formulate robustness maximization as an exact convex semi-infinite program, eliminating rank constraints without a relaxation gap, and construct a decoder that attains the minimum uniform noise-amplification constant. For several function classes, including empirical variance, we derive closed-form robustness-optimal encoders and scores. Under mild regularity conditions, we also establish sharp limits on the Gaussian mean squared error at high signal-to-noise ratios and show how using the input distribution reduces average error at the expense of uniform robustness.

fields
Information Theory, Optimization and Control
pdf ↗ page arXiv:2610.21536 ↗

The direct methods of calculus of variations have had a great success in the classical Newtonian $N$-body problem in this century. First in finding periodic solutions and more recently in the existence of parabolic, hyperbolic and hyperbolic-parabolic solutions. The foundation of many of these results is Marchal's lemma, which guarantee's fixed-end local minimizers are collision-free and hence classical solutions of the $N$-body problem. Meanwhile almost no result is available in the restricted problem through the direct methods of calculus of variations, mainly due to the lack of a Marchal-type lemma. In this paper such a lemma and its variation will be obtained for the restricted problem. We then apply them to the restricted problem to study unbounded (hyperbolic, parabolic, bi-hyperbolic and bi-parabolic) solutions, and bounded (periodic and quasi-periodic) solutions. We emphasize that the extension to the restricted case does not follow from the standard arguement for the unrestricted case.

comments
30 pages, 3 figures with some typos fixed. Comments are welcome!
fields
Dynamical Systems
pdf ↗ page arXiv:2610.22849 ↗

We prove that the ordered GIT quotient of seven points on $\mathbb P^1$ with equal weights is not isomorphic to the Baily–Borel compactification of any arithmetic complex ball quotient. Together with our companion paper and the known positive examples, this determines the equal weight case for every $n\ge6$: such a realization exists precisely for $n=6,8$. The lattice and the algebraic isomorphism are unrestricted, and torsion and divisorial ramification are allowed. The proof combines arithmetic reductions with identities involving Chern classes of fixed subvarieties and intersection inequalities for nef divisors on the GIT quotient. The final comparison includes all fixed components and requires no bound on the degree of an individual fixed curve.

comments
44 pages
fields
Algebraic Geometry, Complex Variables, Number Theory
pdf ↗ page arXiv:2610.23227 ↗

We prove that, for every $n\ge9$, no ordered GIT quotient of $n$ points on $\mathbb P^1$ with positive rational weights smaller than one and of total weight two is isomorphic to the Baily–Borel compactification of an arithmetic complex ball quotient. The weights may lie on stability walls, and the lattice may have torsion and ramify along divisors. No compatibility with a prescribed period map is assumed. The proof compares homomorphisms from the tangent bundle to line bundles of negative degree with normal bundles of arithmetic fixed subvarieties. We describe these homomorphisms through coincident marked points and prove vanishing statements for their kernels. The comparison, together with integrality of Chern classes and the invariant theory of finite reflection groups, gives the nonexistence theorem without a Fano assumption on the GIT quotient.

comments
57 pages
fields
Algebraic Geometry, Complex Variables, Number Theory
pdf ↗ page arXiv:2610.23228 ↗

We classify boundary blow-ups of minimizers of the Mumford-Shah functional on bounded planar domains with Dirichlet boundary conditions. At every point where the discontinuity set meets the boundary of the domain, we prove that the discontinuity set of each blow-up is the tangent line to the boundary at that point. This settles the conjecture formulated by the author in 2024.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.23513 ↗

Let $\mathfrak{GH}$ denote the Gromov–Hausdorff class of isometry classes of nonempty metric spaces, equipped with the Gromov–Hausdorff extended pseudometric. We prove that every nonempty extended metric space is isometrically embeddable into $\mathfrak{GH}$. More precisely, for every nonempty extended metric space $X$, all image points may be represented by metrics on a common set of cardinality $\max\{\operatorname{dens} X,\aleph_0\}$. The main ingredient is a realization theorem for function spaces: for every nonempty set $I$, the space of all real-valued functions on $I$, endowed with the uniform extended metric, can be realized as a family of metrics on a fixed underlying set. Under this realization, the Gromov–Hausdorff distance agrees exactly with the uniform extended distance, including the case of infinite distance. The resulting metric spaces are complete, uniformly discrete, and unbounded, and have cardinality $\max\{|I|,\aleph_0\}$. If $I$ is countable, they may moreover be chosen proper and of bounded geometry. As further consequences, we obtain sharp cardinality results for clouds and universality results for cardinality-restricted parts of $\mathfrak{GH}$ and for spheres in the associated zero-distance quotients.

comments
19 pages
fields
Metric Geometry
pdf ↗ page arXiv:2610.24279 ↗

We prove local rigidity of smooth conformally Euclidean metrics $g_0$ on smooth compact domains $M\subset\mathbb{R}^n$, $n\ge3$, for the anisotropic Helmholtz equation $(-Δ_g -k^2)u =0$ at a fixed energy $k^2>0$ which is not a Dirichlet eigenvalue of $-Δ_{g_0}$ in $M$. Any smooth metric sufficiently close to $g_0$ in $H^s(M)$, with integer $s>n/2+1$, and having the same Dirichlet-to-Neumann map, is its pullback by a smooth diffeomorphism fixing the boundary pointwise. The proof uses a Helmholtz embedding to fix the gauge and a compactness argument together with a characterization of the kernel of the linearization.

