Paperlane

no. 6 · from the archive← Fri 30 Oct archive Tue 3 Nov →

Monday, 2 November 2026

639 papers in 32 fields

Combinatorics58

Median genomes serve as candidates for ancestral reconstruction and help study the diversity of parsimonious rearrangement solutions. However, the breakpoint median problem is NP-hard for unichromosomal genomes, and medians are non-unique, making their diversity important to study. In this paper, we develop tree-based algorithms for studying breakpoint medians of linear unichromosomal genomes represented as unsigned permutations. The algorithms construct rooted labeled trees whose leaves encode candidate median genomes satisfying local adjacency constraints derived from the inputs. We show that the resulting reduced search spaces contain all breakpoint medians while being much smaller than the full permutation space. We investigate these reduced spaces experimentally for three-genome instances. Our method finds exact medians and many near-optimal approximations, while early sampling of leaves often produces near-medians quickly. We also introduce constrained sampling procedures for targeting balanced near-medians, whose distances to the input genomes are more evenly distributed and hence are particularly relevant for ancestral reconstruction. The resulting framework provides a unified method for exact enumeration, efficient sampling, and targeted exploration of median and near-median genomes. In contrast to Traveling Salesman Problem-based approaches, which typically return a single solution and do not reveal the diversity or balance of the median set, our method can enumerate all medians, sample near-medians early, and target structurally distinct candidates under the breakpoint distance.

fields
Combinatorics
pdf ↗ page arXiv:2610.40667 ↗

We consider an undirected edge-traversal game on a finite bouquet of simple cycles. Each of two players controls a token, both tokens begin at the common root, and the players alternate traversing edges that have not been used by either token. A player with no legal move loses. We prove that the first player wins precisely when the number of cycles of odd edge-length is odd, without assuming equal cycle lengths. The proof uses a simultaneous induction on root and defense configurations, with an exact analysis of the forced race between two opened cycles. We also explain how the classification corresponds to a completed Lean 4 formalization of the underlying game on finite edge sets.

code
github.com/adamtilepe/RootedTrailTrapHeterogeneousBouquets ∎ Lean project (lakefile.toml, lean-toolchain) builds, no sorry at f95b7ff, checked 2026-11-02 build details
comments
Complete Lean 4 formalization available at https://github.com/adamtilepe/RootedTrailTrapHeterogeneousBouquets
fields
Combinatorics
pdf ↗ page arXiv:2610.40683 ↗

We prove that every graph with $m>0$ edges contains at most $\frac{3}{2\sqrt{2}}(\frac{m}{5})^{5/2}$ induced pentagons. Equality holds precisely for balanced blow-ups of the Clebsch graph, together with isolated vertices. The bound and equality classification extend to arbitrary nonnegative vertex weights and to all graphons, including fractional graphons. Moreover, a deficit of $\varepsilon$ in the normalized ratio can be repaired by $O(\sqrt{\varepsilon}m)$ edge edits on the original vertex set, producing an exactly balanced Clebsch blow-up and isolates. The exponent $1/2$ is optimal. We further separate the linear cost of adjacency errors from the quadratic cost of class imbalance. This yields an arithmetic description of the finite-edge deficit and its explicit second-order term, with a uniform power-saving remainder. For each fixed integer $j$, the exact extremal count at $m=40k^2+j$ is eventually an integer coefficient polynomial in $k$, computable by a finite enumeration; we do not quantify its threshold of validity. The proof combines global variational conditions, an active-density bound, and an exact rational certificate on the $1,044$ graphs of order seven. A positive-slope identity permits a single rational reference density. An exact conditional span excludes triangles at equality, after which degree and codegree identities determine the Clebsch structure. All finite certificates admit independent verification using integer arithmetic.

comments
59 pages, 1 figure
fields
Combinatorics
pdf ↗ page arXiv:2610.40691 ↗

For positive integers $v_1,\dots,v_n$ let $\mathrm{ML}(v_1,\dots,v_n)=\max_t\min_i\|v_it\|$, where $\|x\|$ is the distance to the nearest integer. Kravitz asked which values below $1/n$ occur. For $n=4$, Jain and Kravitz showed that the values below $1/4$ are $m/(4m+2)$, $m\ge2$, up to finitely many undetermined exceptions. We settle the four-runner case completely. A primitive quadruple has $\mathrm{ML}<1/4$ if and only if it is, up to order, $\{1,2,3,4k\}$, or $\{x,y,x+y,x+2y\}$ with $\gcd(x,y)=1$ and $y\not\equiv x,2x\pmod 4$, or $\{1,3,4,14\}$. The values below $1/4$ are exactly $m/(4m+2)$ with $m\ge2$, $m\ne3$, so $3/14$ is the only exception, and Conjecture 4.1 of Fan and Sun holds. A density lemma bounds the length of a quadruple on neither critical subtorus by $512$, and an exact search over 800,816,099 quadruples completes the proof. The same method shows that the two-dimensional subtori of $(\mathbb{R}/\mathbb{Z})^4$ with loneliness below $2/5$ take exactly the three-runner values in that range, which between $1/3$ and $2/5$ are the fractions $p/q$ with $1\le 2q-5p\le 4$. Hence the accumulation points of the four-runner spectrum in $(0,2/5]$ are the three-runner values there, as Giri and Kravitz conjectured, and $9/25$ is not the loneliness of a two-dimensional subtorus. For every $n$, assuming the conjecture for $n-1$ and $n-2$ speeds, we prove $\mathrm{ML}(v)>1/n-(n(n-1))^{n-2}/(2|v|)$, which is sharp in $|v|$, and that every $n$-tuple with $\mathrm{ML}<1/n$ and $|v|\ge n(n(n-1))^{2n-4}$ lies on a two-dimensional subtorus of loneliness $1/n$. This answers a question of Cordella. For five speeds, a two-dimensional subtorus of $(\mathbb{R}/\mathbb{Z})^5$ with loneliness below $2/9$ is the product of a coordinate circle with the circle of $\{1,2,3,4\}$ or $\{1,3,4,7\}$. We also give infinitely many subtori whose maximizers lie off every circle with a repeated speed.

comments
45 pages. Code, data and verification scripts are included as ancillary files
fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2610.40694 ↗

Let $f_{p+1}(n,m)$ denote the minimum number of $K_{p+1}$-saturating edges in an $n$-vertex $K_{p+1}$-free graph with $m$ edges. In this paper, we investigate the stability of this parameter near the Turán thresholds, with particular emphasis on the role of the minimum degree. We introduce the refined parameter $f_{p+1,p}(n,m,δ)$. It is defined for $K_{p+1}$-free graphs on $n$ vertices with $m$ edges, minimum degree $δ$, and containing at least a copy of $K_p$. For every fixed integer $p\ge3$, we establish a stability result when the number of edges is $\mathrm{ex}(n,K_p)+o(n^2)$, showing that \[ f_{p+1,p} (n,m,δ)=\left(\frac{(p-2)^2}{2(p-1)^3}+o(1)\right)n^2, \ \text{for }\ m=\mathrm{ex}(n,K_p)+o(n^2), \ δ\geq \left(\frac{p-2}{p-1}+o(1)\right)n. \] We further obtain a stability theorem in the regime where the number of edges is quadratically smaller than $\operatorname{ex}(n,K_{p+1})$. More precisely, for $0 \leq c< \frac{1}{2p(3p-1)}$, \[ f_{p+1,p} (n,m,δ)=\left(\frac{1}{2p}+o(1)\right)n^2, \ \text{for }\ m=\operatorname{ex}(n,K_{p+1})-(c+o(1))n^2, \ δ\geq \left(\frac{p-1}{p}-2c+o(1)\right)n. \]

fields
Combinatorics
pdf ↗ page arXiv:2610.40696 ↗

Hibi and Li conjectured that the order polytope $\mathcal{O}(P)$ of a finite poset $P$ has at most as many faces of each dimension as its chain polytope $\mathcal{C}(P)$, and that equality in some dimension $k$ with $1\leq k\leq|P|-1$ forces the two polytopes to be unimodularly equivalent. Fourier proved the inequality in a 2026 preprint. Let $X$ be the five-element poset in which an element lies above two incomparable elements and below two other incomparable elements. Building on Fourier's proof, we show that if an $n$-element poset contains $X$ as an induced subposet, then, for every $k$ with $2\leq k\leq n-1$, its chain polytope has at least $\binom{n-3}{k-2}$ more $k$-dimensional faces than its order polytope, and that for every $n\geq5$ some $n$-element poset containing $X$ attains all these bounds at once. Since Hibi and Li showed that $\mathcal{O}(P)$ and $\mathcal{C}(P)$ are unimodularly equivalent exactly when $P$ does not contain $X$, this proves the equality statement for these $k$. For $k=1$ the equality statement is false: $\mathcal{O}(X)$ and $\mathcal{C}(X)$ have equally many edges (Hibi, Li, Sahara and Shikama) but are not unimodularly equivalent. With Fourier's inequality, these results settle the conjecture. Every result is formally verified in Lean 4, with only published results and two results of Fourier's preprint as axioms.

code
github.com/david44hello/papers_supporting_materials/tree/hibi-li-2016-conjecture-v2.10/hibi-li-2016-conjecture ∎ Lean project (cranston-li-su-wang-xu-2026-conjecture/lean/lakefile.toml, cranston-li-su-wang-xu-2026-conjecture/lean/lean-toolchain) builds, no sorry at 15495bd, checked 2026-11-13 build details
fields
Combinatorics, Metric Geometry
pdf ↗ page arXiv:2610.40703 ↗

Let $S$ be a subset of $n$ elements of a group $G$. An ordering $x_1,\ldots,x_n$ of its elements is valid if the partial sums $x_1,x_1+x_2,\ldots,x_1+\cdots+x_n$ are pairwise distinct. This notion has a long history in combinatorial group theory, dated back to Gordon's works in early 1960. Alspach-Liversidge conjecture asserts that any subset of the nonzero elements of a finite abelian group $G$ has a valid ordering. This generalizes Graham's conjecture, which considers the case when the group $G$ is $\mathbb{F}_p$, the cyclic group of prime order. Graham's conjecture has been solved recently, as a combination of results by Kravitz, Bedert and Kravitz, Müyesser and Pokrovskiy, Bedert, Bucić, Kravitz, Montgomery, and Müyesser, and Pham and Sauermann. Beside this, the only other case where Alspach-Liversidge conjecture has been confirmed is $G=\mathbb{F}_2^k$, thanks to a recent result by Bedert, Bucić, Kravitz, Montgomery, and Müyesser. We prove the Alspach-Liversidge conjecture in its full generality. Our proof combines ideas from many previous papers. In particular, we will use the probabilistic approach and the absorption method. Our new main technical ingredient is a Fourier-moment estimate, which is of independent interest. In particular, we can use it to show that the distribution of certain kinds of random sums in an abelian group has an almost uniform distribution.

fields
Combinatorics
pdf ↗ page arXiv:2610.40712 ↗

Let $A$ be a finite abelian group. A nowhere-zero $A$-flow of a graph gives every edge a direction and a non-zero value in $A$, so that at every vertex the sum of the values on the entering edges equals the sum on the leaving edges. Such a flow can be changed along a cycle: going around the cycle, add a fixed non-zero element of $A$ to the value of every edge that points forward and subtract it from the value of every edge that points backward. The Cranston–Li–Su–Wang–Xu (2026) conjecture states that if $A$ has at least four elements, then in every graph in which all degrees are even, any two nowhere-zero $A$-flows are joined by a chain of such changes through nowhere-zero $A$-flows. It also states that, for every integer $k\ge 4$, the same holds for integer flows whose values are between $1$ and $k-1$ in absolute value, where each change adds a non-zero integer. We prove this conjecture. The core of the proof is an induction on the number of edges: if a flow enters a vertex $v$ along an edge $uv$ and leaves $v$ along an edge $vw$ with the same value, the two edges can be replaced by one edge $uw$, and the induction hypothesis applies to the smaller graph. So the main task is to change every flow into a flow with such a pair of edges. For nowhere-zero $A$-flows with $|A|\ge 4$, replacing vertices of degree more than $4$ by triangles reduces the problem to graphs with all degrees at most $4$. For integer flows, a counting argument joins any two nowhere-zero flows with values in the cyclic group $\mathbb{Z}_k$, $k\ge 4$, by changes that add $1$, $-1$, $2$ or $-2$; every such chain gives a chain of integer flows with values between $1$ and $k-1$ in absolute value, and every integer flow of this chain reduces modulo $k$ to a flow of the given chain. All results of this paper are also proved in the Lean 4 proof assistant, from Lean's standard axioms alone.

code
github.com/david44hello/papers_supporting_materials/tree/cranston-li-su-wang-xu-2026-conjecture-v1.59/cranston-li-su-wang-xu-2026-conjecture ∎ Lean project (cranston-li-su-wang-xu-2026-conjecture/lean/lakefile.toml, cranston-li-su-wang-xu-2026-conjecture/lean/lean-toolchain) builds, no sorry at 15495bd, checked 2026-11-13 build details
fields
Combinatorics
pdf ↗ page arXiv:2610.40863 ↗

Using a unified method of identifying a telescoping function for each statistic-action pair, we prove conjectures of La Croix and Roby, and of Sheridan-Rossi that various permutation statistics, such as the number of fixed points, are homomesic under Foatic actions: compositions of the Foata bijection and dihedral symmetries of one-line notation. Underlying these results is a family of bijective identities relating permutation statistics across the Foata bijection; one of them --- an unexpected invariance of the number of right-to-left minima under reverse-complement, proved by an explicit bijection --- may be of independent interest. The telescoping functions also turn out to be shared among some Foatic actions, implying that mixing certain actions still yields a constant average on closed walks.

comments
37 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.40882 ↗

Kostochka and Thomason independently proved that every graph with density at least $β\,t\sqrt{\log t}$ has a $K_t$ minor, for some constant $β$. Thomason later proved that the least such density is $(0.319\ldots+o(1))\,t\sqrt{\log t}$. However, the $o(1)$ term in this formula is not explicit, so the formula gives no bound for any specific value of $t$. The goal of this paper is a bound of the form $β\,t\sqrt{\log t}$ that holds for every $t$, with $β$ small. We prove that, for every integer $t\geqslant2$, every graph with density at least $4.5\,t\sqrt{\log t}$ has a $K_t$ minor.

fields
Combinatorics
pdf ↗ page arXiv:2610.41060 ↗

We extend a theorem of Geelen, Kroeker, and Nelson on unavoidable flats in $\operatorname{GF}(q)$-representable matroids. For all integers $1\le a\le b$ and $n\ge1$, we prove that every simple matroid of sufficiently large rank with no $U_{a,b}$-minor has a rank-$n$ flat whose restriction is free or isomorphic to a projective or affine geometry over a finite field.

comments
9 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41096 ↗

We prove that the size of a random independent set in every nonempty forest has variance strictly smaller than its mean at common activity $0<t\le1$. A rooted quadratic reserve gives quantitative forms of this inequality. Mean matching transfers the bounds to fixed-weight canonical completion laws, and exact binomial moments prove strict log-concavity for every forest corona and every well-covered forest. The stronger factorial normalization $((m-j)!i_j)_{j=0}^m$ is strictly log-concave; we give explicit coefficient margins, terminal sharpness, one-less smoothing, and the exact universal three-index binomial ULC prefix. An exact attachment argument doubles the rooted reserve for every nontrivial unweighted tree. For a single marked root, a square reserve holds through $Θ=\tfrac12e^{3-2W(1/e)}$ and for every positive mark at roots of degree at most four. A stronger reserve holds through mark $15/14$. These estimates yield marked clique, adjacent-pair, and repulsive two-piece variance bounds with fixed-weight canonical transfer. Finally, arbitrary positive site activities in $[3t/4,t]$ satisfy a uniform rooted reserve on every forest. This proves strict log-concavity for weighted forest coronas with compatible base marks. An explicit weighted-star obstruction shows why an unrestricted heterogeneous extension is impossible. Complete proofs are included, together with the required classical graph characterization and variance comparison.

comments
44 pages, 1 figure. Complete proofs and exact rational certificate files. Relationship to arXiv:2610.22098 is disclosed in the Introduction
fields
Combinatorics, Probability
pdf ↗ page arXiv:2610.41102 ↗

For a binary word $x\in \{0,1\}^+$, the $k$-deck of $x$ is the multiset of all length-$k$ subsequences counted with multiplicity. The deck reconstruction problem asks for the minimum $k$ such that every word of a fixed length is uniquely determined by its $k$-deck. In this paper, we study deck reconstruction for several restricted classes of binary words, including words with exactly $s$ ones, words with at most $\ell$ runs of ones, and words where the indices of ones are given by a monotone polynomial of degree $d$. We prove that for these classes of words, their reconstruction thresholds are bounded by a function of the structural complexity of the class. We also study weak deck reconstruction, in which subwords may contain "wildcard" symbols that map to either 0 or 1. By generalizing a "template lemma" of Dudík and Schulman, we improve the upper bound in an open question about weak reconstruction from Dudík and Schulman. Finally, for the class of periodic words with period $d$, we prove that its in-class reconstruction threshold is equal to the general reconstruction threshold for words of length $d$, and give a general template construction producing further classes with the same threshold-preservation phenomenon.