comments
The introduction has been improved
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.25314 ↗

We show that the geometric control condition (GCC) does not guarantee internal observability for a wave equation with a smooth time-dependent potential. On a bounded connected domain with analytic boundary, we construct a nonempty open observation set such that the pair satisfies GCC and the free Dirichlet wave is observable. For this same domain and observation set, we construct a smooth $2π$-periodic potential for which there exists a nonzero smooth Dirichlet wave that vanishes on the observed cylinder for all time. The wave is the unique mild solution in the space $X_Ω=L^2(Ω)\times H^{-1}(Ω)$ on every finite time interval, so no finite observation horizon yields an observability inequality for this fixed potential. The construction starts from a real-valued local counterexample to unique continuation across a noncharacteristic hyperplane. This local statement is of independent interest, since earlier constructions for the wave operator use complex-valued waves and potentials. The global counterexample is then obtained by joining flat oscillatory solutions along the edges of a polygonal observation set, repairing their sum at the vertices without changing its zero set, and using a spatial transition to a common time-independent pair near the boundary. The planar construction also yields counterexamples in every space dimension $n\ge3$. For comparison, and to identify a geometry that supplies both propagation and continuation, we show on the unit disk that an observation region containing a central disk and a neighborhood of an embedded cross-cut yields observability for every weak wave at sufficiently large observation time, with constants uniform over bounded measurable potentials. The proof uses a cut, logarithmic polar coordinates, and a Carleman weight whose favorable flux is confined to the observed artificial boundary.

fields
Optimization and Control, Analysis of PDEs
pdf ↗ page arXiv:2610.25327 ↗

Indirect and noisy observations are difficult to invert when the forward operator smooths out the unknown signal. In this case, the unstable modes are precisely the modes that the data reveals least, thus the success of any reconstruction method depends on how prior information interacts with the geometry of the operator across regularity scales. Diffusion priors have become ubiquitous for such problems; however, the current theory still does not determine when a learned score remains informative under ill-posed forward operators. In this paper, we study posterior diffusion for linear inverse problems on Hilbert scales and derive a stability theory under an explicit score-amplification condition. Under a smoothing condition on the forward map, a source condition on the truth, and weak-norm control of score error, we derive non-asymptotic recovery bounds in \(\Hs^s\) that separate observation noise, score approximation, and reverse-time discretization. These bounds give a recoverable regularity threshold determined jointly by operator smoothing, truth regularity, score regularity, and the posterior correction. The threshold specifies the Hilbert-scale norms \(\Hs^s\) in which weak-norm score error remains controlled after propagation through the posterior reverse dynamics. For larger regularity indices \(s\), the inverse geometry magnifies the same error, and the Hilbert-scale recovery estimate no longer gives uniform stability. We also prove that a posterior correction adapted to the forward operator improves stability relative to isotropic guidance in the regularity ranges where the smoothing action of the forward map matters most. The recovery estimates give a regularity-sensitive theory of posterior diffusion and show how learned score error, operator smoothing, and reconstruction norm jointly determine the stable range predicted by the analysis.

journal
2026 Asilomar Conference on Signals, Systems, and Computers
fields
Information Theory
pdf ↗ page arXiv:2610.27030 ↗

It is known that all group codes of dimension at most $3$ are permutation equivalent to abelian group codes. We prove that twisted group codes of dimension at most $3$ are monomially equivalent either to an abelian group code or to an irreducible constacyclic code, and in the second case we describe the code explicitly by means of a trace map. Monomial equivalence cannot be replaced by permutation equivalence, and the second case cannot be avoided: the ternary tetracode is an ideal of $\mathbb F_3[x]/(x^4+1)$ but it is not monomially equivalent to any group code. More generally, we give two infinite families, in dimensions $2$ and $3$, of irreducible constacyclic codes which attain the Griesmer bound, whereas every group code with the same length and dimension has smaller minimum distance.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.28297 ↗

This book has several fundamental aims, one of which is to serve as a reference book for computing unstable homotopy groups using methods of the Toda school. The Toda bracket serves the art of constructing homotopy liftings and extensions of maps. It plays a fundamental role in dealing with composition relations of homotopy classes. This book brings together in Appendix A.3 all known results on the $2$-primary components of unstable homotopy groups of spheres obtained using the art of Toda brackets as the principal tool, with explicit generators. It also explains in Appendix A.2 the conventions used to name the generators. The book mainly presents calculations of the $2$-primary components of the homotopy groups $π_i(\mathit{SO}(n))$ for all $23\leq i\leq 28$ and all $n,$ as well as some $π_{29}(\mathit{SO}(n)),$ with explicit generators, while retaining two possible group structures for $π_{28}(\mathit{SO}(11)).$ In degree $29,$ the cases $7\leq n\leq14$ remain under investigation. The odd-primary components of the homotopy groups $π_i(\mathit{SO}(n))$ in the above range can be computed with little effort; these computations essentially amount to looking up the relevant entries in existing tables.

fields
Algebraic Topology
pdf ↗ page arXiv:2610.28447 ↗

The classical Newton-Maclaurin's inequalities for the elementary symmetric means are well known. In this paper, we derive a new class of Newton-type inequalities and generalize Maclaurin's inequality by connecting the elementary symmetric functions with their counterparts. Our idea is mainly motivated by the inclusion relations of G$\mathring{\text{a}}$rding's cones and $k$-convex cones.

fields
Classical Analysis and ODEs, Combinatorics
pdf ↗ page arXiv:2610.28544 ↗

Pinching antenna systems (PASS) provide flexible transmitter-side radiation reconfiguration for high-frequency wireless communications. This paper applies PASS to non-orthogonal multiple access (NOMA) covert communication networks, where activated pinching antennas (PAs) transmit superimposed signals to a pair of users stochastically distributed within a rectangle. Two passive wardens respectively monitor the downlink transmissions intended for two users. More specifically, we derive the closed-form detection error probability (DEP) expressions for two wardens under bounded noise uncertainty. Then we analytically determine the optimal detection thresholds and minimum DEPs in the worst-case covert detection scenarios. The covert transmission rates of two users are further characterized by taking into consideration ideal successive interference cancellation and non-ideal successive interference cancellation at near user and intra-NOMA interference at far user. Numerical results are presented to confirm that: 1) PASS-NOMA is capable of providing higher covert rates relative to conventional orthogonal multiple access OMA schemes; 2) As the number of activated PAs begun to increase, the near user has ability to obtain a larger covert rate gain than far user; and 3) The DEP of PASS-NOMA networks is larger than that of both PASS-OMA and traditional antenna based NOMA in specific output power regions.

fields
Information Theory
pdf ↗ page arXiv:2610.29433 ↗

We recover a shared constitutive generator from equilibrium locations on weighted graphs. For an even $C^2$ log-coordinate potential with positive curvature, the complete two-site response map determines the generator up to positive scale. A sweep over all positive boundary values at one fixed, possibly unknown, unequal weight ratio gives global identification through a Schröder-Koenigs iteration. We extend this result to a three-vertex chain with two observed free vertices under a sweep condition. For variable-ratio control, we prove finite-target recovery, local Lipschitz stability, and noisy-bisection bounds. Fixed-ratio stability and nonparametric estimation under uncertain controls remain open. In the reciprocal family $J_p$, an exact Gaussian model-selection probability benchmarks synthetic experiments under observation noise and errors in variables. These are identifiability results under controls; no measured data or physical system is analyzed.

comments
36 pages, 4 figures, 3 tables. Supplemental material (19 pages) and reproducibility code included as ancillary files. v2: Substantially revised, with a new fixed-ratio identification theorem, a two-free-vertex graph extension, finite-query and stability bounds, expanded literature comparisons, and numerical validation
fields
Optimization and Control
pdf ↗ page arXiv:2610.30146 ↗

Let \[ D(f,h)=\sum_{k=1}^n a_k f(b_kh) \] be a finite-difference operator, where the coefficients may be complex and the nodes are real. We study the asymptotic implication \[ D(f,h)=o(h)\quad\Longrightarrow\quad f(h)=o(h) \] for Lebesgue measurable functions. In 2017, Ash, Catoiu, and Csörnyei \cite{ACC} showed that this implication holds for every measurable function precisely for operators belonging to a very restricted class of finite differences. To establish that every operator outside the exceptional class admits a counterexample, Ash, Catoiu, and Csörnyei analyze an infinite system of linear equations. Their argument is therefore existential in nature and does not provide an explicit description of the counterexamples. By contrast, we give a substantially simpler and elementary construction based on zeros of finite Laurent polynomials and multiplicative characters on finitely generated multiplicative subgroups. This provides a direct recipe for constructing measurable functions \(f\) such that \[ D(f,h)=0\qquad\text{for every }h\in\mathbb C, \] while, in fact, \[ f(h)\not=o(1)\qquad\text{as }h\to0. \]

fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.30583 ↗

For $p\geq2$ and $r\geq 1$, we prove the sharp strong $r$-variation estimate for solutions to the free Schrödinger equation $$ \bigl\|V^r_{[0,1]}\{e^{itΔ}f\}\bigr\|_{L^p(\mathbb R^2)} \lesssim \|f\|_{H^s(\mathbb R^2)}. $$ The estimate holds if $s> s(p,r)$ and fails if $s< s(p,r)$, where $$s(p,r)=\max\left\{ \frac{2}{r}+\frac{1}{p}-\frac12,\, \frac{2}{r}-\frac{2}{p},\, \frac{1}{p}+\frac{1}{r},\, \frac{2}{p}-\frac13,\, 1-\frac{2}{p} \right\}.$$ In particular, $s(3,\infty)=1/3$. Thus our result yields a sharp variational extension of the two-dimensional Schrödinger maximal estimate of Du, Guth, and Li [{Ann. Math.}, 186, 607–640 (2017)]. The proof employs fractal restriction estimates to eliminate the regularity loss originating from the summation in the variation norm.

comments
35 pages, We have revised the proofs of Lemma 3.1, modified the definition of $g_{R,j}$ in Theorem 4.1 and corrected some typographical errors
fields
Classical Analysis and ODEs, Analysis of PDEs
pdf ↗ page arXiv:2610.31099 ↗

An integrable elliptic generalization of the closed $D^{(2)}_2$ quantum spin chain is constructed. Taking the trigonometric limit, we have shown that the corresponding R-matrix reduces to that of the $D^{(2)}_2$ one, up to a similarity transformation. Basing on the very fact of the factorization for the R-matrix and the corresponding transfer matrix which indeed ensures the integrability of the model, we give the exact solution of the model which can be expressed in terms of a homogeneous T-Q relation.