comments
20 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41110 ↗

With the expansion of complex transport infrastructure, new paths for linking different destinations have emerged, allowing the definition of multiple routes. The introduction of new roads may also lead to disruptions in traffic flow. This problem can be generalized into a special case of VRP, which aims to find the optimal route for a fleet of vehicles within a network of destinations. Finding solutions to this problem requires an exponential number of operations, hindering the ability of exact solvers to deal with large instances, which must therefore be tackled using heuristics, often leading to suboptimal results. Quantum-based algorithms may allow us to find exact solutions in polynomial time with high probability, offering a potential future alternative to classical solvers and in this work, we focus on the QAOA. Our novel contributions are a parameter-initialization strategy that improves the convergence probability and the integration of the subtour constraints directly into the optimization without using additional qubits by employing an approximation of exponential equality constraints. We believe that our results pave the way for effectively solving instances of VRP using quantum-based solutions, leveraging the space-efficient problem encoding and our initialization strategy, which limits the effect of barren-plateau under random parameter initialization.

comments
12 pages, 6 figures, 1 table
fields
Combinatorics, Optimization and Control
pdf ↗ page arXiv:2610.41120 ↗

A dessin is a 2-cell embedding of a connected bipartite graph into an orientable closed surface. It is regular if the colour- and orientation-preserving automorphism group acts regularly on edges, and circular if every face boundary is a simple cycle. We classify and enumerate circular regular dessins of the complete bipartite multigraph ${\bf K}_{2^e,2^f}^{(2^d)}$ ($e\ge f\ge 2$, $d\ge 1$) whose automorphism group is {\it metacyclic}. This extends the work of Fan (circular, metacyclic) and Cai–Hu (regular, metacyclic) from simple graphs ${\bf K}_{2^e,2^f}$ to multigraphs ${\bf K}_{2^e,2^f}^{(2^d)}$. Such dessins correspond to non-exact bicyclic factorisations $G=ła\rłb\r\cong\ZZ_{2^{e+d}}\ZZ_{2^{f+d}}$, where $ła\r\capłb\r\cong\ZZ_{2^d}$ and $łba\r\cap ła\r=łba\r\cap łb\r=1$. In general, a group admitting such a factorisation need not be metacyclic; in this paper we impose the additional condition that $G$ is a metacyclic 2-group. Under this assumption, we obtain explicit enumeration formulas for the number of such dessins up to isomorphism. Together with the known classifications for simple graphs (the case $d=0$), this completes the enumeration of the metacyclic case for 2-power complete bipartite multigraphs.

fields
Combinatorics
pdf ↗ page arXiv:2610.41123 ↗

We study the transition coefficients $b_{λ,μ}(q,t)$ of the Hall–Littlewood $P$-functions $P_λ(t)$ expanded in the basis $P_μ(q)$. We find a tableaux formula when $λ$ is a hook shape which generalizes term-by-term Lascoux-Schutzenberger's charge formula as well as Macdonald's monomial expansion formula for Hall-Littlewood functions. We conjecture that $b_{λ,μ}(t+1,t)\in\mathbb{Z}_{\ge0}[t]$ and prove it when $λ$ is a hook shape, when $λ$ has two-columns and some more simple cases. We prove that $b_{λ,μ}(t+n,n)\in\mathbb{Z}_{\ge0}[t]$ for every $n\in\mathbb{Z}_{\ge0}$, extending work of Alexandersson, and we record a stability property and some reduction principles.

fields
Combinatorics, Representation Theory
pdf ↗ page arXiv:2610.41155 ↗

We give a deterministic algorithm for minimum-cost Hamiltonian cycle in undirected graphs with arbitrary real edge costs and an independent set of size at least $αn$, for any fixed $3/7<α\le1/2$. It runs in $O^*(γ(α)^n)$ time for an explicit constant $γ(α)<2$; for bipartite graphs, the bound is $O^*((1+2\sqrt2)^{n/2})\approx O^*(1.957^n)$. The algorithm is based on a restricted Held–Karp dynamic program. Its state enumeration uses structural properties of minimum-cost assignments arising from a regular triangulation of a product of simplices.

comments
13 pages, 2 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.41178 ↗

Let $\mathbf{G}=\{G_1,\dots,G_{n-1}\}$ be a collection of not necessarily distinct graphs of order $n$ on the same vertex set $V$. A path $P$ with $V(P)\subseteq V$ is called \emph{rainbow} in $\mathbf{G}$, if there exists an injection $φ\colon E(P)\to [n-1]$ such that $e\in E(G_{φ(e)})$ for each $e\in E(P)$. In this paper, we show that if $n\ge3$, $e(G_i)> \binom{n-2}{2}+2$ for each $i\in[n-1]$ and $G_j$ is connected for some $j\in[n-1]$, then there exists a rainbow Hamiltonian path in $\mathbf{G}$ unless $n=6$ with $G_1=G_2=\dots=G_{5}\cong K_2\vee 4K_1$, or $n=8$ with $G_1=G_2=\dots=G_{7}\cong K_3\vee 5K_1$. This corrects Theorem 6 in [Graphs Combin., \textbf{41}(6)(2025), 119] and relaxes its hypothesis from requiring all $G_i$ to be connected to only requiring that at least one $G_i$ is connected, where $i\in[n-1]$. Moreover, we show that if $n\ge2$ and $e(G_i)\ge \binom{n-1}{2}$ for each $i\in[n-1]$, then $\mathbf{G}$ contains a rainbow Hamiltonian path unless $G_1=G_2=\dots=G_{n-1}\cong K_{n-1}\cup K_1$, or $n=4$ with $G_1=G_2=G_3\cong K_{1,3}$. As a consequence, Theorem 2 of [Discrete Math., \textbf{347}(2024), 114128] follows immediately.

fields
Combinatorics
pdf ↗ page arXiv:2610.41203 ↗

Delcourt and Postle recently introduced the `discharging in the dual' proof technique, leading to a breakthrough proof of Nash-Williams' conjecture on triangle decompositions. Independently, and almost simultaneously, the first and second authors built on the `condensation' method of Bryant, Dukes, Horsley, Maenhaut and Montgomery to determine the decomposition threshold of all odd cycles of length at least $73$. In this note, we explain how the `discharging in the dual' and `condensation' methods can be combined to determine the decomposition threshold of cycles of any odd length, as well as the approximate decomposition threshold of a range of other $3$-chromatic graphs.

comments
10 pages, 1 figure
fields
Combinatorics
pdf ↗ page arXiv:2610.41266 ↗

Let $T$ be a fixed tree on at least two vertices. We determine the Turán number and all extremal graphs for the Cartesian product of an odd cycle $C_{2m+1}$ and $T$, for every fixed integer $m\ge1$ and all sufficiently large $n$, in terms of its decomposition family. Every extremal graph is the join of an independent set and an extremal graph for that family. The proof combines the Erdős–Sós theorem with weak stability and progressive induction.

comments
16 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41286 ↗

We study inverse-closed generator additions of minimum size that make the optimal pebbling number of a finite Cayley graph at most three while preserving its specified regular action. A weighted vertex-cover formulation on inverse orbits yields an $O(|Γ|^2)$ algorithm for abelian groups in the explicit multiplication-table model. We derive formulas for central involutions and cubelike graphs, with an inverse-closure correction determined by square roots in the uncovered set. The exact thresholds are $2^{d-1}-d$ for hypercubes ($d\ge1$), and $3$ for standard Keller graphs when $d=2$ and $2^{d-1}$ when $d\ge3$. Every admissible augmentation of either family has ordinary pebbling number equal to its order.

fields
Combinatorics
pdf ↗ page arXiv:2610.41306 ↗

In this paper we establish the following result for clique-factors in randomly perturbed hypergraphs when the cliques are large enough. For integers $k \geq 3$ and $t\geq t_0(k)$, we show that for every $α> 0$, there exists $C>0$ such that for sufficiently large $n\in(t+1)\mathbb{N}$, the union of a $k$-uniform hypergraph on $n$ vertices with minimum codegree at least $αn$ and a binomial random $k$-uniform hypergraph $H^{(k)}(n,p)$ with $p\geq Cn^{-{t/\binom{t+1}{k}}}$ on the same vertex set contains a $K_{t+1}^{(k)}$-factor asymptotically almost surely. The bound on $p$ is asymptotically best possible and extends the graph case, resolved by Balogh, Treglown and Wagner. In particular, we have $t_0(3)=3$, which implies that our result applies to every $t\ge 3$ for $3$-uniform hypergraphs. Together with the previously established result for perfect matchings, this covers all possible clique sizes for $3$-uniform hypergraphs.

fields
Combinatorics
pdf ↗ page arXiv:2610.41340 ↗

For fixed integers $t,k\ge 2$, put $N=(t-1)(k-1)+1$. We prove that the maximum universe size of a $t$-perfect hash family of length $N$ over an alphabet of size $q$ satisfies $ p_t(N,q)\ge q^{k-o(1)}$. Together with monotonicity in the length, this gives a positive answer to a question of Blackburn, Etzion, Stinson and Zaverucha for every choice of feasible parameters. Our proof combines the Nash–Williams–Tutte theorem, a graph-theoretic elimination argument, and a solution-free-set theorem established in our previous work.

comments
17pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41353 ↗

Let us color the points of a projective plane of odd order $n$ with two colors, say red and blue. Erdős asked the question of determining the maximum number of balanced lines (that have the same number of red and blue points). The answer is $n^2$, and in the corresponding trivial coloring the unbalanced lines go through a point. In this paper, we consider the Galois plane $\mathrm{PG}(2,q)$ and determine the number of unbalanced lines for a non-trivial coloring. The problem can also be generalized to $(r,b)$-colored lines (see Definition 1.1) and in many cases we extend the results to the generalized problem. Our strongest result for Galois planes of prime order $p$ describes all 2-colorings with at most $\max\{3p+1,4p-22\}$ not $(r,b)$-colored lines (see Theorem 5.13). Finally, in case of the generalized problem, we allow even more not $(r,b)$-colored lines if $r$ is small and the order is prime (for more details, see Theorem 5.18).

comments
23 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41357 ↗

Fix an integer $r\ge 3$. Let $K_r$ denote the complete graph on $r$ vertices and $t_{r-1}(n)$ the Turán number for $K_r$. For each integer $k$ satisfying $0\le k \le \binom{n}{2}-t_{r-1}(n)$, $\rev{φ_r(n,k)}$ denotes the minimum, over all $n$-vertex graphs with $t_{r-1}(n)+k$ edges, of the maximum size of a $K_r$-packing \rev{(a family of edge-disjoint copies of $K_r$)} in the graph. The problem of characterizing the behavior of $φ_r(n,k)$ has a long history and dates back to Erdős for the case $r=3$. \rev{Győri} proved that $φ_r(n,k)=(1-o(1))k$ whenever $k=o(n^2)$. We characterize the range of $k$ for which $φ_r(n,k)=(1-o(1))k$, showing that this equality holds if and only if $k=o(n^2)$. Moreover, we obtain an \rev{asymptotic} upper bound (which falls below $1$) for $φ_r(n,k)/k$ whenever $k/n^2$ converges to a constant $δ$ satisfying $0 < δ\le \frac{1}{2(r-1)}$.

comments
10 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41371 ↗

Given a digraph \(D\), let \(c(D)\) be the largest size of a collection of openly disjoint directed cycles through a common vertex, and let \(c_r\) be the minimum of \(c(D)\) over all \(r\)-regular digraphs \(D\). In 2005, Seymour asked whether every \(r\)-regular digraph satisfies \(c(D)=r\). Mader answered Seymour's problem negatively for every \(r\ge 8\), but conjectured that nevertheless \(\lim_{r\to\infty}c_r=\infty\). Recently, Steiner confirmed this conjecture by proving \[ \left\lceil \frac{3r}{22}\right\rceil \le c_r \le 7\left\lceil \frac{r}{8}\right\rceil. \] In this paper, we first answer Seymour's problem negatively for $r\in \{6,7\}$, and then improve Steiner's bound to \[ \left\lceil \frac{r}{6}\right\rceil \le c_r \le 5\left\lceil \frac{r}{6}\right\rceil.\]

fields
Combinatorics
pdf ↗ page arXiv:2610.41372 ↗

The Clarke-Cormack-Burkowski lattice $\mathcal{E}_n$ is the lattice of antichains of intervals of the chain $\{0,1,\dots,n-1\}$, the algebraic structure underlying minimal-interval semantics, a query semantics used in information retrieval. We give an explicit, elementary anti-isomorphism between $\mathcal{E}_n$ minus its top element and the Stanley lattice of Dyck words of semilength $n+1$ (an anti-isomorphism whose existence is implicit in work of Ferrari and Munarini), under which the rank of an antichain is $\binom{n+1}{2}$ minus the area of the associated Dyck word. Hence the rank generating function of $\mathcal{E}_n\setminus\{\top\}$ is the reverse of the Carlitz-Riordan $q$-Catalan polynomial. Using a local limit theorem for the area of random-walk excursions, we confirm unconditionally an asymptotic formula of Takács: the size $λ_n$ of the largest level set of $\mathcal{E}_n$ satisfies $λ_n\simκ\,4^{n+1}/(n+1)^3$, where $κ\approx1.0807$. Finally, for the antichains of intervals of an arbitrary finite poset $P$ we study the sweep dimension, the minimum number of linear orders needed to separate every interval from every element outside it. It equals the Dushnik-Miller dimension of the lattice of convex subsets of $P$, it is the least $k$ for which $P$ maps into $\mathbb{R}^k$ so that every interval is the preimage of an upper orthant, and it lies between $\max(2,w)$ and $2w$, where $w$ is the width of $P$, both bounds being tight. Antichains of intervals embed, preserving joins, into $k$-dimensional staircase regions.

fields
Combinatorics
pdf ↗ page arXiv:2610.41385 ↗

Chromatic quasisymmetric functions and unicellular LLT polynomials give symmetric-function interpretations of graph orientations counted by sinks. For natural unit interval graphs, we prove that the acyclic sink polynomial is real-rooted for every nonnegative ascent weight. Equivalently, the elementary coefficients of the chromatic quasisymmetric function, grouped by partition length, form a real-rooted polynomial. The Carlsson–Mellit relation expresses this polynomial as a plethystic specialization of the corresponding LLT polynomial, and the Hessenberg correspondence gives an interpretation in cohomology. More generally, the acyclic sink polynomial is real-rooted for chordal claw-free graphs, and the all-orientations sink polynomial is real-rooted for every claw-free graph. We also study a real-rooted polynomial counting joint minima in local edge orders. It agrees with the acyclic sink polynomial of the line graph exactly for forests. An exact formula for unicyclic roots gives a line-graph counterexample to acyclic sink real-rootedness.

comments
30 pages, 5 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.41502 ↗

Equivariant Ehrhart theory was introduced by Stapledon (2011) and has since been extensively studied. A central question is whether the numerator in a certain rational function expression for the equivariant Ehrhart series is both a polynomial and effective. In this paper, we propose a modification of the denominator in the rational function expression for the equivariant Ehrhart series of rational polytopes. For rational simplices and rational cross-polytopes whose vertices have coordinates that are reciprocals of integers, we show that the resulting numerator is a polynomial and is effective. As an application of our results, we compute the equivariant series of the characteristic quasi-polynomials of root systems.

comments
17 pages, 1 table, no figures
fields
Combinatorics, Representation Theory
pdf ↗ page arXiv:2610.41529 ↗

We study self-orthogonal $t$-$(v,k,λ)$ designs satisfying the condition that the codes generated by the designs are self-orthogonal codes whose codewords of minimum weight $k$ correspond to the blocks. A classification of such designs is given under the conditions that $t\geq \lfloor k/4\rfloor+1\geq2$ and $t \le 11$. This yields a classification of self-orthogonal codes of minimum weight $k$ whose codewords of minimum weight form a $t$-design under the same conditions.

comments
31 pages
fields
Combinatorics, Information Theory
pdf ↗ page arXiv:2610.41545 ↗