comments
18 pages,1 table
fields
Mathematical Physics
pdf ↗ page arXiv:2610.31381 ↗

Let $k$ be a number field. We prove that for every quadratic extension $F/k$, every real zero $β\in(0,1)$ of $L(s,χ_{F/k})$ satisfies $(1-β)\log d_F\ge c(k)$ with an effective constant $c(k)>0$, and we deduce (through a reduction of Stark) that the Brauer-Siegel theorem holds effectively and uniformly over all Galois extensions $K$ of $k$, in the form $h_K R_K \ge c(k)\, d_K^{1/2}/((2π)^{[K:\mathbb{Q}]}\log d_K)$. Stark's effective cases of the Brauer-Siegel theorem excluded the fields with quadratic subfields, since those subfields carried the ineffectivity of Siegel's theorem; over a base field $k$, the first result removes this obstruction. Both rely on a uniform bound for Chebotarev sums. Let $K/k$ be a Galois extension with group $G$, and suppose $S\subset G$ contains the identity. For any prime $\mathfrak{p}$ of $k$, let $δ_{\mathfrak{p}}$ be the proportion of the primes of $K$ above $\mathfrak{p}$ whose Frobenius lies in $S$. Then $\sum_{N\mathfrak{p}\le X}δ_{\mathfrak{p}}\,\log N\mathfrak{p}/N\mathfrak{p}\le(|S|/|G|+\varepsilon)\log X+O(1)$ holds uniformly in $K$ as soon as $X\ge d_K^{C}$, with $C$ and the implied constant depending only on $\varepsilon$, $C$, and the degree $[K:\mathbb{Q}]$. The proof is the argument of OpenAI in the form given by Kerr, simple vanishing at the points $(ιθ)_{ι\in S}$, with $θ$ in a box in $\mathcal{O}_K$, and Masser's zero estimate. The field enters only through Masser's invariant of the lattice of points, which we compute to be $[K:\mathbb{Q}]/|S|$ by averaging over the Galois group of the normal closure. Over $\mathbb{Q}$, the determinant method gives sharper Chebotarev bounds, and we explain the comparison.

comments
v2: references corrected and verified; related work (Frei and Kai; Cureton and Thorner, Liu) added; attribution of the determinant method revised; mathematics unchanged. 19 pages, 6 diagrams
fields
Number Theory
pdf ↗ page arXiv:2610.31402 ↗

In this paper we study the convergence of solutions of the Stokes equations in a domain perforated by slender bodies placed on a fixed periodic grid of size $2ε$ as the size of the bodies goes to zero and the number of bodies goes to infinity. We assume that the width $b_ε$ of the bodies vanishes faster than the length $a_ε$ of the particles. We identify the critical scaling regimes for $a_ε$ and $b_ε$ and prove that in this case the system is effectively described by the Brinkman equations.

comments
45 pages, 2 figures, Keywords: Homogenization, slender bodies, Brinkman limit
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.31480 ↗

In this paper, we prove that an annular capillary minimal surface outside the unit ball in $\mathbb{R}^3$ with one complete embedded end and finite total curvature must be either part of the plane or part of the catenoid. The proofs are based on the use of the Weierstrass data through the Hopf differential and associated holomorphic functions. For a planar end, we control the second fundamental form, while for a catenoidal end, we analyze the Gauss map in more detail using the boundary condition. We also prove nonexistence results for annular free boundary minimal surfaces with one end asymptotic to the end of Enneper's surface, under appropriate conditions. Finally, we establish an exterior analogue of Choe's result on capillary minimal surfaces with vertices.

comments
This submission is withdrawn because it was posted as a new submission rather than as a replacement of arXiv:1905.07887. The revised manuscript will appear at arXiv:1905.07887
fields
Differential Geometry
pdf ↗ page arXiv:2610.32447 ↗

We prove that every finite simple maximal planar graph of maximum degree at most six admits a total eight-coloring. We first show that every proper vertex four-coloring of a subcubic graph extends to a total five-coloring. A curvature inequality for boundary walks then gives a matching decomposition of four-connected triangulations with degrees five and six, extending their prescribed vertex four-colorings to total eight-colorings. In a minimum counterexample, two local boundary colorings force the degree-four vertices to have disjoint neighborhoods. A characterization of the local extension obstructions permits these vertices to be restored simultaneously. The proof uses Hamiltonian-connectedness of four-connected planar graphs and Kardoš's Hamiltonicity theorem for three-connected cubic planar graphs.

comments
51pages, 3 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.32717 ↗

We consider the homogenization problem of the stationary Stokes equations in a periodically perforated domain whose solid structure consists of a thin truss. That is, the width $a_ε$ of the truss tends to zero faster than the period size $ε$. We identify the different scaling regimes and derive the corresponding effective equations. In the critical regime, we obtain the Brinkman equations, while in the subcritical regime, we recover the unchanged Stokes equations. In a subrange of the supercritical regime, we derive Darcy's law as the effective equation. Furthermore, we explicitly compute the effective coefficients in the Darcy and Brinkman limits. Moreover, as a tool in the analysis, we establish the analogous homogenization result for a periodic forest of cylinders. Finally, we also consider the homogenization result for a fence of cylinders.

comments
66 pages, 12 figures, Keywords: Homogenization, connected structures, Stokes-Brinkman-Darcy limits
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.32804 ↗

We investigate the fourth-order Jordan-Moore-Gibson-Thompson equation \begin{equation*} u_{tttt}+\al u_{ttt}+\bet u_{tt}-γ\Del u_{tt}-\del \Del u_{t}-\varrho \Del u=(f(u))_{ttt}+(f(u))_{tt} \end{equation*} under homogeneous Dirichlet boundary conditions in a smooth bounded domain $\Om \subset \R^n$ with $n\leq 3$, where $\al \in \R$ and $\bet,γ,\del,\varrho>0$. First, for any suitably regular initial data and assuming that $f\in C^4(\R)$ with $f(0)=0$, we prove that the corresponding initial-boundary value problem admits a unique maximal strong solution $u$ on $(0, \tm)$. Second, by combining three time-integrated formulations with a series of resolvent-regularized energy estimates, we show that gradient blow-up is ruled out in the sense that \begin{equation}\label{EE} \tm<\infty \quad\,\, \Rightarrow\quad\,\, \limsup_{t\nearrow \tm}\|u(\cd,t)\|_{L^{\infty}(\Om)}=\infty. \tag{$\star$} \end{equation} Finally, assuming in addition that $f$ is convex and grows superlinearly in the sense that \begin{equation*} f''(ξ)\geq 0 \;\; \mbox{ for all}\;\;ξ\in\R, \qquad \frac{f(ξ)}ξ\rightarrow +\infty\;\;\mbox{ as}\;\; ξ\rightarrow +\infty , \qquad \int_{ξ_{0}}^{\infty}\frac{dξ}{f(ξ)}<\infty \;\; \mbox{ for some}\;\; ξ_{0}>0, \end{equation*} we employ the extensibility criterion (\ref{EE}) to prove that, for a suitable class of initial data, the solution blows up in finite time in the $L^\infty$-norm. To the best of our knowledge, this provides the first rigorous description of finite-time blow-up for a genuinely nonlinear equation of fourth order in time.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.33213 ↗

This paper presents a unified downlink architec- ture for serving indoor and outdoor users from a shared multi-antenna rooftop base station (BS). To overcome severe building-penetration loss, the BS transfers information and radio- frequency (RF) energy to an indoor access point (AP) using simultaneous wireless information and power transfer (SWIPT). The AP harvests this energy and performs amplify-and-forward (AF) relaying, enabling self-sustained indoor communication. A movable waveguide-mounted pinching antenna (PA) further improves indoor coverage by adapting its position to the prop- agation environment. For outdoor transmission, zero-forcing (ZF) beamforming suppresses inter-group interference, while power-domain non-orthogonal multiple access (NOMA) supports multiple users within each group. We maximize the outdoor weighted sum rate (WSR) while satisfying stringent indoor and outdoor quality-of-service (QoS) requirements. The resulting nonconvex problem jointly optimizes the BS power fraction, SWIPT power-splitting ratio, PA position, and NOMA power allocation. To obtain an efficient solution, we develop a two-stage method based on alternating optimization (AO) and sequential quadratic programming (SQP). Numerical results show that the proposed architecture improves outdoor spectral efficiency while maintaining reliable, energy-sustainable indoor coverage. Increasing the BS array size further raises the indoor QoS- satisfaction probability toward unity, even under demanding rate targets.

fields
Information Theory
pdf ↗ page arXiv:2610.33367 ↗

Let $X$ be a curve over $\mathbb{F}_q$. Denote by $N(X)$ and $g(X)$ the number of $\mathbb{F}_q$-rational points and the genus of $X$ respectively. The Ihara constant \begin{equation*} A(q)=\limsup_{g(X)\to\infty}\frac{N(X)}{g(X)} \end{equation*} measures the asymptotic abundance of rational points on curves over $\mathbb{F}_q$. It is a central problem in the study of algebraic geometry and has also attracted broader mathematical interest. In particular, the constant $A(2)$ is discussed in the \emph{Optimization Constants in Mathematics} collection maintained by Terence Tao et al \cite[Number 33]{DIT26}. In this paper, we study the Zassenhaus filtration of pro-$p$ groups arising from class field towers. By combining Golod–Shafarevich theorem, embedding problems and central class field theory, we establish a new criterion for the infinitude of class field towers. As an application, we obtain an improvement on the best known lower bounds of Ihara constant $A(2)$ and $A(3)$ which are explicitly given by \[ A(2)\geq 0.317779\cdots \text{ and } A(3)\geq 0.496074\cdots. \]

comments
29 pages
fields
Number Theory
pdf ↗ page arXiv:2610.33649 ↗

In communication networks, network coding allows intermediate nodes to combine messages, achieving transmission rates unattainable by routing alone in many multicast settings. In multiple unicast, where each source sends a message only to its prescribed destination, a central question is under what conditions fractional routing suffices to attain the transmission rates supported by network coding. We prove that, for any given collection of multiple-unicast demands in an undirected network, if the maximum flow between every pair of distinct destinations is at least the total demand rate, then the following are equivalent in feasibility: multiple-unicast coding, multiple-unicast fractional routing, and simultaneous fractional routing of all source rates to any fixed destination. Verifying this pairwise maximum-flow condition and, when it holds, deciding feasibility of the given multiple-unicast demands require at most as many maximum-flow computations as there are destinations. As a direct corollary, we prove the Langberg–Médard conjecture: if a directed network supports coded multicast of independent source messages at a common rate to every destination, then removing arc directions without changing capacities permits simultaneous fractional routing at that rate for all prescribed source–destination pairs.