Haiman conjectured that for every partition $λ\vdash n$ and every $w\in S_n$ the polynomial $c_{λ,w}(q)$, a monomial character of the Hecke algebra evaluated at the Kazhdan-Lusztig basis, has non-negative integer coefficients and is symmetric and unimodal, and he verified this by computer for $n\le7$. We verify it for every $n\le10$. A stability lemma reduces the statement to $n=10$, which we settle by an exact computation that stores only the 1,252,173,030 pairs $(x,w)$ in $S_{10}$ with $x$ maximal in its double coset for the descent sets of $w$, and takes the traces of the standard basis from the class polynomials of Geck and Pfeiffer. Every stored Kazhdan-Lusztig polynomial agrees with du Cloux's Coxeter3, and the coefficients pass several independent checks, including a comparison at all 16,796 codominant $w$ with chromatic quasisymmetric functions computed from colourings. Consequently, for $n\le10$ every monomial immanant is a non-negative integer combination of Kazhdan-Lusztig immanants, Stembridge's monomial immanant conjecture holds, and a class-function immanant is non-negative on totally non-negative matrices if and only if it is non-negative at the block matrices $J_μ$. We use this to describe the order of normalised irreducible-character immanants on totally non-negative matrices for $n\le10$, and to prove Pate's inequalities there. On the full span of permutation monomials we give a Newton-polytope criterion for the coefficient of $\mathrm{Imm}_w$ to be non-negative throughout the totally non-negative cone, show that for $n=4,5$ this happens exactly when $X_w$ is smooth, and check by exact computation for $n\le10$ the proof of this for all $n$ given in a companion paper.

comments
Companion to arXiv:2610.39021. Code: doi:10.5281/zenodo.23730750; data: doi:10.5281/zenodo.23532289
fields
Combinatorics, Representation Theory
pdf ↗ page arXiv:2610.41555 ↗

We characterize adjacency perfect state transfer in connected unicyclic graphs: the transfer vertices are opposite degree-two vertices of a four-cycle, and the two attached trees satisfy a characteristic-polynomial condition. Except for the four-cycle, the transfer pair is unique and the minimum transfer time is $π/\sqrt2$. Outside the four-cycle, we obtain sharp lower bounds of $46$ on the order and $10$ on the maximum degree, and classify all examples of maximum degree at most $20$. A finite inverse construction enumerates all realizations of a prescribed eigenvalue support at the transfer vertices. For supports of size at least seven, we establish the sharp maximum-degree lower bound of $21$. Every seven-element support $\{0,\pm\sqrt2,\pm u\sqrt2,\pm v\sqrt2\}$, with odd integers $3\leq u<v$, is realizable, and every such realization has attachment depth at most four. Explicit branching rules give a minimum-order recurrence. Among all connected unicyclic graphs admitting perfect state transfer with seven-element eigenvalue support, the minimum order is $218$, attained uniquely up to isomorphism.

comments
46 pages, 5 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.41572 ↗

Let $H$ be a simple $3$-uniform hypergraph on $n\ge7$ vertices, and put $τ_n=\binom{\lfloor(n-1)/2\rfloor}{2}$. We prove that $H$ has a Hamiltonian Berge cycle if its $2$-shadow is $2$-connected, every pair at Berge distance two has an endpoint of hypergraph degree at least $τ_n+1$, and every pair of codegree one has both endpoints above $τ_n$. The distance-two threshold is sharp, and the order range cannot be extended to $n=4,5,6$. The proof separates two structural arguments. A low-degree endpoint of an extremal longest Berge path forces Hamiltonicity, independently of the order and the degree threshold. For high-degree endpoints, injective charging and its equality cases force pairs of codegree one that permit further rotations. The cases $n=7,8$ are settled by exhaustive verification, with a completeness proof and a second implementation using direct vertex–hyperedge cycle search. Keywords: Hamiltonian Berge cycle; Fan-type condition; 3-uniform hypergraph; longest-path rotation; exhaustive verification.

fields
Combinatorics
pdf ↗ page arXiv:2610.41577 ↗

A graph is said to be 1-universally rigid if, for any generic realization of its vertices on the real line, the realization cannot be deformed out of the line in any higher-dimensional Euclidean space while preserving all edge lengths. In this paper, we investigate the sparsity of 1-universally rigid graphs in different manners. First, we show that an Erdős-Rényi random graph $G(n,p)$ is 1-universally rigid with high probability whenever $p \geq C\log n \cdot \log\log n/n$ for some sufficiently large constant $C>0$. These techniques also imply the existence of 1-universally rigid graphs of arbitrarily large girth, which resolves a problem suggested by Dantas e Moura, Jordán, and Silverman in a non-constructive manner. We also study how sparse a 1-universally rigid graph can be, proving that there exists an absolute constant $\varepsilon>0$ such that every sufficiently large 1-universally rigid graph on $n$ vertices has at least $(3/2+\varepsilon)n$ edges. This breaks the natural $3/2$ barrier arising from degree-2 extensions and makes progress towards a question of Jordán and Nguyen.

fields
Combinatorics
pdf ↗ page arXiv:2610.41581 ↗

A class of graphs $\mathcal{G}$ is tame if there is a constant c such that every $n$-vertex graph in $\mathcal{G}$ has at most $n^c$ minimal separators. Many important graph classes are tame, including chordal graphs, permutation graphs, and circle graphs. Several fundamental problems, most notably Max Weight Independent Set, can be solved in polynomial time on every tame class, which explains in a unified way their tractability on all these classes. We show that List 3-Coloring can be solved in quasipolynomial time on every tame class of graphs.

fields
Combinatorics
pdf ↗ page arXiv:2610.41620 ↗

We prove the numerical Kawamata–Viehweg conjecture for every matroid of rank three: the Euler characteristic of the inverse of an integral big and nef line bundle class is nonnegative. The proof reduces the conjecture to an inequality for a stone game.

comments
8 pages
fields
Combinatorics, Algebraic Geometry
pdf ↗ page arXiv:2610.41631 ↗

A proper coloring $\varphi$ of a graph $G$ is called proper conflict-free if, for every non-isolated vertex $v$, some color appears exactly once in its open neighborhood $N_G(v)$. For a nonnegative integer $k$, a graph $G$ is proper conflict-free $(\mathrm{degree}+k)$-choosable if, for every list assignment $L$ satisfying $|L(v)|\ge d_G(v)+k$ for each $v\in V(G)$, there exists a proper conflict-free coloring $\varphi$ of $G$ with $\varphi(v)\in L(v)$ for each $v\in V(G)$. We prove that every connected graph other than the $5$-cycle $C_5$ with maximum average degree less than $8/3$ is proper conflict-free $(\mathrm{degree}+2)$-choosable. This improves the maximum average degree bound $18/7$ of Kashima, Škrekovski and Xu, and removes the assumption $g(G)\ge7$ from the $8/3$ theorem of Qi, Song and Cao. As an immediate consequence, our theorem also yields that every planar graph of girth at least $8$ is proper conflict-free $(\mathrm{degree}+2)$-choosable.

fields
Combinatorics
pdf ↗ page arXiv:2610.41668 ↗

We obtain series of new completely regular codes in Cayley graphs of the groups $Z_q^n$ with the Lee metric. Combined with binary linear programming methods, the constructions lead to enumeration of all completely regular codes in such graphs for the groups $Z_q^4$, $q\leq 7$ up to parameter matrices.

fields
Combinatorics
pdf ↗ page arXiv:2610.41716 ↗

Phylogenetic networks provide a natural framework for representing reticulate evolutionary events such as horizontal gene transfer and hybridization. In the present manuscript, we investigate the identifiability of phylogenetic networks under (hardwired) parsimony, a natural extension of Maximum Parsimony from trees to networks. More precisely, we ask which structural features of a network can be uniquely inferred from so-called $A_k$-alignments under this criterion. While identifiability has been studied extensively for phylogenetic trees under parsimony, little is known about the identifiability of networks under hardwired parsimony. We take a first step towards addressing this gap by developing a general framework for studying identifiability under hardwired parsimony. We then use this framework to show that the tree of blobs of a phylogenetic network can be reconstructed from its $A_1$-alignment. This result can be viewed as a natural extension of Buneman's well-known splits equivalence theorem. Finally, we show that, under hardwired parsimony, cycles of length at least $5$ cannot in general be uniquely identified from character data, establishing a fundamental limitation of hardwired parsimony.

fields
Combinatorics
pdf ↗ page arXiv:2610.41725 ↗

Optimising the vertex correspondence can substantially increase the number of edges shared by two independent realisations of the same random graph. We quantify this phenomenon for uniformly random $d$-regular graphs and derive corresponding results for their edge-based graph edit distance. For every fixed $d\geq 3$, we show that the maximum overlap is $n+Θ_{\mathbb P}(n/\log n)$, where the leading term $n$ arises from Hamiltonicity, while the surplus captures the optimal order of the excess beyond it. We establish explicit upper and lower bounds on the surplus. The upper bound is obtained by counting common cores of minimum degree at least two, while the lower bound is established using spanning factors consisting of cycles of the same logarithmic length. For the special case $d=2$, we show that the maximum overlap is $n-(1+o_{\mathbb P}(1))\log n$. The additional overlap achievable is also studied by rotating aligned Hamilton cycles, using a contiguity argument. As a by-product, we determine the asymptotic maximum overlap between unions of independent random perfect matchings under cyclic shifts.

fields
Combinatorics, Probability
pdf ↗ page arXiv:2610.41735 ↗

For a positive rational number $x=p/q$, where $p$ and $q$ are coprime positive integers, let $\operatorname{den}(x)=q$. For a positive integer $n$, let $\operatorname{sd}(n)=\max\{\operatorname{den}(χ_f(G)): |V(G)|=n\}$. Let $D(n)=2^{-n}(n+1)^{(n+1)/2}$. It was proved by Alon, Tuza and Voigt that $\operatorname{sd}(n)\le D(n)$. In this paper, we prove that $\operatorname{sd}(n)\le D(n)n^{-\frac12\log_2 n+\frac14\log_2\log_2 n+O(1)}$ for all sufficiently large $n$. For the lower bound on $\operatorname{sd}(n)$, it was recently proved by Bonamy, Hylasová, Kaiser and Sereni that $\limsup_{n\to\infty}\operatorname{sd}(n)^{1/n}\ge\sqrt{2}$. We prove that $\limsup_{n\to\infty}\operatorname{sd}(n)^{1/n}\ge 3^{1/3}>\sqrt{2}$.

comments
18 pages, 5 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.41771 ↗

Let $σ_k(H)$ be the minimum sum of the vertex degrees over all non-edges of an $n$-vertex $k$-uniform hypergraph $H$. For every $k\ge3$, $n\geq ks$ and sufficiently large $n$, we establish a sharp Ore-type degree condition that forces a matching of size $s$ in $H$. Specifically, we prove that if \[ σ_k(H)>\max\left\{ k\left(\binom{n-1}{k-1}-\binom{n-s}{k-1}\right), (k-c)\binom{ks-2}{k-1}\right\}, \] then $H$ has a matching of size $s$, where $c=n-ks+1$. In particular, this proves a conjecture of Balogh, Palmer and Raeisi for $k\geq 3$ and sufficiently large $n$.

comments
19 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41787 ↗

This paper studies the Alon-Tarsi number of connected $4$-regular circulant graphs $C_n(s,t)$, where $1\le s<t<n/2$. We prove that if $n=\ell p^r$, where $p$ is an odd prime, $r\ge 1$, and $t\equiv\pm s\pmod{\ell}$, then $\operatorname{AT}(C_n(s,t))=3$ if $\ell$ is even, or if $\ell\ge 3$ is odd and $p\nmid 2^\ell-2$. We also prove that $\operatorname{AT}(C_{6m}(2,3))=3$ for $m\ge 2$, and $\operatorname{AT}(C_n(1,3))=3$ for $n\ge 8$.

comments
12 pages, 0 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.41820 ↗

A digraph is $h$-semicomplete if every vertex has at most $h$ non-neighbors. In semicomplete digraphs, $k$-triples are obstructions to bounded directed pathwidth with many algorithmic applications: Fradkin and Seymour [JCTB 2013] showed that a large triple contains a subdivision of every small digraph, and triples were used for irrelevant vertex arguments in routing problems by Fomin and Pilipczuk [JCTB 2019] and Gomes, Lopes, and Sau [ICALP 2025]. Fradkin and Seymour also showed that there is a function $f(k)$ such that every semicomplete digraph of directed pathwidth at least $f(k)$ contains a $k$-triple, and their proof gives $f(k)$ triply exponential in $k$. For $h$-semicomplete digraphs, a result of Kitsunai et al. [ESA 2015] also yields a $k$-triple, but only existentially, and finding one in fixed-parameter time remained open. We improve the bound on $f(k)$ to a single exponential and extend the result to $h$-semicomplete digraphs: we show that every $h$-semicomplete digraph of directed pathwidth greater than $(h+1) \cdot k \cdot 2^{27k+9}$ contains a $k$-triple, which we find in polynomial time, and no bound subexponential in $k$ or sublinear in $h$ is possible. We also extend the approximation algorithm for directed pathwidth of Fomin and Pilipczuk to $h$-semicomplete digraphs, so its certificates of large directed pathwidth remain valid. Our proof is much shorter than the analysis of Kitsunai et al., and while their algorithm is exact, ours runs in polynomial time. As an application, we show that Arc-Disjoint Paths on $h$-semicomplete digraphs is fixed-parameter tractable parameterized by the number of paths and $h$, generalizing the semicomplete case. Finally, unlike in semicomplete digraphs, directed pathwidth does not bound directed cliquewidth: for every $k$, some acyclic $1$-semicomplete digraph, so of directed pathwidth zero, has directed cliquewidth at least $k$.

fields
Combinatorics
pdf ↗ page arXiv:2610.41840 ↗

We study a problem posed by Kardoš asking whether the truncation of every connected simple bridgeless cubic graph has strong chromatic index six. The truncation $T(G)$ is obtained by replacing each vertex of $G$ with a triangle and attaching its incident edges to distinct vertices of that triangle. An affirmative answer was previously established for prism bases by Han and Cui. We prove that $\sidx(T(G))=6$ whenever $G$ is a connected simple cubic graph that is bipartite or is both bridgeless and claw-free. Both proofs are constructive and use a common compatibility criterion for combining colourings of smaller pieces. The bipartite result also extends to non-det-extremal multigraphs.

comments
26 pages, 5 figures, 4 tables
fields
Combinatorics
pdf ↗ page arXiv:2610.41950 ↗

For $1\le i\le j\le k$, let $gr^3_2(K_i, K_j, K_k)$ be the least $n\in \mathbb{N}$ such that every rainbow triangle-free $3$-coloring of $K_n$ (using red, blue, and green) contains a red/blue $K_i$, a red/green $K_j$, or a blue/green $K_k$. In this paper, we completely evaluate $gr^3_2(K_i, K_j, K_k)$ for all values of $i$, $j$, and $k$ in which $r(K_i,K_j)$ is known.

comments
25 pages, 4 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.41971 ↗

We disprove the analogue of Hedetniemi's conjecture for Shannon OR-capacity, which asserts that the capacity of a categorical product equals the smaller capacity of its factors. For every sufficiently large complete graph, we construct a second graph of larger capacity such that their categorical product has capacity strictly below that of the complete graph. Further constructions show that, for every real threshold above an explicit constant, both factor capacities can exceed that threshold while the product capacity lies below it. We also construct graph pairs, again with one complete factor, whose capacities tend to infinity while their product capacity is subpolynomial in the smaller factor capacity. Whether the product capacity must nevertheless tend to infinity whenever both factor capacities do remains open. The proofs use spectral interpolation, which also yields a single spectral point violating Hedetniemi's identity throughout the complete-factor family. Finally, we identify the categorical product as the distinguishability graph of a random-switching model for zero-error communication. In this model, our constructions show that switching can reduce the zero-error rate to a vanishing fraction of the smaller individual rate.

comments
46 pages: 18 pages main text including references and 28 pages supplementary appendices; 2 figures, 8 tables
fields
Combinatorics
pdf ↗ page arXiv:2610.41977 ↗

Every graph obtained from a regular Class $1$ graph by a vertex-splitting is overfull and hence is of Class $2$; the central question is whether the resulting graph is edge-chromatic critical. We consider a revised version of the Hilton-Zhao vertex-splitting conjecture for triangle-free base graphs, excluding the known cubic exception that arises from the Wagner graph. Let $G$ be a connected $n$-vertex triangle-free $Δ$-regular Class $1$ graph, and let $G^{*}$ be obtained from $G$ by splitting one vertex into two adjacent vertices. We first show that any non-extreme vertex-splitting produces a $Δ$-critical graph provided $Δ>\frac{n-1}{3}$. Consequently, for non-extreme vertex-splittings, the revised version of the conjecture is valid across its entire conjectured degree range, and indeed even under a slightly weaker degree condition. We next turn to arbitrary vertex-splittings, extreme ones included. It is shown that $G^*$ is $Δ$-critical provided $n \ge 10$ and $Δ>\frac{2n-3}{5}$. In addition, for $n>24$, the non-strict condition $Δ\ge \frac{2n-3}{5}$ is already sufficient. This extends the bipartite range $Δ>\frac{2n}{5}$ to the boundary range that the bipartite argument does not cover. In particular, our arguments give a complete treatment of the case of triangle-free $6$-regular base graphs in the conjectured range. The proofs bring together structural properties of full-deficiency pairs, homomorphism theorems for dense triangle-free graphs, and Hamilton decompositions built via graph detachment.

fields
Combinatorics
pdf ↗ page arXiv:2610.41984 ↗

We present the game of Matroid Bingo and prove that it gets more suspenseful the longer the game goes on.