comments
14 pages, 3 figures. Revised exposition and proof presentation
fields
Information Theory
pdf ↗ page arXiv:2610.33655 ↗

We prove that a polynomial mapping $\mathbb{R}^2 \to \mathbb{R}^2$ with nowhere zero Jacobian determinant is a global diffeomorphism provided the preimage of large circles are connected. In higher dimensions this result fails and we present families of counterexamples. Yet, in dimension $3$ with codimension of Jelonek set greater than one, we prove that if the preimages of large spheres are connected, then the Jelonek set of the mapping is neither homeomorphic to a line nor can have a single end at infinity.

fields
Algebraic Geometry, Dynamical Systems
pdf ↗ page arXiv:2610.33875 ↗

Optimal designs for regression with asymmetric errors build on the second-order least squares estimator (SLSE) of Wang and Leblanc (2008), whose gain over ordinary least squares (OLS) comes from the error skewness. With i.i.d. errors the optimally weighted SLSE solves the same estimating equation as the degree-two Polynomial Maximization Method (PMM), a family whose degree-$S$ location estimator has asymptotic variance $g_S\le1$ times that of the sample mean. We derive optimal designs for this family when the errors are independent with constant variance but have standardised cumulants of order three and higher that vary with the regressor (cumulant heteroscedasticity). For a known reduction field $g_S(x)$, the $g_S$-weighted information matrix $M_{g_S}$ obeys the Kiefer-Wolfowitz equivalence theorem. A constant field leaves the OLS-optimal design unchanged; a non-constant field can change it but need not. For symmetric errors with varying kurtosis, the PMM$_2$ and conditional-SLSE designs cannot depart from the OLS-optimal one, whereas a degree-three design can. A stratify-and-pool PMM$_S$ estimator attains $N\mathrm{Cov}(\hatθ)\toσ^2M_{g_S}^{-1}$ at every degree as all strata grow; its per-stratum reduction is below one only for symmetric errors with the symmetry imposed. For a quadratic model on $[0,1]$ with symmetric Beta errors of unit variance and dose-varying shape, the third-order design for the known field has $A$-efficiency $1.047$ relative to the OLS-optimal design, and $1.001$ under the $D$-criterion. A two-stage procedure that estimates the field from a one-third pilot builds designs of $A$-efficiency $1.014$-$1.020$ at $N=600$-$2400$; its realised mean-squared-error ratio against the OLS-optimal design ranges from $0.89$ (a loss) to $1.08$.

comments
36 pages, 4 figures, 7 tables. Companion to arXiv:2606.11421 (second-order least squares as the degree-two polynomial maximization method). v2: added references to published related work; code availability points to the CRAN package
fields
Statistics Theory
pdf ↗ page arXiv:2610.33921 ↗

Using the analytic machinery and numerical inputs of OpenAI's proofs of the Haldane gap, we prove uniqueness of the translation-invariant ground state of the infinite $S=1$ Heisenberg antiferromagnetic chain and of the $J$-$D$ model for sufficiently small $|D|$. Uniqueness is among all translation-invariant ground states in the operator-algebraic sense, without assuming a gap or spin-rotation symmetry for a competing state. The proof combines concentration of the spatial transfer spectrum, an exponentially small low-temperature free-energy correction, and a Golden–Thompson comparison for separated local observables. The unique translation-invariant state is pure, locally unique and gapped, and continuous in $D$. The ground states of even periodic chains and odd open chains with equal positive endpoint fields converge to this state along their full sequences, uniformly in $D$, with a stretched-exponential local rate. Every choice of ground states of open chains without endpoint magnetic fields has the same local limit. These conclusions do not establish uniqueness among arbitrary nontranslation-invariant ground states of the infinite chain. The author formulated the problems and proposed the core strategies. ChatGPT Astra made substantial contributions to developing the proofs and preparing the manuscript.

comments
46 pages, 2 figures. Version 2: The title has been changed. We prove uniqueness among translation-invariant ground states of the infinite chain, a substantially stronger result than the boundary-condition independence established in version 1
fields
Mathematical Physics
pdf ↗ page arXiv:2610.34330 ↗

We count Boolean functions represented as the sign of a real polynomial of bounded degree on the $n$-dimensional cube. We prove that the logarithm of their number is asymptotic to that of the classical hyperplane-arrangement upper bound, uniformly over all degrees, with relative error $O((\log n)^2/n)$. We also obtain sharper additive and relative error estimates in particular degree ranges, including an additive error $O(n^2)$ between the logarithms throughout a small linear degree range and improved bounds just below the central degree. The proof combines probabilistic estimates for Reed–Muller codes with a deletion argument for sets of evaluation vectors, an entropy count, and a sharp-threshold estimate for expected rank deficiency.

comments
39 pages. Revised exposition and organisation, with expanded explanations. Main results and theorem statements unchanged
fields
Information Theory, Combinatorics
pdf ↗ page arXiv:2610.34378 ↗