comments
6 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41985 ↗

We determine all currently unknown values of the generalized two-color Ramsey number $R(K_{1,4},K_{m,n})$ for $\min(m,n) \leq 8$. We also prove that the number $R(K_{1,4},K_{m,m})$ equals $26, 29, 31, 33$ for $m = 9, 10, 11, 12$ respectively.

comments
18 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.41999 ↗

We construct an explicit 64-point subset of $\mathbb F_2^9$ which tiles by translation but has no linear-subspace tiling complement. Together with our earlier dimension-eight theorem and a product argument, this shows that every nonempty tile in $\mathbb F_2^d$ has a linear-subspace complement precisely when $d\le8$. The example is spectral, so it is not a counterexample to Fuglede's conjecture. Its construction uses six three-dimensional subspaces of $\mathbb F_2^6$, specified by a small table. We classify all its tiling complements and spectra: there are exactly 512 complements and 64 spectra, with one translation class of each. Every spectrum has normalized Wiener norm $15/8$. An appendix proves that Goyal's related construction is affinely inequivalent but has the same tiling complements and spectra. The construction and both classifications are proved from first principles. We also give elementary proofs of the positive assertion in dimensions at most seven and for all tile sizes other than sixteen in dimension eight. The remaining case uses our previously established theorem based on the classification of perfect binary codes of length fifteen.

comments
28 pages, 3 tables, no figures; includes a comparison appendix and ancillary Python verification scripts and data
fields
Combinatorics, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.42005 ↗

Let $\mathcal M$ be a finite spanning multiset of points in ${\rm PG}(k-1,q)$, and let ${\rm Char}_{\mathcal M}$ be its character function. For $1\le r\le k$, we study the maximum number of distinct values attained by ${\rm Char}_{\mathcal M}$ on the codimension-$r$ flats under a total-multiplicity budget. The spectrum is subject to two natural upper bounds, whose extremal cases lead respectively to $r$-maximum character spectrum ($r$-MCS) and $r$-full character spectrum ($r$-FCS) systems. We establish the existence of $r$-FCS systems in a broad parameter range, and solve the size-constrained extremal problem completely when $r=k-1$, including the minimum-cost multiplicity profiles. For $1\le r<k$, put $M_r=\qbinom{k}{r}$ and $B_r=\qbinom{k-1}{r}$. The minimum size $ν_r(k,q)$ of an $r$-MCS system satisfies \[\left\lceil\frac{\binom{M_r}{2}}{B_r}\right\rceil\le ν_r(k,q)\le \binom{M_r-1}{2}+\frac{q^k-q^{k-r}}{q-1}.\] The upper bound is at most $\binom{M_r}{2}$, strictly less when $r\le k-2$, and exact when $r=k-1$. Its proof applies positive additive labelings to the complement of one flat, whose trace is empty while all remaining traces are distinct. For $r=1$, we improve the known upper bounds on the minimum length of such systems. We also develop global moment identities and local flat-incidence bounds that impose strong structural constraints on large character spectra. Under the projective-system correspondence, complementary character spectra are the support-weight spectra of $r$-dimensional subcodes.

fields
Combinatorics, Information Theory
pdf ↗ page arXiv:2610.42006 ↗

A modulo $k$-orientation of a signed graph is an orientation in which the outdegree and indegree are congruent modulo $k$ at every vertex. We show that, for every odd integer $k\ge3$, every essentially $k$-unbalanced $(6k-10)$-edge-connected signed graph admits a modulo $k$-orientation. This improves the edge-connectivity bound of Zhu [JCTB, 112 (2015)] from $6k-7$ to $6k-10$.

fields
Combinatorics
pdf ↗ page arXiv:2610.42019 ↗

Treedepth is an important parameter in algorithmic graph theory. It is known that the problem of computing the treedepth of a graph is NP-hard. It has been an open problem whether treedepth can be computed in polynomial time when the input graph is restricted to be planar, or of bounded degree. Assuming $P \neq NP$, we answer both questions in the negative: Under polynomial-time Turing reductions, it is NP-hard to compute the treedepth of planar graphs with maximum degree 3. This degree bound is best possible, since treedepth can be computed efficiently on graphs of maximum degree 2.

comments
14 pages, 5 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.42033 ↗

In 2001, Berenstein and Zelevinsky presented a polyhedral model for the problem of computing tensor-product multiplicities for semi\-simple Lie algebras. We provide explicit linear equations and inequalities for Berenstein–Zelevinsky (BZ) polytopes of type-$\mathrm{G}_{2}$ Lie algebras. We prove that the denominator of a type-$\mathrm{G}_{2}$ BZ~polytope divides $180$, and that this bound is sharp, while the quasi-period of the Ehrhart quasipolynomial of such a polytope is always a divisor of $6$. We also show that BZ~polytopes of type-$\mathrm{G}_{2}$ Lie algebras can fail to satisfy Ehrhart positivity, in contrast to a conjectured property of BZ~polytopes for all classical types.

comments
35 pages, 10 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.42038 ↗

Given a simple undirected graph on $n$ vertices, random $k$-out sampling forms a spanning subgraph by having each vertex independently choose $k$ incident edges uniformly with replacement. We prove that, for every integer $k\geq2$, the expected number of original edges joining distinct sampled components is $O(n/k)$. This resolves the conjecture of Holm, King, Thorup, Zamir, and Zwick in its full range, improving their logarithmic sampling requirement to the optimal threshold of two choices per vertex. The bound fails for $k=1$, and the dependence $O(n/k)$ is tight up to constant factors. Our result holds with replacement and consequently also without replacement. In particular, this implies that contracting just two independent local choices per vertex suffices to reduce an arbitrary simple graph to a quotient with $O(n)$ edges in expectation, which is crucial in reducing the instance size for many contraction-based algorithms. Our proof uses switching arguments to relate intercomponent edges to bridges and a degree-weighted component potential, whose expectation we also bound with a switching argument.

fields
Combinatorics
pdf ↗ page arXiv:2610.42067 ↗

For an integer $k\ge 2$ and $r^2>2(2k-1)\ln (2k-1)$, we construct a bipartite graph $G$ whose two vertex classes are disjoint copies of the family of all $k$-subsets of a $(2k-1)$-element set, two vertices being adjacent when their sets intersect in at most $r$ elements. Assigning each vertex its corresponding subset as its color list, $G$ is not colorable, and thus its choice number $\ch(G)\ge k+1$. This implies that there are graphs with maximum degree $Δ$ and $$\ch(G)\ge\Bigl(\frac1{32}-o(1)\Bigr)\frac{(\lnΔ)^2}{(\ln\lnΔ)^3}.$$ When $k\to\infty$, consequently $\ch(G)/\lnΔ$ tends to infinity, which disproves the Alon–Krivelevich conjecture. In order to show the uncolorability, we use tools from probability (conditional product measure, nested coupling, and Doob martingale), which we write in an elementary language of weighted counting, and topology (Brouwer fixed-point theorem).

comments
17 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.42094 ↗

We consider Tutte's $3$-flow conjecture in the class of Cayley graphs on groups $G$ of order $p^2n$, where $p$ is a prime and $n$ is square-free and coprime to $p$. The conjecture has been proved by J. Zhang and Z. Zhang in the case where $n$ is a prime (Discrete Mathematics, 2023); and by the authors in the case where $p=2$ (arXiv:2603.24175). In this paper, we prove the conjecture in the case where $p > 2$ and either a Sylow $p$-subgroup or a Hall $n$-subgroup of $G$ is a cyclic group.

fields
Combinatorics
pdf ↗ page arXiv:2610.42099 ↗

Optimization and Control51

In this work, we study Burer-Monteiro factorizations of semidefinite programs with unit diagonal constraints. For a real symmetric cost matrix whose rank-width over $\mathbb{R}$ is at most $k$, we prove that every non-flat second-order stationary point of the factorized problem is globally optimal whenever the factorization rank is at least $2k$, independently of the matrix size.

comments
17 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.40872 ↗

We hereby fill in a gap in an induction argument used in the proof of a theorem on state-covariance controllability of linear stochastic systems in an earlier paper by the authors. The mathematical statement is true. The missing point in the proof concerns the assertion that, in the canonical single-input case, the control can be assigned arbitrary endpoint values. We amend the argument by proving a slightly stronger induction statement that permits arbitrary finite endpoint jets of the control. An independent approach to homogeneous state-covariance controllability was provided in a paper by Abdelgalil and Georgiou (2025) as a corollary of strong controllability results on $\GLp(n,\mathbb R)$, and shown herein that it extends to the case of additive noise as well.

comments
4 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.40888 ↗

Gradient flows of twice continuously differentiable strongly convex and convex objectives are, respectively, incrementally exponentially stable and non-expansive in the Euclidean metric. In this work, we investigate to what extent these properties extend to nonconvex objectives satisfying generalized Polyak-Łojasiewicz inequalities. For objectives with a unique minimizer, we establish conditions under which the gradient flow is contracting in a state-dependent Riemannian metric and incrementally exponentially stable. For objectives with nonunique minimizers, we generalize the non-expansivity of convex objectives to incremental stability for more general nonconvex objectives. We focus on incremental stability due to its implication to convergence to equilibrium points; when paired with a Lyapunov function it implies pointwise convergence to a minimizer and continuity of the endpoint map. Along the way, we derive new structural results for incrementally stable nonlinear systems and extend the analysis to Heavy-Ball and Dynamical Inertial Newton flows.

fields
Optimization and Control
pdf ↗ page arXiv:2610.40897 ↗

We introduce a fast novel algorithm to commit hydro-generators. We show that our algorithm can commit 56 generators for 200 days 10,000 times faster than Hexaly, the leading commercial solver for scheduling problems, without sacrificing solution quality. Our method can commit 1792 generators for an entire year in approximately 20 seconds on commodity hardware. Hexaly is unable to solve problems of this size in reasonable time. We attain this runtime improvement by introducing a greedy-heuristic tailored for hydro units. We provide a direction for a possible proof that the error our solver makes decreases with the problem size. This is supported by our experimental results. We prove that our solver's runtime scales log-linearly in the number of generators and time periods.

comments
5 pages (Conference limit), 5 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.40918 ↗

Can learned latent dynamics models support the design of feedback controllers that provably stabilize the original system? For linear time-invariant systems, we show that learning linear autoencoders and latent dynamics with multi-step prediction-reconstruction losses $\unicode{x2014}$ which we term "backward losses" as they encourage latent trajectories to reconstruct full state trajectories $\unicode{x2014}$ provably recovers the structure needed for stabilization. Specifically, a sufficiently long reconstruction horizon ensures that a minimum ${H}_\infty$-norm controller designed using the learned latent model stabilizes the original system through output-feedback via the corresponding encoder, provided the latent dimension is large enough to capture all non-stable modes of the underlying system. We also show why both reconstruction and a multi-step horizon are essential: forward prediction losses that require only low prediction error along latent trajectories admit global loss minimizers that discard unstable modes regardless of horizon length, while single-step reconstruction losses can yield latent models whose stabilizing controllers fail to stabilize the original system. Finally, experiments on a non-linear flexible-arm system illustrate the practical implications of our results for learning non-linear autoencoders and latent dynamics.

fields
Optimization and Control
pdf ↗ page arXiv:2610.40931 ↗

Childcare shortages have limited families' access to quality care, with consequences for children's development and parents' careers. We link two decades of childcare supply data to longitudinal surveys of U.S. scientists. We find workplace childcare deserts are widespread: more than half of scientists at universities and research institutions work in deserts, where supply falls significantly short of demand. We further find mothers working in deserts are less likely to attain tenure. They are also more likely to move to industry and part-time positions. Reduced work time partly accounts for the tenure association but other reasons remain. Expanding childcare availability near workplaces may help retain scientific talent.

fields
Optimization and Control
pdf ↗ page arXiv:2610.40951 ↗

A 2BSDE system is written down for a $(D+1)$-dimensional Black–Scholes partial differential equation (PDE) with the Leland transaction cost model which incorporates the sign of the curvature. The 2BSDE system takes input from the ``speed'' (derivative of the curvature). This system is solved using a single ResNet neural network which learns both the solution to the PDE as well as the degrees of freedom in the curvature matrix. The network is used to learn the price of derivatives such as call options and butterfly spreads on geometric baskets for $D=10,100$ achieving a time-averaged RMSE of less than $0.1$ in each case.

comments
Comments are very welcome!
fields
Optimization and Control
pdf ↗ page arXiv:2610.41019 ↗

Quadratic programs (QPs) are fundamental building blocks of many robotic tasks such as inverse kinematics, whole-body control, and model predictive control. Extending these methods to large-scale robot learning and control requires solving many QPs concurrently and, for end-to-end learning, differentiating through their solutions. Efficient batched optimization is therefore essential to bringing established model-based methods into GPU-based robotics pipelines. We present \textsc{cuPIQP}, a batched and differentiable QP solver on GPUs based on a proximal interior-point method, with unified support for dense, sparse, and multistage problems. The solver exploits parallelism across independent QPs and their internal structure while preserving per-problem numerical updates and termination decisions. Benchmarks with up to 16384 QPs per batch demonstrate solve-time speedups of up to $18.8\times$ over \textsc{Moreau} and $4.26\times$ over eight-core \textsc{PIQP}. In an end-to-end learning task, \textsc{cuPIQP} accelerates training by $4.6\times$ over \textsc{QPAX} and $7.8\times$ over \textsc{Moreau}. By reducing the computational cost, \textsc{cuPIQP} provides efficient, reusable infrastructure for integrating model-based optimization into large-scale robot learning and control.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41021 ↗

There exists little mathematical theory explaining why Transformer architectures are empirically successful even for non-Markovian data. To this end, this paper studies next-token prediction via stochastic optimal control theoretically trained Transformers for such sources, where we consider either stationary and ergodic non-Markovian sources or those which are hidden Markov models of some finite order (possibly non-stationary). In either case, we first obtain error bounds via finite models with finite window approximations. For stationary ergodic sources, the resulting error is bounded by an entropy gap; for $N$th order Markovian sources, it decays at an exponential rate $α^{\lfloor M/N\rfloor}$, where $M$ is the window length and $α$ is determined by filter stability (and the Dobrushin coefficients of the transition and measurement kernels). We then construct a Transformer of depth $M+1$, one head per layer, realizing the truncated filter, and we optimally train it with stochastic control developed in earlier work, which rigorously returns a near optimal Transformer. In particular, we show that for a non-Markovian source, either under stationarity and ergodicity for general models or under filter stability for hidden Markov models, an optimally trained Transformer has arbitrarily small performance error with increasing context length, data, and approximation refinement. Numerical studies accompany the theory.

comments
62 pages, 12 figures
fields
Optimization and Control, Probability
pdf ↗ page arXiv:2610.41038 ↗

Linear inequalities can generate large constraint systems in semidefinite relaxations of binary quadratic programs, making the quadratic subproblems of splitting methods expensive to solve. We develop a box-cone representation of bound-factor reformulation-linearization technique (RLT) constraints in semidefinite relaxations of binary quadratic programs. These constraints arise by multiplying each original linear inequality by every binary variable and its complement and linearizing the resulting products. Our key observation is that all products generated by one inequality share a common slack and admit an exact representation in the standard cone over the unit box. This box-cone representation processes the inequalities through independent explicit projections, without introducing a separate nonnegative slack for each scalar RLT inequality. We illustrate the computational use of the representation within standard two-block alternating direction method of multipliers (ADMM). The resulting subproblems require linear solves with a reusable factorization and explicit cone projections, without enlarging the semidefinite matrix. Experiments on 257 binary quadratic programs compare this implementation with SDPNAL+. A second application groups Max-Cut triangle inequalities into box cones and uses the resulting bounds within an existing branch-and-bound framework. Tests on 220 graphs show competitive performance with MADAM using only triangle cuts. Together, these applications illustrate the representation's usefulness as a building block for first-order methods.

comments
50 pages, including 32 pages of supplementary numerical results
fields
Optimization and Control
pdf ↗ page arXiv:2610.41051 ↗

We consider the situation that the grand set of all players, in a cooperative game in characteristic function form with a nonempty core, consists of recognizable subsets of players. The underlying assumption is that the players of each subset have something in common between themselves which is not shared by the players of other subsets. For example, the members of these subsets are distinguishable by their nationalities or by their geographic locations or by some other market-related economic or social property. In addition to the agreement of all players to play the game to reach a point in the core, the members of each subset want to steer the iterative orbit of the bargaining process to a point in the core that will reflect the strength of their solidarity within their subset. This is why we call such subsets solidarity sets. The solidarity means that players of a solidarity set are willing to participate in coalitions that include only members of their solidarity set and are not willing to participate in coalitions of other solidarity sets. The question is how can such a game be played? Is there an iterative bargaining scheme that will generate a sequence of payoffs that converges to a point in the core which reflects the solidarity of the members of the solidarity sets? We propose bargaining schemes that allow the members of each solidarity set to perform, independently from other solidarity sets and, therefore, simultaneously in time, a full but finite sweep of intercoalitional collaboration. Then the end-points of all these sweeps are combined into a single payoff vector via a convex combination wherein the coefficients of the convex combination may reflect some a priori knowledge about the relative strengths of the different solidarity sets. These bargaining schemes are applied to a new type of cooperative game that we call solidarity cooperative game (SCG).

fields
Optimization and Control
pdf ↗ page arXiv:2610.41062 ↗

Much of the existing convergence guarantees and rate statements for stochastic variational inequality (SVI) problems are limited to stochastic approximation schemes that employ constant or diminishing steplength sequences bounded by some reciprocal of the Lipschitz constant $L$ of the map. This requires both knowing such constants and may lead to debilitatingly small steplengths, afflicting the scheme with potentially poor performance. We propose a parameter-free stochastic extragradient method with periodically restarted linesearch (\textbf{SEG-LS}) for solving pseudomonotone and monotone SVI problems. The resulting stepsize sequence is neither monotone nor requires prior knowledge of the Lipschitz constant $L$. We show that the iteration and sample complexities for computing a point whose residual has norm at most $ε$ in expectation are $\mathcal{O}(ε^{-2})$ and $\mathcal{O}(ε^{-4})$, respectively. These results improve to $\mathcal{O}(ε^{-1})$ and $\mathcal{O}(ε^{-2})$ in the monotone setting. In addition, we propose a periodically restarted biased stochastic linesearch method (\textbf{B-SLAM}) for solving non-monotone SVIs under strict copositivity. The resulting iteration and sample complexities are $\mathcal{O}(ε^{-2})$ and $\mathcal{O}(ε^{-4})$, respectively. Preliminary numerics support these findings and display promise.

comments
43 pages, 4 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.41065 ↗

We review models of inexact computational oracles in smooth continuous optimization and their implications for the convergence, computational complexity, and attainable accuracy of optimization algorithms. We adopt a framework in which zeroth-,first-, and second-order oracles are characterized by quantitative error bounds and may accept additional control parameters that influence both their accuracy and computational cost. We introduce deterministic and stochastic oracle models and illustrate their properties through a variety of examples, including finite-difference approximations, stochastic gradients, and randomized derivative estimators. Particular attention is given to the distinction between absolute and relative errors, bias and variance, and stochastic function evaluations with independent or coupled randomness. A central theme of the review is the use of common analytical tools to recover and compare convergence and complexity guarantees under different oracle models. For deterministic and stochastic first-order methods, we show how a general descent inequality leads to complexity bounds under various assumptions on the objective function and all proposed oracle models. In the stochastic setting, we develop bounds on expected stopping times and the accuracy that can be guaranteed under different oracle conditions. We also examine the total computational complexity of algorithms relying on specific oracle constructions and briefly review models of inexact second- order information and their use in optimization algorithms.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41086 ↗

An evaluated candidate can improve a local search step, alter the population used by later steps, and affect a solver's final result. We distinguish these three questions using Da Gou Gun Optimization (DGO), a population optimizer designed for this study. Its JQ100 configuration evaluates an interpolation candidate during a two-sided probe. We define local budget value, population commitment effects and complete-policy performance, and give an exact example in which a locally better commitment leads to a worse continuation. Short-horizon matched comparisons favor interpolation on eligible quadratic states. Paired continuations yield mixed effect estimates, with primary group intervals spanning zero. Complete-policy comparisons favor JQ100 over the tested classical references in aggregate; a matched-start comparison has favorable means but intervals spanning parity. On 24 boxed BBOB functions, target-truncated geometric-mean error ratios favor JQ100 over the tested classical policies and favor L-SHADE and CMA-ES over JQ100 under the stated initialization and stopping rules. Local evaluation gains, commitment effects and solver competitiveness therefore require separate comparisons.

comments
44 pages, 12 figures, 20 tables, including appendices
fields
Optimization and Control
pdf ↗ page arXiv:2610.41099 ↗

Continuous-time exploratory stochastic control provides a stochastic control framework for the exploration–exploitation trade-off in reinforcement learning. In this paper, we study the long-time behavior of finite-horizon exploratory stochastic linear-quadratic (LQ) optimal control problems as the time horizon $T\to\infty$, and establish so-called turnpike properties for such problems. In the finite-horizon setting, we derive a feedback representation of the exploratory stochastic LQ problem and analyze its long-time structure. We formulate an exploratory static distribution optimization problem analogous to the classical static optimization problem and identify the corresponding static reference object. We prove a turnpike property for the optimal state process relative to a stochastic reference process, as well as a Wasserstein turnpike property for the optimal control distribution. In addition to the exploratory control formulation, another key feature of our results is that, by introducing a joint coercivity condition, we allow the cost functional to be indefinite and the controlled state system to be merely stabilizable. Consequently, the derived turnpike property is novel even within the classical stochastic LQ control framework.

comments
35 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.41139 ↗

While the Koopman operator-based lifted representations of nonlinear autonomous and controlled systems have been well covered in the literature, the direct controller synthesis problem largely remains open. As the controlled system is generally a bilinear (instead of fully linear) Koopman model, controller synthesis appears to lose the linear–quadratic structures of the classical Lyapunov and Riccati methods for finite-dimensional linear systems. However, in infinite dimensions, bilinearity in the feature map should be merely artifactual and hence reducible to linearity in principle. In this work, by considering the operator semigroups in the reproducing kernel Hilbert space with a linear–Sobolev kernel and utilizing the Yosida approximation of their generators, the synthesis problems of a stabilizing controller and an input-unconstrained optimal controller can be formulated as an operator Lyapunov inequality and an operator Riccati equation, respectively. These formulations recover the desirable linear–quadratic structures, and hence, with sample data, they are solvable efficiently in analogous ways to usual linear systems by semidefinite programming solvers and Kleinman iterations, respectively. In addition to proving control performance bounds dependent on the sampling denseness and Yosida regularization, through numerical comparison to linearized LQR, bilinear EDMD-based synthesis, nonlinear MPC, and Koopman MPC methods on numerical models and process control examples, this work demonstrates the proposed Koopman–Riccati synthesis as a computationally efficient approach to obtain a near-optimal state-feedback control law.

comments
56 pages, 9 figure, 5 tables
fields
Optimization and Control
pdf ↗ page arXiv:2610.41141 ↗

Classical exactness conditions for conic relaxations of the DistFlow model rely, among other assumptions, either on all nodal injections being adjustable or on a network-wide path condition. In practice, these conditions may not hold uniformly across a feeder: controllable distributed energy resources (DERs) sit at some buses and fixed loads at others, while reverse flows from DER-rich regions can violate the path condition. This paper certifies exactness region by region and composes the regional certificates, accounting for perturbations that propagate downstream through voltages. We first introduce a subtree voltage stability region, in which the network and all compatible subtrees remain regular. Within it, two endpoint sign conditions, one relaxing the nonnegative-flow assumption and one implied by the classical path condition, yield monotone responses of voltages, currents, and losses to root voltages. Using duality, the implicit function theorem, and sign properties of inverse sensitivity matrices, we recast both classical arguments as individual sign certificates on branch multipliers. This turns exactness into a branchwise property that can be certified region by region. A partition rule then splits the network into dispatchable and passive subtrees, and a hybrid certificate combines their local conditions into exactness of the full-network relaxation, under additional operating-limit and objective conditions. In particular, placing controllable DERs at both ends of a branch with large reverse flow exempts that branch from the passive sign conditions.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41220 ↗

We study the non-asymptotic convergence of Stochastic Mirror Descent (SMD) algorithm when the noisy first-order oracle is driven by dependent observations. Departing from the classical i.i.d. assumption underlying most existing guarantees, we introduce a flexible ``forgetting" condition that accommodates geometrically or polynomially dependent sequences. Under this condition we derive explicit finite-sample accuracy bounds for the averaged SMD iterates, recovering the classical $O(1/T)$ and $O(1/\sqrt{T})$ rates in the large- and small-initial-error regimes. We instantiate the general framework on two applications: high-dimensional linear regression with dependent regressors and nonparametric estimation of the drift function of a functional autoregressive model.

comments
31 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.41251 ↗

In this paper, we propose a sparse descent direction derived from the Newton direction within a low-dimensional subspace for sparse optimization problems. Building upon this direction and incorporating hard thresholding and orthogonal projection, we develop a novel algorithm termed the sparse descent subspace-Newton thresholding (SDSNT) algorithm. We further establish its global convergence under the restricted isometry property $δ_{3k}\leq \sqrt{3/20}$ of the measurement matrix. Numerical experiments on random sparse vector recovery and audio signal reconstruction reveal that SDSNT achieves comparable performance to several state-of-the-art algorithms. Notably, relative to most of them, it delivers superior recovery at high sparsity levels and improved efficiency at low sparsity levels.

fields
Optimization and Control, Information Theory
pdf ↗ page arXiv:2610.41318 ↗

Differentiable optimal control embeds constrained optimization problems in data-driven pipelines by differentiating their optimality conditions and propagating task-loss gradients to model or objective parameters. A key observation is that forward optimization and backward sensitivity computation involve related stage-structured linear systems, making reuse of forward factorization information natural. However, direct reuse is generally invalid, because differences in linearization, barrier terms, regularization, and curvature can create a mismatch between the retained forward operator and the target operator defining the derivative. We therefore propose RapidDOC, a factorization-reuse method that leaves the target adjoint equation unchanged. The retained cache produces a candidate adjoint and, when its target residual exceeds the prescribed tolerance, serves as a fixed right preconditioner for GMRES. A cache-action error operator characterizes directional exactness, the preconditioned iteration, and the relation between target residual and outer-gradient error. An ablation on three constrained robotic OCPs shows that the proposed method achieves lower backward time than both target-system refactorization and a no-reuse iterative method, demonstrating the computational benefit of reusing the retained forward factorizations. Across open- and closed-loop parameter-learning experiments, the proposed method achieves learning performance comparable to the baselines, while yielding the lowest open-loop backward time on all three tasks and reducing total closed-loop solver-call time by 40.5-83.1% relative to the fastest competing method. Finally, deploying the learned controller on a physical vehicle reduces the median figure-eight tracking RMSE from 22.75 to 6.09 cm, demonstrating its potential for hardware deployment.

comments
8 pages, 6 figures, 2 tables. Code: https://github.com/opt-man/RapidDOC ; Supplementary video: https://youtu.be/b53gcrjdS7c
fields
Optimization and Control
pdf ↗ page arXiv:2610.41324 ↗

Accelerated Riemannian optimization methods often require a priori assumptions that their iterates remain in a region where geometric bounds hold. We develop a parameterized inertial dynamical system and an explicit accelerated gradient method for smooth geodesically convex optimization on Riemannian manifolds with locally bounded sectional curvature. Our setting allows both positive and negative curvature without requiring completeness of the ambient manifold. Starting from a curvature-corrected continuous model, we construct the discrete updates by eliminating indefinite cross terms in the Lyapunov energy difference and derive verifiable conditions for a fixed step size. The resulting energy analysis establishes preservation of the initial geodesic ball without assuming iterate containment in advance. We further establish accelerated convergence rates for both geodesically convex and strongly geodesically convex objectives, without requiring restarts.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41331 ↗

We study Riemannian optimization on an unknown manifold accessed only through feasible samples. The central challenge is to turn feasible samples into optimization steps, i.e., the samples are drawn from the unknown feasible manifold, but do not directly provide the tangent spaces, projections, or retractions required by classic Riemannian methods. Moreover, regions with low objective values may be sparsely represented in the data, so directly sampling from a distribution fitted to these observations can require many draws to find better feasible solutions. We introduce Proximal Refinement on Implicit Smooth Manifolds (PRISM), which turns feasible-sample access into an iterative optimization procedure by repeatedly sampling an objective-aware proximal distribution conditioned on the current iterate. Theoretically, we show that PRISM recovers the effect of Riemannian-gradient refinement, leading to provable objective descent, feasibility (manifold adherence), and convergence toward first-order stationarity. We demonstrate the effectiveness of our approach through synthetic experiments illustrating its geometric properties and applications to data-driven building control and diverse real world optimization tasks, highlighting its potential as a general framework for data-driven optimization across scientific and engineering domains.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41379 ↗

Integrated railroad blocking is a central planning problem in freight railway operations, which requires coordinated decisions on block design, blocking sequences, and block routing. We propose a dual-guided column-generation-based matheuristic (DG-CGM) that uses linear programming (LP) dual information to identify promising transportation paths and incorporates them into restricted mixed-integer programs to improve feasible solutions. Starting from an initial plan built by construction, DG-CGM alternates destroy and local repair with path generation, which uses dual prices from restricted LPs of a Dantzig-Wolfe reformulation to identify promising transportation paths. When local improvements continue to stagnate, we further apply joint repair to combine alternatives across demand groups. To address the coupling between logical and physical path pricing, we introduce a joint reduced cost that eliminates path-specific dual variables from pricing, enabling more efficient column identification through exact labeling algorithms. We evaluate the performance of DG-CGM on both generated instances and real-world cases. On 150 generated instances with 50-150 nodes, DG-CGM achieves the optimal cost on 89 of the 91 instances solved to proven optimality by Gurobi, with a geometric mean runtime equal to 5.2% of that for Gurobi and a mean relative cost reduction of 3.4%. On six full-scale scenarios derived from the 2026 INFORMS Railway Applications Section competition data, each with 47,193 nodes, DG-CGM obtains feasible solutions within two hours that serve 95.9%-99.6% of the requested freight volume, providing practical blocking plans for large-scale railway operations. The code, instances, and computational results are publicly available at https://github.com/Zhengzhong-You/IRBP.

comments
49 pages, 5 figures, 14 tables, including a 15-page electronic companion
fields
Optimization and Control
pdf ↗ page arXiv:2610.41403 ↗

We study deterministic and stochastic optimization over the Stiefel manifold, explicitly allowing iterates to deviate from the manifold rather than enforcing feasibility via retraction at each step. We introduce Landom, a Landing-type method that constructs momentum in the relative-gradient direction, maps this update to its compact polar factor, and combines the resulting tangential step with a step along the negative gradient of the orthogonality penalty to improve feasibility. Under operator-to-nuclear smoothness, Landom achieves an $O(T^{-1/2})$ bound on the averaged nuclear norm of the relative gradient with exact gradients, and an $O(T^{-1/4})$ bound in expectation with stochastic momentum, bounded variance, and a fixed batch size. Our analysis also provides explicit pathwise bounds on the orthogonality residual, including its dependence on the step size and initial infeasibility. To address the coupling between objective and feasibility correction in Landom, we further propose Landom+, a two-timescale variant in which the tangential and normal step sizes are decoupled. By keeping the normal step size fixed, Landom+ achieves tighter control over the orthogonality residual following an initial transient while maintaining the respective convergence rates for stationarity. Numerical experiments across a range of applications illustrate that the proposed algorithms remain competitive with established manifold optimization approaches.

comments
32 pages, 9 figures
fields
Optimization and Control
pdf ↗ page arXiv:2610.41406 ↗

Percutaneous needle insertion is a common practice in minimally invasive procedures whose clinical success depends on accurate needle placement. However, achieving the necessary precision is a challenge for manual operators. Robotic assistance can improve targeting accuracy and reduce operator dependence, but residual millimeter-scale errors persist because nominal kinematic models do not fully represent the physical manipulator. This paper presents a robust bound-constrained kinematic calibration method designed to improve positional accuracy for a 6-Degree-of-Freedom (6-DoF) robotic needle-guidance manipulator. The approach uses physically plausible parameter ranges to regularize a compact 19-parameter geometric identification model. By treating these bounds as active optimization constraints rather than mere numerical safeguards, the framework effectively mitigates parameter coupling and overfitting, ensuring highly stable parameter identification. In simulation, the proposed method reduced end-effector Root Mean Square Error (RMSE) from 3.422 mm to 0.690 mm (a 79.8% improvement). In benchtop hardware experiments, the calibrated model achieved sub-millimeter validation accuracy, including 0.865 mm RMSE in experimental trajectory validation and 0.610 mm RMSE in a coordinate-lattice verification task. These results demonstrate that physically constrained kinematic calibration can enhance robotic needle-targeting accuracy, establishing a high-precision foundation for future clinical applications.

comments
6 pages, 4 figures, 4 tables
fields
Optimization and Control
pdf ↗ page arXiv:2610.41429 ↗

We revisit the relation between discrete- and continuous-time stationary dynamic optimization, using both the Bellman and Hamiltonian descriptions. We study the problem of embedding discrete Bellman–Lax–Oleinik operators and the associated Hamiltonian cascades into continuous-time dynamics. Rather subtle obstructions to such embeddings emerge even under the most natural assumptions.

comments
Dedicated to the memory of Zvi Artstein. 43 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.41436 ↗

We study the stability of ideal Nash equilibria in generalized multi-criteria games, in which each player's vector-valued payoff is ordered by a closed cone and the feasible strategy sets depend on the other players' strategies. Under compactness and continuity assumptions, the ideal Nash equilibrium correspondence is upper semicontinuous on the class of games admitting an ideal equilibrium. Our main result does not require the ordering cones to be convex or pointed. In fact, the ideal equilibria are exactly the common equilibria of associated scalar games when all the cones are convex. As consequences of our results, the stability of interval-valued games is derived and existence theorems for single-leader-multi-follower problems are proved for both optimistic and pessimistic sense.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41439 ↗