When a product is sold out, customers who prefer it may buy a substitute or leave without purchasing. Efficient demand estimation in this setting is challenging when inventory is observed only at replenishments, because the stockout times, and hence the products available to each customer, are unobserved. Estimating demand efficiently under stockout substitution has been a challenging problem since Anupindi et al. (1998), and it has become more pressing as the scale of such problem instances grows thanks to innovations in retailing such as unmanned stores. We provide a complete solution to this problem for the multinomial logit (MNL) model, the most widely used model of customer choice. Using a continuous-time representation of the purchase process developed in Zhou et al. (2026), we show that the exact likelihood and its gradient are one-dimensional integrals, no matter how many products, customers, or stockouts there are. Both can be evaluated at a cost linear in the numbers of products and periods. Our method estimates demand for 2,000 products from 2,000 selling periods in under ten minutes.

fields
Optimization and Control
pdf ↗ page arXiv:2610.34547 ↗

We show that if $κ^{<κ} = κ$, then there is a $κ$-strategically closed, $κ^+$-c.c. partial order that forces $\square_κ$. We discuss consequences for generic large cardinals.

fields
Logic
pdf ↗ page arXiv:2610.34900 ↗

We construct closed mean curvature flows with arbitrarily large genus increase in every smooth Riemannian three-manifold, giving counterexamples to Ilmanen's genus-reduction conjecture. We prove that a compact hypersurface with a single isolated conical singularity and a power rate of convergence to its cone admits a mean curvature flow whose forward rescaled limit is a prescribed expander in the tangent space at the singular point, smoothly asymptotic to that cone. Combining this theorem with closed mean curvature flows with prescribed tangent flows (arXiv:2405.15577, arXiv:2607.23879) and the noncompact counterexample (arXiv:2605.09657) gives the desired examples.

comments
21 pages. Comments are welcome! v2: Added the Riemannian case
fields
Differential Geometry, Analysis of PDEs
pdf ↗ page arXiv:2610.35000 ↗

Banderier, Marchal, and Wallner studied the enumeration problem for Young tableaux with walls. These are similar to standard Young tableaux, except that local decreases are allowed at some walls. In 2021, Banderier and Wallner counted Young tableaux with periodic walls, in which the elementary blocks are Young diagrams of shape $(2,2)$. They obtained exact counting formulas for $30$ of the $32$ classes up to symmetry. In this paper, we first derive recurrences for counting the remaining two classes of Young tableaux with periodic walls. We then prove that their counting exponential generating functions are differentially transcendental. Consequently, they are neither D-algebraic nor D-finite. This disproves a conjecture of Banderier and Wallner.

fields
Combinatorics
pdf ↗ page arXiv:2610.35067 ↗

We study the global regularity of solutions to the homogeneous Dirichlet problem for a class of Monge-Ampère equations, which become degenerate or singular on the boundary due to the zero boundary value and the distance function. The degeneracy and singularity are described by the index - the sum of the powers of the distance function and the solution. We give a complete classification of the index, and prove that the solution is globally Hölder continuous, log-Lipschitz continuous, $C^{1,η}$, $C^{2,α}$ and $W^{2,p}$ respectively, depending on the size of the index. We also obtain estimates related to the five regularity results. All these results are optimal.

comments
32 pages, no figures. Submitted for publication
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.35192 ↗

We prove global well-posedness and regularization for the isotropic consensus-based optimization equation with initial probability measures of finite fourth moment. Under suitable weight decay assumptions, after any nonconstant consensus motion, both the density and the consensus trajectory become smooth, and all spatial Sobolev norms are locally bounded in time. The proof uses a Fourier commutator estimate, localization and iteration, followed by a nonlinear time bootstrap. Counterexamples show that smoothing can fail at a stationary consensus point. For non-Dirac initial data, we also prove positivity away from the consensus point and full support at every positive time in all dimensions.

comments
add remark 1.4 and expand the introduction
fields
Analysis of PDEs, Optimization and Control
pdf ↗ page arXiv:2610.35266 ↗

We study how a functional inequality satisfied by the one-particle conditional distributions of a Gibbs measure, with a constant independent of the positions of the other particles, can be propagated to the whole N -particle system, with a constant independent of N , when the interaction is singular. Our model is the Coulomb gas on the torus T 3 in the mean-field regime. We distinguish two mechanisms. The first one is a dependent tensorization, governed by Dobrushin-Wu interdependence coefficients. It yields uniform Poincar{é} and transport-information W 2 I inequalities with fully explicit constants. We explain why this mechanism does not extend to the log-Sobolev inequality: the Zegarlinski-type coefficients, which require pointwise bounds on the interaction, are infinite. The second mechanism is a tensorization with a defect. Using a uniform bound on the partition function due to Duerinckx and Jabin, it yields a uniform log-Sobolev inequality for the Coulomb gas on T 3 at small charge. We identify the assumptions on the interaction under which each mechanism applies. All the results are then extended to the two-dimensional Coulomb gas and to periodic Riesz gases 1/|x|^s , 0 < s $\le$ 1. The extension to d $\ge$ 4 is also discussed.