Non-convex optimization is central to machine learning and AI, yet unlike its convex counterpart, it still lacks a general and principled framework. In this work, we try to highlight the importance of combining noisy perturbations with over-parametrization mathematically under the classic setting of matrix completion (MC). Matrix completion is an important non-convex recovery problem in which only a subset of a matrix's entries is observed, and the remaining entries must be recovered using low-rank constraints. This problem is notoriously difficult because observations are limited. However, we show that by injecting $ε$-level perturbations into the nullspace of the observation mask, we can construct noisy matrix sensing surrogates whose solutions remain $\mathcal{O}(ε)$-close to the ground truth with valid restricted isometry property (RIP) constants. The existence of RIP further enables powerful tensor frameworks to be applied with theoretical guarantees. This approach reflects a broader empirical trend in which stochasticity or noise improves the curvature of the optimization landscape, allowing over-parameterized (a.k.a large) models to better separate signal from noise. Assuming each matrix entry is observed independently, we establish quantitative recovery guarantees without explicit requirements on sampling rate or incoherence, and discuss how such assumptions can further strengthen our results.

comments
30 pages, accepted at NeurIPS'26
fields
Optimization and Control
pdf ↗ page arXiv:2610.41455 ↗

In periodic-review stochastic inventory control, replenishment policies must balance holding and backlogging costs while responding to uncertain demands and evolving forecasts that may be arbitrarily correlated and nonstationary. In this general information setting, Levi, P{á}l, Roundy, and Shmoys (Mathematics of Operations Research 2007) established that the fully adaptive optimum can be approximated within factor $2$, which had remained the best-known performance guarantee since then. Their marginal cost accounting and dual-balancing framework made this comparison possible without computing optimal decisions across future trajectories, opening a broad research program on approximation algorithms for stochastic inventory systems. We break this long-standing barrier for periodic-review stochastic inventory control, showing how to compete against the fully adaptive benchmark within factor $\frac{ e }{ e-1 } \approx 1.582$ while retaining general demand correlations and evolving forecasts. We attain these advances by discovering a structural connection between marginal inventory accounting and weighted stopping problems, where each unit's conditional holding charge forms a supermartingale, while postponing its order incurs a history-dependent backlog charge. To leverage the random thresholding approach of Ma and Tzamos (ICML 2023) in this context, we develop a potential-function analysis, leading to an $\frac{ e }{ e-1 }$-approximation with respect to any unit's stopping problem. Subsequently, our common-threshold policy coordinates these marginal decisions into a global replenishment policy without additional loss.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41475 ↗

A polynomial optimization problem is said to have an SPLD structure if the polynomials defining its objective and constraints admit decompositions into separable and nonseparable parts, with the maximum degree of the separable parts exceeding that of the nonseparable parts. In this paper, we propose a bounded degree SDSOS hierarchy for globally solving a class of such problems. We establish the asymptotic convergence of this hierarchy over compact feasible sets and provide a sufficient condition for its finite convergence. Numerical experiments compare the proposed hierarchy with the BSDSOS hierarchy (Chuong et al. in J. Global Optim., 75(4):885-919, 2019).

fields
Optimization and Control
pdf ↗ page arXiv:2610.41483 ↗

Robust principal component analysis concerns the recovery of a low-rank matrix $Z$ and a sparse matrix $S$ from their observed sum $D = Z + S$. In this paper, we study this recovery problem under an $\ell_0$ sparsity penalty and a rank constraint on the low-rank component. By exploiting the entrywise separable structure of the sparse subproblem, we obtain an exact closed-form solution via hard thresholding with box clipping, avoiding the shrinkage bias of $\ell_1$-based methods. The low-rank subproblem is solved by truncated SVD. We prove block-coordinate convergence of the resulting alternating minimization scheme, establish uniqueness of block minimizers under non-degeneracy and singular value gap conditions, establish local linear convergence under a contraction condition on the Fréchet derivative of the truncated SVD operator restricted to the sparse support, and provide a second-order optimality analysis via the Riemannian Hessian. Experiments on synthetic data, images, and video show that the proposed algorithm outperforms the existing convex and nonconvex algorithms, most notably in support recovery and in the accuracy-efficiency trade-off.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41492 ↗

This paper focuses on the use of learning nonlinear operators via Deep Operator Networks (DeepONets) for approximating the solutions of optimal control problems. We study two operator learning strategies, namely, indirect and direct, both based on DeepONets, for parametric families of optimal control problems where the same problem must be solved many times for different initial data, different parameters, different cost function weights, and different bounds defining the set of the admissible controls. In the indirect strategy two DeepONets approximate the solution operators of the state and adjoint systems, and their predictions are used inside iterative projected gradient algorithms for finding the optimal control. In the direct strategy, a single DeepONet approximates the operator that maps directly to the optimal control generated by a reference solver; consequently, a new control is obtained through a single evaluation of the network. In order to analyze the computational cost, the efficiency, the speed and the accuracy of the learning strategies, we consider an optimal control problem related to an ordinary differential system that describes the temporal evolution of the glioblastoma growth subject to therapeutic interventions. For this specific optimal control problem we propose a discretize-then-optimize strategy, based on an implicit-explicit finite difference method, and then, we develop several numerical experiments in order to compare the two DeepONet learning strategies for this problem. The proposed operator learning DeepONet strategies can be adapted to approximate large classes of optimal control problems related to diverse differential systems.

comments
38 pages, 9 figures, 12 tables
fields
Optimization and Control
pdf ↗ page arXiv:2610.41532 ↗

The performance of many distributed optimization algorithms over peer-to-peer networks strongly depends on the choice of their hyperparameters. However, existing selection rules often require global quantities, yield conservative choices, or are tied to specific algorithms and hyperparameters. We propose Distributed Grid Search (D-GS), a consensus-based search method that evaluates and selects hyperparameter configurations across the network, requiring only the ability to run the tuned algorithm and evaluate local gradients. We also propose a zoom-in variant, Distributed Zoom Grid Search (D-ZGS), which improves empirical performance at a reduced cost. We prove that both methods terminate in finite time with all agents holding the same hyperparameter configuration, and quantify the (sub)optimality of this configuration. Numerical experiments on linear and logistic regression show that D-GS closely matches a centralized reference and that tuning and deployment cost a fraction of deploying algorithms with hyperparameters tuned using theoretical results.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41537 ↗

The Fixed Cardinality Binary Quadratic Program (FCBQP) is an optimization problem that requires a prescribed number of variables to take the value 1. We propose a parallel CPU-based heuristic that combines impact-based variable selection, tabu search, guided restarts, and a shared elite pool. Computational experiments on 342 benchmark instances from the maximum dispersion problem, k-clique problem, and minimum graph bisection problem show that the new algorithm is competitive and can reach best-known or reference values on instances with more than 20000 variables within one minute.

comments
7 pages, Proceedings for the Annual International Conference of the German Operations Research Society (GOR2026)
fields
Optimization and Control
pdf ↗ page arXiv:2610.41546 ↗

Collision-free motion planning for multi-agent robotic systems remains an active research area because of the complexity of coordinating robot trajectories, with applications ranging from coordinated drone shows to autonomous operations on land, in the air, and on water. Importantly, the analytical difficulty of nonconvex collision-avoidance constraints has so far driven the development of exclusively numerical methods. We derive a complete analytical characterization of the optimal trajectories for two robots that start and end at rest, under double-integrator dynamics and a squared-acceleration objective. We prove that these trajectories belong to eight distinct contact families and determine which family applies for each configuration of initial and final positions. We analyze the geometrical properties of the optimal trajectories, provide a numerical algorithm for computing them, and discuss how these insights may improve the performance of existing numerical trajectory optimizers.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41558 ↗

Quadratic Unconstrained Binary Optimization (QUBO) problems and their variant with a fixed cardinality constraint are both well-known NP-hard problems. While certain structural properties of the cost matrix can make some QUBO instances solvable in polynomial time, this advantage typically disappears once a cardinality constraint is added. In this work, we investigate specific structural conditions on the cost matrix under which the constrained version-quadratic binary optimization with a fixed cardinality-remains tractable.

comments
7 pages, it was accepted for publication for Operations Research Proceedings 2025/ Annual International Conference of the German Operations Research Society
fields
Optimization and Control
pdf ↗ page arXiv:2610.41560 ↗

Finding optimal transport protocols for underdamped stochastic processes is a difficult yet important problem in control theory. In this paper we develop a direct method, based on machine learning, to find the optimal solution of the problem in the presence of non-linear boundary conditions, including those modelling the Landauer bit-erasure problem. The dynamics is steered from a single-peak to a double-peak distribution in position space, by only acting on a controllable external potential. The problem is highly non-linear, and analytical solutions are not available in general. We benchmark our numerical results with a perturbative expansion, which is expected to hold in a suitable overdamped limit.

comments
Main: 12 pages, 6 figures. Supplemental material: 11 pages
fields
Optimization and Control
pdf ↗ page arXiv:2610.41599 ↗

We study decentralized composite finite-sum optimization over a fixed connected undirected network, where agents access local component functions through function evaluations and share a possibly nonsmooth convex regularizer. In this setting, standard two-point gradient estimators can retain nonvanishing variance at a solution, complicating their integration with gradient tracking. We propose Distributed Zeroth-Order Incremental Variance Reduction (Distributed ZIVR), which combines incrementally updated component-gradient estimates with a proximal gradient-tracking scheme. The method admits an implementation using a constant number of function evaluations per agent at every iteration, without intermittent full-component or full-coordinate refreshes. For convex smooth component functions and strongly convex local averages, we establish linear convergence up to a controllable finite-difference error. With the finite-difference radius chosen for the target accuracy, Chebyshev-accelerated mixing yields a per-agent oracle complexity of $\widetilde{O}(d(m+κ)\log(1/ε))$ for expected squared solution error $ε$, where $d$ is the dimension, $m$ is the number of components per agent, and $κ$ is the condition number. This matches the centralized ZIVR oracle bound, while the communication bound has only an inverse square-root dependence on the network spectral gap. Experiments on logistic and robust regression demonstrate rapid reduction of solution and consensus errors.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41615 ↗

Autonomous inspection of a target spacecraft is typically the initial step in on-orbit servicing. Planning close-proximity observation viewpoints involves a complex combinatorial problem constrained by coverage, fuel, and safety considerations. This paper models the viewpoint selection as a Constrained Markov Decision Process (CMDP) and applies Proximal Policy Optimization with Lagrangian relaxation to solve it. The system combines a high-level planner, which sequences camera viewpoints on an observation sphere around the target, with a low-level model that assesses each transfer using the Clohessy–Wiltshire (CW) equations for relative motion. The deterministic constraints on single-impulse reachability and $Δv$ cap are enforced directly on the action set. Safety margins are affected by navigation and manoeuvre errors, making the actual clearance stochastic. To address this, a chance constraint on the realised trajectory is implemented, with a Lagrangian multiplier governed by PID control to adapt the penalty during training. The approach is tested on a Galileo-class navigation satellite in medium Earth orbit, modelled as a point cloud where visibility from each viewpoint considers field-of-view and line-of-sight occlusion. Across various risk budgets, the trained policy maintains the keep-out violation rate within the specified limit, achieving coverage close to its maximum and staying within the mission $Δv$ budget. Compared to the usual practice of applying a fixed margin to the keep-out radius and planning deterministically, the chance-constrained policy attains less than half the violation rate at comparable coverage. This makes the collision risk an explicit design parameter, eliminating the need for re-tuning per case. The learned policy selects a viewpoint without the repeated Monte Carlo risk estimates required by a greedy, chance-constrained planner at every decision point.

journal
77th International Astronautical Congress (IAC), International Astronautical Federation (IAF), Oct 2026, Antalya (TR), Turkey
fields
Optimization and Control
pdf ↗ page arXiv:2610.41639 ↗

Multi-target on-orbit servicing missions require planning tools that can adapt to evolving operational conditions while accounting for the outcome variability inherent to safety-critical spaceflight. This paper presents an adaptive planning approach for multi-target servicing campaigns in low Earth orbit and applies it to inspection missions flown by a low-thrust servicer. The approach is based on Distributional Reinforcement Learning, which learns the full distribution of mission returns and thereby supports explicit risk-sensitive decision-making at each planning step. Specifically, a Quantile-Regression Deep Q-Network (QR-DQN) agent selects which target to inspect next by jointly considering the expectation and the variance of the predicted return through a mean-variance criterion. The mission environment models circular orbits with a spread in inclination and uses Edelbaum's combined-manoeuvre low-thrust transfer estimates. The observation comprises the orbital elements, the visit history, the observed risk state of each target and the remaining mission-time budget. The risk state of each target changes stochastically during the campaign, making the outcome of an inspection uncertain and creating operationally meaningful distinctions between risk-neutral and risk-averse planning strategies. Because the mission-time budget generally admits only a subset of the candidate targets, the planner must also select which targets to inspect. Numerical experiments on scenarios with five target satellites show that one risk-aversion weight moves the policy across a range of mean return and return variance, over an unchanged environment and reward. Risk-averse agents reduce the variance of the episodic return by approximately one half relative to risk-neutral agents and lower the probability of a mission loss, at a cost in mean return. To validate the learned policies, the efficient frontier, which gives the highest mean return attainable at each level of return variance, is computed exactly by enumerating the admissible visit orders. At equal variance, the learned policies lie close to this frontier in mean return, and the risk-averse policies have a lower probability of a mission loss than the frontier policies.

journal
77th International Astronautical Congress (IAC), International Astronautical Federation (IAF), Oct 2026, Antalya (Turkey), Turkey
fields
Optimization and Control
pdf ↗ page arXiv:2610.41647 ↗

Quantum Neural Networks (QNNs) offer a framework for learning functions from classical data, but their expressive power depends critically on how inputs are encoded into quantum states. Angle encoding can require resources that scale with the number of features, whereas amplitude encoding represents $d$ features using only $\mathcal{O}(\log_2 d)$ qubits. This compactness comes at a representational cost: normalization makes conventional amplitude encoding invariant to nonzero real rescaling of the input, up to an unobservable global phase. We show that this invariance induces an exact, architecture-independent restriction on the resulting hypothesis class. No downstream quantum circuit can distinguish inputs differing by nonzero real scaling. We develop a general condition under which deterministic preprocessing can remove this ambiguity and show that nonlinearity alone is insufficient. Based on this analysis, we introduce a fixed nonlinear feature map that provably distinguishes all real inputs under subsequent amplitude encoding, without adding trainable parameters and while retaining logarithmic qubit scaling. We evaluate the resulting QNN on three canonical classification benchmarks and a nonlinear synthetic dataset against conventional amplitude and angle encoding, as well as a classical neural network. Across these tasks, the proposed encoding consistently improves classification accuracy over both quantum encoding baselines by 21$-$44 %. Our results establish normalization-induced scale invariance as a concrete representational bottleneck of amplitude encoding and demonstrate that it can be removed while preserving a compact quantum representation, marking a step toward quantum models capable of capturing arbitrary relationships in classical data.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41824 ↗

Recent developments in both high-thrust (HT) and low-thrust (LT) propulsion technologies broaden design options for on-orbit servicing (OOS) missions. However, incorporating LT transfers into OOS mission planning requires accounting for the trade-off between prescribed transfer duration and propellant consumption. Multi-servicing, multi-transfer OOS mission planning is inherently sequential and requires adaptive planning tools that account for propulsion-dependent transfer durations and mass-dependent propellant consumption. Thus, transfer dynamics must be integrated into mission planning and scheduling to increase flexibility and adaptability. In addition, the planning is often done with imperfect information about the system's state and stochastic service demands. This paper focuses on modeling orbital transfers, particularly LT transfers, within a Partially Observable Markov Decision Process (POMDP) framework for OOS mission planning. To enable offline policy computation, the transfer model must provide explicit state transitions and rewards for all admissible actions. The state space is therefore augmented with the servicing vehicle's remaining propellant to enable consistent mass updates. HT transfers are modeled using a precomputed transfer matrix that maps each transfer to the time of flight and propellant consumed. LT transfers are represented using an analytical GEO phasing model with prescribed transfer-duration alternatives. For each alternative, the model calculates propellant consumption and maneuver feasibility as functions of the required phase change and the current servicing-vehicle mass. The proposed modeling approach is evaluated on simplified OOS planning scenarios using circular phasing with HT and LT propulsion. The resulting POMDPs are solved using the Successive Approximations of the Reachable Space under Optimal Policies (SARSOP) algorithm, and the policies obtained under HT and LT are compared. This comparison provides a quantitative assessment of the feasibility and operational impact of LT options, particularly regarding the potential role of advanced electric propulsion in future servicing architectures.

journal
77th International Astronautical Congress (IAC), Oct 2026, Antalya (Turkey), Turkey
fields
Optimization and Control
pdf ↗ page arXiv:2610.41839 ↗