fields
Functional Analysis, Mathematical Physics
pdf ↗ page arXiv:2610.35350 ↗

Let $V_{g,n}$ denote the Weil–Petersson volume of the moduli space of genus-$g$ hyperbolic surfaces with $n$ cusps. Mirzakhani and Zograf obtained a complete asymptotic expansion as $g\to\infty$ for fixed $n$; its terms are effectively powers of $n^2/g$, so it degenerates once $n$ is of order $\sqrt g$. We prove an asymptotic expansion in inverse powers of the genus that is uniform in all $n\ge0$. The leading term is an explicit saddle-point expression whose saddle moves with the cusp density $n/g$. It has relative error $O(g^{-1})$, and truncation at an order proportional to $\sqrt g$ gives relative error $O(\exp(-a\sqrt g))$ for some $a>0$. For fixed $n$, the leading term confirms Zograf's conjecture that the constant in the Mirzakhani–Zograf asymptotics is $1/\sqrtπ$. We also determine the asymptotic profiles of volume ratios, uniformly in the number of cusps, and of adjacent coefficients of the one-boundary volume polynomial, uniformly in the degree; this answers two questions of Mirzakhani. The proof uses a positive recursion for mixed $κ_1$--$ψ$ intersection numbers, normalized at the pure-$ψ$ endpoint, and a local limit theorem in the number of cusps.

comments
84 pages, no figures. v2: added uniform boundary and tight-spectrum results; expanded proofs and updated references, including removal of a withdrawn dependency. An earlier version was first made public on Zenodo on 16 October 2026: https://zenodo.org/records/23414478
fields
Differential Geometry, Algebraic Geometry
pdf ↗ page arXiv:2610.35492 ↗

Disjunctions over the cells of a triangulation arise in formulations of piecewise-linear functions. For a poset $P$ on $[d]$, we study the convex hull $D$ of the simplices of Stanley's triangulation of the order polytope of $P$, each lifted by its own indicator vector. Via the Cayley trick, for any disjunction of full-dimensional polytopes, the facets of the convex hull other than those described by the bounds on the indicator variables correspond to the facet normals of the Minkowski sum of the polytopes, and each is described by the full optimal big-M lifting of the supporting inequality with that normal of any one polytope. For Stanley's triangulation, with $M(P)$ the Minkowski sum of its simplices, lifting only the facet-describing inequalities of the simplices need not suffice, unlike some other cases; we show that it suffices if and only if $M(P)$ is alcoved, if and only if deleting one element of $P$ leaves a chain. Otherwise, the coefficients of the additional facet-describing inequalities are governed by the least number $\ell(P)$ of elements whose deletion leaves a chain: for fixed $\ell(P)$, the largest entry of a primitive facet normal grows like $d^{\ell(P)-1}$, and it can be exponential in $d$ if $\ell(P)$ grows with $d$. For $d \geq 3$ and $\ell(P)\leq2$ the sharp bound is $d-2$.

fields
Combinatorics
pdf ↗ page arXiv:2610.35668 ↗

Staircase simplices are the simplices of the standard (Freudenthal–Kuhn) triangulation of $[0,1]^d$ and of Stanley's triangulations of order polytopes. We study the facet normals of their Minkowski sums and how the number of summands constrains the size of primitive integer normals. We characterize the normals attainable with a prescribed number of summands in terms of $d-1$ linearly independent zero-one vectors orthogonal to the normal, whose supports have an intersection graph that can be colored with one color per summand. This gives a combinatorial characterization of the least number of summands needed to realize a normal. For two summands, all primitive normals have entries in $\{-1,0,1\}$, and we determine which of them occur. For three summands we bound every entry in absolute value by $3^{(d-2)/3}$ and construct normals with entries of order $θ^d$, $θ$ the plastic number; for four summands a second construction gives entries of order $ψ^d$, $ψ$ the supergolden ratio. Since $ψ>3^{1/3}$, the largest coefficients for three and four summands are exponentially separated. For vectors with at most two distinct nonzero entry values, we determine the least number of summands exactly, using a matching-rank formula proved by exterior algebra and representations of symmetric groups. With no restriction on the number of summands, the maximum coefficient equals the classical maximum for facets of $0/1$-polytopes in the same dimension.

fields
Combinatorics
pdf ↗ page arXiv:2610.35689 ↗

We give a fourth-order Calabi-Yau differential operator, which seems to have been missed by earlier tabulations. We draw attention to this operator because of its clear geometric interpretation and evidence that it possesses a rank-two attractor, conjecturally giving a new example of a Calabi-Yau threefold with modularity properties analogous to those of an elliptic curve over $\mathbb{Q}$.

comments
10 pages. v2: Added reference to Master's thesis of Paul Blesse
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.35798 ↗

Let $s(C_n^r)$ be the Erdős–Ginzburg–Ziv constant of $C_n^r$ and $η(C_n^r)$ the short zero-sum constant. For even $n$, Harborth's construction gives $s(C_n^r)\ge2^r(n-1)+1$, with equality when $n$ is a power of two. Wang and Zeng recently showed that this bound is strict for every $r\ge4$ and every even $n$ that is not a power of two, with an excess of $2^{r-3}(m-1)$, where $m$ is the odd part of $n$; this disproves Properties C and D. We show that the excess grows faster than $2^r$. For every even $n=2^am$ with odd $m\ge3$ and every $r\ge4$, \[ s(C_n^r)\ge2^r(n-1)+(m-1)\,2^{r-3}\lfloor r/4\rfloor+1 , \] so that $s(C_n^r)-2^r(n-1)\ge c_n\,r\,2^r$.

comments
Comments welcome
fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2610.35819 ↗

Get each issue by email →