This paper develops a safety verification framework for inverter-based radial distribution networks under sensor and actuator attacks. Each customer is equipped with a droop-based controller to regulate its voltage within a desirable bound of a reference, i.e., to achieve safety. In our setting, attacks and disturbances enter the droop-based controllers, while actuator attacks corrupt the reactive power injected into the network. Consequently, invariance of the controller state alone does not guarantee compliance with inverter capacities or customer voltage limits. To address this challenge, we derive conditions to guarantee that the reactive power state remains inside a predefined ellipsoid, while at the same time ensuring that the attacked reactive power injections and customer voltages respect predetermined operating limits that hold uniformly for the entire predefined ellipsoid. Together, these conditions guarantee safety under all admissible disturbances and state-dependent sensor and actuator attacks. Finally, we express the resulting containment condition as a linear matrix inequality (LMI) feasibility problem. A five-customer radial network based on the CIGRE benchmark corroborates our theoretical developments in simulations.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41910 ↗

Multiple local updates are widely used in decentralized optimization to enhance communication efficiency, often in conjunction with stochastic gradients when exact gradient computation is costly. However, such algorithms generally converge only to a neighborhood of the optimum, whose size depends on the stepsize and the variance of gradient noise. Variance reduction has been incorporated into ADMM-based methods to achieve exact convergence in expectation, but their stepsizes typically scale inversely with both the square root of the network size and the cube of the number of local updates, leading to slow convergence as the network size or the number of local updates increases. In this paper, we propose a novel primal-dual-based decentralized optimization algorithm that integrates variance reduction with multiple local updates, with a stepsize independent of the network size and inversely proportional to the number of local updates (rather than its cube), thereby enabling significantly larger stepsizes and faster convergence. Moreover, to the best of our knowledge, we establish the first linear convergence guarantee in expectation for decentralized optimization with multiple local updates under stochastic gradients in the strongly convex setting. Beyond the theoretical guarantee, the proposed method exchanges only one variable per round, halving the communication cost compared with gradient-tracking-based methods. Numerical experiments further demonstrate its advantages in convergence speed and accuracy.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41927 ↗

This paper establishes the existence, in every dimension, of universal Minkowski classes of configuration-constrained polytopes, with direct applications to set-theoretic control. For any prescribed accuracy, such a class contains an approximation of every compact convex set to that accuracy and admits a jointly affine parameterization of facets and vertices. The construction is explicit and relies on a combinatorial averaging procedure that produces minimal configuration enclosures. In control, these polytopic templates enable convex reformulations of otherwise non-convex problems, including invariant-set computation and control Lyapunov function design. Tutorial examples illustrate both the geometric construction and its control applications.

fields
Optimization and Control
pdf ↗ page arXiv:2610.41966 ↗

We study the stochastic second-order oracle complexity of solving smooth convex finite-sum optimization with $n$ components. Very recently, Tsedenov \& Kovalev (2026) showed that the minimax optimal complexity under averaged smoothness and convexity is $\tildeΘ(n + n^{6/7} ε^{-2/7})$ by establishing nearly matching upper and lower bounds. In this paper, we show that individual smoothness and convexity provably enable further acceleration, where the lower complexity bound is $Ω(n+ n^{5/7} ε^{-2/7})$ under this stronger assumption. Then, by establishing a novel second-order self-bounding inequality, we design a new Taylor-corrected lazy Newton method, which can lead to a better complexity of $\tilde{\mathcal{O}}(n + n^{5/7} ε^{-2/7})$ by applying the optimal Monteiro-Svaiter acceleration, matching our lower bound up to only logarithmic factors.

fields
Optimization and Control
pdf ↗ page arXiv:2610.42040 ↗

This work considers the design of stepsize schedules for stochastic gradient descent in smooth convex optimization. Throughout, we assume that gradient noise is conditionally zero-mean with bounded variance. Among problems with dimension $d\geq N+1$, we provide an exact worst-case analysis of the expected suboptimality of stochastic gradient descent's final iterate. Our analysis applies to any (potentially nonconstant) schedule of $N$ stepsizes in $[0,1.5]^N$. Here, exact means that we provide a matching objective function and noise distributions where our guarantee is attained. Whenever the variance bound is at least $Ω(1/\sqrt{N})$, our theory provides the minimax optimal stepsize schedule for stochastic gradient descent and proves its uniqueness. These optimal stepsizes monotonically decrease, ruling out any accelerations in expected final suboptimality from fixed fractal schedules observed in the deterministic case.

fields
Optimization and Control
pdf ↗ page arXiv:2610.42048 ↗

Quantum algorithms for linear algebra that access a matrix $A$ through quantum random access memory (QRAM) prepare quantum states proportional to the solutions of linear systems and related problems. Dequantized classical algorithms, when given sampling and query access to $A$, return sampling and query access to an approximate solution. Both kinds of algorithms use a number of queries that is polynomial in the normalized Frobenius norm $ρ=\|A\|_F/\|A\|$ and does not directly depend on the dimension. For linear systems at fixed condition number and accuracy, quantum algorithms use $\widetilde O(ρ)$ queries, whereas the best dequantized algorithms use $\widetilde O(ρ^4)$. We prove that this quartic gap is tight in the worst case, up to a logarithmic factor. For every constant $κ_0>1$, we construct dense linear systems with condition number at most $κ_0$ for which sampling from the solution within a sufficiently small constant total variation distance requires $Ω(ρ^4/\logρ)$ classical queries but only $Θ(ρ)$ quantum queries. The same separation holds, for errors below smaller thresholds, for positive-definite systems, least squares, ridge regression, matrix powers and square roots, imaginary-time and real-time Hamiltonian evolution, and sampling from the top principal component. Our results therefore show that many quantum algorithms for linear algebra, achieve a near-quartic speedup in the worst case, over classical algorithms with sampling and query access. All of these classical lower bounds follow from the same framework, which is based on a lower bound for sampling from distributions that depend on a fixed set of Fourier coefficients of an unknown Boolean function.

fields
Optimization and Control
pdf ↗ page arXiv:2610.42069 ↗

We consider the adjoint-free random-plane iteration algorithm for finding the top eigenvector of a matrix proposed in (SIAM J. Matrix Anal. Appl. 47 (2026), 1349-1374). Unlike standard methods such as the power method or gradient descent, the above algorithm lacked linear convergence guarantees in the presence of a spectral gap. In this paper, we establish explicit spectral-gap-dependent global linear convergence rates in expectation, providing quantitative guarantees for the empirically observed convergence.

fields
Optimization and Control, Numerical Analysis, Probability
pdf ↗ page arXiv:2610.42071 ↗

Freeze-thaw Bayesian optimization (BO) evaluates candidates in resumable increments but usually stops at a budget or convergence test. We maximize a budget-free utility, the best observed score minus a cost per increment, with a one-step lookahead: advance the candidate with the largest expected improvement (EI) under a fixed surrogate, and stop once no EI exceeds the cost. This rule, EI stopping, never falls below stopping immediately in expected utility, yet it can stop too early. We construct a certificate, a lower bound on the value of two-step lookahead, that proves such a stop premature, and repeating this correction never lowers expected utility under the surrogate. Under the surrogate, EI stops are premature on average at low costs and lose little at high costs. On three hyperparameter optimization benchmarks, a pre-registered test finds EI stopping noninferior to the correction in most high-cost settings, while the correction gains at low costs on two of the three; more information sharing in the surrogate raises EI stopping's utility. On the surrogates of two published methods, EI stopping trades lower scores for fewer training increments. EI stopping is a simple cost-aware reference whose utility depends on the surrogate.

fields
Optimization and Control
pdf ↗ page arXiv:2610.42085 ↗

Distributionally robust optimization (DRO) minimizes worst-case risk under distributional uncertainty, but its decisions can be suboptimal in terms of their excess risk, or regret. We study this gap for multivariate least-squares regression, assuming that the observed source law and unknown target law differ by an adversarial 2-Wasserstein perturbation of size $\varepsilon$. We first identify the asymptotic complexity of robust decision-making in the $\varepsilon \to 0$ limit, characterized by a problem-specific condition number $κ$. Here, a wide spectrum of simple algorithms, including DRO, achieve the instance-optimal rate of $O(κ\varepsilon^2)$. For fixed $\varepsilon > 0$, we prove that DRO still achieves the minimax rate if the problem has bounded residual effective rank. On the other hand, we identify significant failure modes where DRO is provably suboptimal by dimension-dependent factors. Fortunately, we prove that two alternative procedures, minimax regret optimization (MRO) and spectral-norm regularization (SNR), attain the minimax rate. We derive an exact convex reformulation of MRO and an accelerated proximal algorithm for SNR, along with high-probability finite-sample guarantees. We support these theoretical results with numerical experiments separating the performance of MRO and SNR from DRO.

comments
Accepted at NeurIPS 2026
fields
Optimization and Control
pdf ↗ page arXiv:2610.42109 ↗

Analysis of PDEs38

Let $p\in[1,\infty)$, and let $(X,ρ,μ)$ be a complete doubling $q$-Poincaré space for some $q\in[1,p]$ with a lower dimension $d$ and an upper dimension $D$. In this article, we establish Poincaré inequalities for two families of nonlocal nonconvex functionals defined using distances or volumes. More precisely, for any ball $B$ with its radius $r_B\in(0,\operatorname{diam} X)$, any $λ\in(0,\infty)$, and any measurable function $f:X\to\mathbb R$, \begin{align*} &\frac{1}{μ(B)^2}\int_B\int_B|f(x)-f(y)|^p\,dμ(x)\,dμ(y)\\ &\quad\lesssimλ^pr_B^pφ(B)^{γp}+\frac{λ^pr_B^p}{μ(B)}\mathop{\int_{ΛB}\int_{ΛB}}\limits_{\genfrac{}{}{}{2}{|f(x)-f(y)|}{ρ(x,y)φ(x,y)^γ}>λ}\frac{φ(x,y)^{γp}}{V(x,y)}\,dμ(x)\,dμ(y), \end{align*} where $V(x,y):=μ(B(x,ρ(x,y)))$ and either $φ=ρ$ with $φ(B)=r_B$ and $γ\in(-\infty,-\frac{\min\{q,D\}}{p}]$ or $φ=V$ with $φ(B)=μ(B)$ and $γ\in(-\infty,-\frac{q}{p\max\{q,d\}}]$. Here both ranges of $γ$ are sharp over the class of $q$-Poincaré spaces with the prescribed lower and upper dimensions, and the constant $Λ\in[1,\infty)$ and the implicit positive constant depend only on $p$, $γ$, and the geometric constants of $X$. We also obtain lower bounds for Sobolev energy and total variation in terms of the limit inferior of these functionals, for $γ$ in either the above negative ranges or $(0,\infty)$. Our proofs use the curve characterization of Poincaré spaces and an adaptive dyadic refinement argument based on medians.

fields
Analysis of PDEs, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.40743 ↗

We prove the existence of an infinite family of conservation laws and, more generally, balance laws, satisfied by plane-symmetric flows of perfect compressible fluids, allowing all components of the velocity vector to be possibly non-vanishing. We cover general pressure laws satisfying solely hyperbolicity and asymptotic conditions. We exhibit an almost quadratic formulation of the Euler equations, which we refer to as the J-formulation, and introduce a reduced Euler system based on an effective mass-energy density and an effective pressure law which account for transverse velocities. We establish the existence of a fundamental kernel which we use to generate all spacetime (quasi-)currents, also called mathematical entropy pairs. We derive explicit formulas for their density and flux components, together with the source-terms of the corresponding balance laws. Furthermore, we establish regularity and asymptotic estimates covering the whole range of physical variables, including vacuum. We prove that the kernel is positive on the interior of its support provided the pressure law satisfies a genuine nonlinearity condition, and we also investigate the future-convexity of the spacetime currents. Throughout, the fluid momentum parallel to the orbits of plane-symmetry is kept arbitrary. In turn, this family of spacetime quasi-currents provides a key ingredient in our companion study of torus-symmetric Einstein-Euler flows with weak regularity.

comments
95 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.40836 ↗

We propose a phenotype-structure form of the Cahn-Hilliard equation with degenerate mobility and analyze the existence of weak solutions for this new variant. This model is motivated by tumor modeling as multiple phenotypes co-exist within the same tumor and are believed to constitute a continuum of phenotypes. The originality of the model lies in its ability to consider an infinite number of phases and it allows for spatial interactions between phenotypes as well as transport in the phenotypic space, i.e., mutations.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.40869 ↗

We prove a sharp, dimension-free quantitative stability theorem for the Gaussian isoperimetric inequality in general Gauss spaces $(\mathbb{R}^n,γ_A)$, where $A$ is an arbitrary positive-definite matrix. More precisely, we show that the anisotropic Gaussian perimeter deficit controls the square of the anisotropic Gaussian asymmetry with a dimension-free constant, recovering the sharp quadratic stability behavior known in the isotropic setting. The central difficulties are the loss of rotational symmetry and the lack of compactness of the isoperimetric sets. The main novelty of our work is a new symmetrization-based reduction to the small-perturbation regime, together with effective Fuglede-type estimates derived from the weighted constant mean curvature equation. These ingredients overcome the principal obstructions created by anisotropy and extend the proof-from-regularity approach to quantitative isoperimetric stability in the anisotropic setting.

comments
73 pages. No AI
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.40876 ↗

In this paper, we establish global existence of strong solutions for the three-dimensional Cahn–Hilliard–Navier–Stokes system on the torus through finite-dimensional scalar modifications of a prescribed external force. For arbitrary initial data in $H^1\times H^2$ and any force that is locally bounded in time, we construct two smooth scalar corrections, one added to the third component of the fluid force and the other to the phase source. These corrections involve three fixed Fourier frequencies in the fluid equation and one in the phase equation, and act only on a set of times of arbitrarily small total measure. The proof combines a quantitative local theory with repeated returns of the solution to a fixed ball around zero. These returns rely on small-time asymptotic limits that balance the different scales of the two equations, together with a saturation procedure carefully adapted to the structure of the coupled system. Fluid transport generates the phase frequencies that the chosen phase forcing cannot produce on its own. As a consequence, we also obtain global strong solutions for the three-dimensional Navier–Stokes equations with a single smooth scalar correction acting on one component of the force at three fixed Fourier frequencies.

comments
Comments are welcome!
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.40898 ↗

We study the eigenvalues and eigenvectors of the Schur complement $\textbf{S}$ of the graph Laplacian, defined on vertices sampled from a domain with $C^2$ boundary. We use a thin boundary layer to approximate the continuum boundary as the number of vertices increases and rely on transportation theory to justify the consistency of a discrete Steklov problem. In particular, $\textbf{S}$ acts on vectors defined over the shrinking boundary layer but behaves asymptotically like a known operator. In this paper, we use a variational argument to determine conditions for which the eigenvalues and eigenvectors of $\textbf{S}$ converge to the eigenvalues and eigenfunctions of a Dirichlet-to-Neumann map, and we provide evidence of this spectral convergence through a numerical example.

comments
25 pages, 3 figures; the author welcomes all comments
fields
Analysis of PDEs, Spectral Theory
pdf ↗ page arXiv:2610.40920 ↗

We introduce a unified optimal transport framework for the Immersed Boundary (IB) electrodiffusion active pump model. By representing membrane chemical potential barriers and active pump potentials via smoothed Eulerian kernels ($Ψ_w, \mathcal{H}_w$), we formulate the passive dynamics as an $L^2$-Wasserstein ($W_2$) gradient flow coupled to non-conservative active pushing fields representing ATP-driven $3:2$ $\text{Na}^+/\text{K}^+$ exchangers. We establish global weak existence of solutions to the continuous Poisson–Nernst–Planck–IB system via a split-step Jordan–Kinderlehrer–Otto (JKO) variational scheme, backed by Calderón–Zygmund bounds for non-local Poisson drift and $H^3(\mathbb{S}^1)$ interface regularity under Stokes flow. We prove the continuous Rayleighian energy balance and establish exponential $W_2$-stability of Non-Equilibrium Steady States (NESS) under a quantitative Bakry–Émery curvature threshold. Furthermore, we analyze the dual singular limits of kernel smoothing width $w \to 0$ and dimensionless Debye length $g = λ_D / L \to 0$. We establish the precise asymptotic regime connecting regularized and sharp models, prove the $Γ$-convergence of regularized energy functionals, derive distributional interface flux jump conditions, and resolve $O(λ_D)$ Debye space-charge boundary layers to confirm bulk electroneutrality.

comments
29 pages
fields
Analysis of PDEs, Numerical Analysis, Optimization and Control
pdf ↗ page arXiv:2610.40942 ↗

We study the linearized Yamabe equation in $\R^n$, or equivalently on $\s^n$, at a low-energy sign-changing solution with two nodal domains. We consider odd dimensions $n=2k-1$ and a solution invariant under $Γ=O(k)\times O(k)$. Its regular level sets are isoparametric hypersurfaces, which foliate the sphere away from the two focal submanifolds. This symmetry reduces the equation for the solution to a singular ODE. In the balanced case, we choose a single-node profile that is odd under reflection across the midpoint. We prove nondegeneracy in the $Γ$-invariant subspace for every $k\ge2$, and in the full space $H^1(\s^n)$ for sufficiently large $k$: every kernel element is generated by a conformal invariance of the equation. The proof uses separation of variables, Wronskian identities and comparison functions adapted to the singular equations for the angular modes.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.40960 ↗

In this work we study two optimal design problems, of hard and soft obstacle type, in fractional Orlicz-Sobolev spaces. We also establish the connection between these problems, and finally we analyze the asymptotic behavior of the quotients $Λ_{s,μ}(α)/μ$ with respect to the energy level $μ$ at the endpoints $0$ and $+\infty$.

comments
29 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.40997 ↗

We prove the weak maximum principle and the Cucker-Smale flocking theorem for a general class of kinetic alignment models with rough communication kernels. This includes the classical Cucker-Smale and Motsch-Tadmor systems with defining kernel $φ$ assumed to be merely continuous or even $φ\in L^\infty$ for solutions in the Lebesgue class $f\in L^1$. For such models the characteristic flow-map is not well-defined and therefore the classical approaches of Cucker-Smale \cite{CS2007a,CS2007b} or Ha-Liu-Tadmor \cite{HL2009,HT2008} break down. We develop a new method adopted to rough settings based on Ambrossio's probabilistic representation of solutions and construction of dynamic barriers. All proofs are performed in the framework of so called bounded communication protocols, a non-smooth variant of regular protocols introduced in \cite{Shv-CPAM, Shv-EA} and representing a broad range of models of collective dynamics.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41047 ↗

We study the time-asymptotic stability of generic Riemann solutions for the one-dimensional radiative Euler equations, where the Riemann pattern consists of a $1$-rarefaction wave, a $2$-contact discontinuity, and a $3$-shock wave. We prove that, for sufficiently small wave strengths and initial perturbations, the Cauchy problem admits a unique global solution, which converges uniformly as $t\to \infty$ to a composite wave of the $1$-rarefaction wave, $2$-viscous contact wave, and $3$-radiative shock profile with a time-dependent shift. As a key ingredient, we establish a comprehensive set of quantitative estimates for weak radiative shock profiles. Our proof is based on the method of $a$-contraction with shifts. In all previous applications of the $a$-contraction method to the study of long-time behavior, the proof has relied essentially on the parabolic structure provided by either viscosity or heat conduction. In this work, we extend the method to a hyperbolic-elliptic system without parabolic structure.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41106 ↗

We study stable solutions of the discrete Liouville equation \[ -Δu=e^u \] on infinite, locally finite graphs with polynomial volume growth. We prove that no stable solution exists whenever the order $d$ of polynomial volume growth satisfies $d<10$.

comments
7 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41137 ↗

We study optimal potentials for the finite-rank fractional Lieb–Thirring problem at the eigenvalue-counting endpoint, for \(0<s<1\) and \(d>2s\). Using the zero-energy Birman–Schwinger formulation, we prove a profile decomposition of normalized optimizing sequences with exact splitting of rank and potential mass, and strong approximation in \(L^{d/(2s)}\). Strict subadditivity yields existence and compactness modulo translations and dilations. At rank two, the optimal constant equals the rank-one constant, the supremum is not attained, and every normalized optimizing sequence splits into two Sobolev potentials. We also prove strict increase of the minimal potential masses and a uniform spectral gap for optimizing sequences. For each optimizer, we derive an Euler–Lagrange system and establish boundedness, local Hölder regularity and \(V(x)=|x|^{-4s}(c_\infty+o(1))\) as \(|x|\to\infty\), with \(c_\infty\geq0\). The conformal orbit of every optimizer contains a representative with a strictly positive asymptotic coefficient.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41148 ↗

We compare the maximum points \(P_Ω\) and \(Q_Ω\) of the torsion function and the positive first Dirichlet eigenfunction. We begin with a systematic analytic description of both hot spots over the full range of opening angles of circular sectors: uniqueness, strict monotonicity with respect to the angle, distinct small-angle asymptotics, and explicit limiting equations in the slit-disk limit. We also prove the existence of a coincidence angle and establish arithmetic properties of the limiting coordinates. We then consider bounded convex planar domains with a line of symmetry, and we prove the sharp diameter-normalized identity \[ \sup_Ω \frac{|P_Ω-Q_Ω|}{\operatorname{diam}(Ω)}=\frac12. \] Its upper-bound argument uses standard localization in the heart and an elementary geometric enclosure. Sharpness follows from thin convex domains with a vanishing relative width perturbation peaking near one endpoint: the torsion hot spot approaches that endpoint, whereas the ground-state hot spot approaches the midpoint.

comments
33 pages, 2 figures
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41205 ↗

We establish improved local smoothing estimates for the Schrödinger equation on waveguides. Our results extend the previously known range of exponents in every dimension and are sharp in certain cases. The proof develops a transference argument that preserves the discrete periodic frequency structure, together with a refined decoupling estimate adapted to waveguides. We also derive consequences for Euclidean local smoothing and almost everywhere convergence.

comments
51 Pages
fields
Analysis of PDEs, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.41219 ↗

We study an unbalanced optimal transport problem with a spatially independent signed source. The target density is continuous on a bounded domain or continuous and integrable on $\mathbb R^n$. We establish sequential weak lower semicontinuity at the critical Sobolev regularity $W^{1,n}$ and existence of minimizers under prescribed distortion and Jacobian mass bounds, provided a finite-energy admissible pair exists. For domain-preserving, uniformly bi-Lipschitz affine flows, we prove an energy-preserving static-dynamic correspondence and stability of the reconstruction. Within this class, sequential $Γ$-convergence identifies the strong source-penalty limit as transport minimization among pairs of minimum source cost. This yields optimal-cost asymptotics and subsequential convergence of minimizers and their reconstructed paths. When a constant source is admissible, optimal sources converge strongly in $L^2$ to the constant determined by the endpoint mass difference. We establish local constant-source feasibility for Hölder endpoint densities near a fixed reference pair and give a quantitative obstruction showing that constant-source feasibility may fail even when admissible pairs exist.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41321 ↗

For first-order port-Hamiltonian systems in the dissipative setting considered here, we characterise exponential stability without any further regularity or block-structure assumption on the Hamiltonian density, thus closing a gap persisting in the literature. The proof requires revisiting the computation of the adjoint generator of the port-Hamiltonian semigroup. Finally, we provide both examples and non-examples showcasing the inconclusiveness of known conditions on the Hamiltonian densities necessary for the application of characterisation results in the literature.

comments
21 pages
fields
Analysis of PDEs, Mathematical Physics, Functional Analysis
pdf ↗ page arXiv:2610.41326 ↗

We establish local boundedness and integral Harnack-type inequalities for locally nonnegative weak solutions to nonlocal doubly nonlinear parabolic equations in the fast diffusion range by combining refined energy estimates with a careful treatment of the nonlocal terms. As an application, quantitative decay estimates for both the local mass and the local supremum of weak solutions are derived near the extinction time. Furthermore, we study the extinction phenomenon for bounded nonnegative weak solutions to the Cauchy–Dirichlet problem and obtain quantitative upper bounds for the extinction time.

comments
39 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41329 ↗

For every integer $d\ge2$ and every subsonic speed $c\in(0,1)$, we construct a nonconstant finite-energy travelling wave for the easy-plane Landau-Lifshitz equation on $\mathbb{R}^d$. This addresses the existence problem raised in the numerical work of Papanicolaou and Spathis. The profiles satisfy the travelling-wave equation in the sense of distributions and are smooth in dimension two.

comments
82 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41470 ↗

We study the dependence on the dimension of stability constants for the Caffarelli–Kohn–Nirenberg inequalities, using the weighted energy distance to the manifold of optimizers. In the subcritical non-degenerate case, i.e., $b_{FS}(a)<b<a+1$ with $a<0$ and $a<b<a+1$ with $a\ge0$, we prove \[ \frac{β_{a,b}}d\le c_{d,a,b} \le \frac{4(1+a-b)}{d+4(1+a-b)} \] with a positive constant depending only on the weights, uniformly over all admissible dimensions. The upper bound shows that this dimensional order is optimal. The critical case $a=b>0$ exhibits a different dimensional scale, \[ \frac{\min\{β_{\mathrm{Sob}},4a\}}{25d^2} \le c_{d,a,a}\le\frac{16a}{d^2}+O_a(d^{-3}), \] where $β_{\mathrm{Sob}}>0$ is a universal Sobolev stability constant. In the degenerate case on the Felli–Schneider curve, for fixed $a<0$, we prove \[ \frac{β_a^{FS}}{d^3} \le\widetilde c_{d,a,b_{FS}(a)} \le\frac{32|a|}{d^3}+O_a(d^{-4}), \] where $\widetilde c$ is the quartic stability constant given in [Frank-Peteranderl, CV\&PDE, 2024] and $β_a^{FS}>0$ is independent of $d$. The three cases thus have optimal dimensional orders $d^{-1}$, $d^{-2}$, and $d^{-3}$, respectively.

comments
46 pages. All comments welcome!
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41474 ↗

The Nonlinear Integrate-and-Fire equation is a fundamental model for neural assemblies, describing the dynamics of the density of neurons with a given potential. Its standard formulation uses a given firing potential, and is known to undergo finite-time blow-up for excitatory connections. To fit experimental observations, it has been advocated to use superlinear drifts. We show how, in this situation, it makes sense to set the equation on the whole line. The main question we address is whether in these circumstances the solution still exhibits finite-time blow-up. We study the construction of solutions by a fixed point and, using the framework of \cite{carrilloQualitativePropertiesSolutions2015}, we develop novel a priori estimates which use specifically the superlinearity of the drift at $+\infty$. In particular, we obtain Ces{à}ro-mean bounds on the activity, as well as exponential convergence towards the unique steady state in the weakly connected regime. Several critical steps rely on controls depending solely on the $L^1$ norm of the solution or coefficients.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41497 ↗

For any $0\leqγ_m<\frac13$ and $0\leqγ_ρ<\frac23$,, we use convex integration to construct a nontrivial weak solution $(ρ,m)$ to the three-dimensional isentropic compressible Euler equations such that $ρ-ρ^*\in C^{γ_ρ}(\mathbb{R}\times\mathbb{T}^3)$ and $m\in C^{γ_m}(\mathbb{R}\times\mathbb{T}^3)$. Here $ρ^*>0$ is a constant, and the density stays bounded away from vacuum. Both $ρ-ρ^*$ and $m$ have compact support in time. The construction applies to smooth pressure laws with positive derivative throughout the density range. This yields nonuniqueness for the constant initial data $(ρ^*,0)$. Each stage of the convex integration scheme consists of finitely many Newton steps followed by a Nash step. Kirchhoff's formula in Lagrangian coordinates and the Piola transform yield coupled density and momentum corrections that treat the acoustic part of the linearized system while preserving the continuity equation exactly. In the Newton step, we combine these corrections with an incompressible correction adapted from Giri–Radu (2024) to reduce the error. In the Nash step, a divergence-free perturbation built from spatially separated Mikado flows treats the gluing stresses on disjoint time intervals without changing the density.

comments
164 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41499 ↗

Chemotactic aggregation can overcome diffusion and lead to finite-time blow-up in Patlak-Keller-Segel models. Mixing induced by suitable fluid flows can counteract this aggregation through enhanced dissipation. Earlier results of Zeng-Zhang-Zi (J. Funct. Anal., 2021) for the coupled system impose additional smallness assumptions on the initial chemical gradient. A natural question is whether this restriction can be removed for the coupled system near a stationary Couette flow. We give a positive answer to this question for the two-dimensional parabolic-parabolic Patlak-Keller-Segel-Navier-Stokes system near the Couette flow $(Ay,0)$ in $\mathbb T\times\mathbb R$. It is proved that, when the Couette flow is strong enough ($A$ is large enough), the system admits a global solution that remains uniformly bounded in time for arbitrary fixed initial data. The main difficulties are the transient growth of the chemical gradient and the nonlinear coupling among the density, chemical field and fluid. We develop a two-stage energy method combining short-time estimates in Kelvin coordinates with long-time estimates in physical coordinates. The key is to split the nonzero chemical modes into two parts while accounting for the nonlinear transport. One part carries the initial data, and the other is a forced component with zero initial data. A weighted coupled energy controls the former through a time-integrable Gronwall coefficient and absorbs the feedback from the latter without increasing the initial energy. Enhanced dissipation then provides the bounds needed to enter the long-time regime, while one-dimensional zero-mode estimates and Moser iteration complete the argument without a mass restriction.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41507 ↗

We apply to the generalized pure power Benjamin, Bona and Mahony equation our method of combining high energy virial estimates with Kato smoothing, which closes the virial estimates in a natural way and allows us to prove a result of asymptotic stability for all the stable solitons with $c>1$ for all the exponents $p>1$.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41549 ↗

Flocking refers to a phenomenon in collective motion where particles align their velocities with each other, commonly modelled using Cucker-Smale dynamics. In this work, we introduce and analyse two general reduced SPDE models for stochastic Cucker-Smale particle dynamics that encompass environmental and i.i.d.-type noises. Our models are computationally cheaper alternatives to the underlying particle system and preserve the flocking behaviour. To derive our reduced models, we consider suitable empirical particle densities which depend on position and time, but feature velocity dependency in a multiplicative fashion: this setup crucially allows for a very intuitive and convenient notion of flocking at the SPDE level. In $d=1$ we prove flocking of the reduced SPDE models, and, for the environmental noise, we also quantify the discrepancy between the reduced SPDE and mollified particle system. We also discuss the technical modifications required to extend the results to higher dimensions. We then provide extensive numerical experiments to verify our theoretical results in $d=1$ and to explore the behaviour of our models in $d=2$.

comments
57 pages, 8 figures
fields
Analysis of PDEs, Numerical Analysis, Probability
pdf ↗ page arXiv:2610.41595 ↗

We design a boundary feedback for the level of the sediment bed of a continuous settler and prove that it exponentially stabilizes the linearized closed loop under a small-gain condition. The unit is modeled by a 2 x 2 system of balance laws in which the bed surface, the interface, is not a smooth part of the solution but a moving shock between a weakly hyperbolic free-settling zone and a strictly hyperbolic compression zone. We describe it by the Rankine-Hugoniot relations and map the problem onto a fixed domain. Moreover, we prove local well-posedness with a continuation criterion given by a single margin function. Then we linearize the system near an equilibrium. The linearization couples a dissipative distributed layer to one scalar, the interface position, whose equation does not contain the control. We solve this by infinitedimensional forwarding: a Sylvester kernel cancels the interface trace and isolates the interface pole, and a proportional feedback then sets its decay rate. Every hypothesis is verified on an illustrative example, a laboratory-scale continuous settler containing a bed of compact polystyrene beads.

fields
Analysis of PDEs, Optimization and Control
pdf ↗ page arXiv:2610.41632 ↗

The global well-posedness for Lions' weak solutions to the inhomogeneous incompressible Navier-Stokes equations in two-dimensional bounded domains is concerned. When the initial density $0\leq ρ_0\in L^\infty(\Om)$ (allowing for vacuum) and initial velocity $u_0\in L^2_σ(Ω)$, the global existence, regularity, and exponential decay-in-time rates of weak solutions are established via a priori estimates, which rely crucially on the duality of BMO and Hardy space for the pressure term, and the classical Sobolev extension theorem. In particular, for the non-vacuous case where $0< ρ_0\in L^\infty(\Om)$ is bounded away from zero, we further prove the uniqueness of weak solutions in the sense of approximation by strong solutions, based on the key estimate \(t^{1/2}\nabla u\in L^2(0,T;L^\infty(Ω))\), which essentially follows from a delicate localized point-wise bound for \(\nabla u\) near the boundary. The latter bound is derived by the Green functions estimate for the nonhomogeneous Dirichlet Stokes problem, and the uniform \(C^1\) boundary estimate on the remaining homogeneous Stokes problem, through a Stokes decomposition on suitably constructed local \(C^2\) domains.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41637 ↗

We construct generalized eigenfunctions and a spectral calculus for the transmission problem on Lipschitz domains. Then, focusing on the specific example of the ball, we prove a high-frequency resonance dichotomy for Maxwell's equations for a ball with positive, piecewise constant permittivity and permeability. If the product of these coefficients is no larger inside the ball than outside, the cutoff resolvent extends holomorphically to a strip of fixed width below the real axis, including the case of equal products. If the interior product is larger, we construct whispering-gallery resonances exponentially close to the real axis. The proofs use estimates for the transverse electric and transverse magnetic determinants that are uniform in angular momentum and complex frequency, together with a quasimode construction and Rouché's theorem. Explicit generalized eigenfunctions identify these determinants with the scattering poles and give the spectral resolution and the normalization relating the scattering amplitude to the Mie T-matrix.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41638 ↗

In this paper, we study the inviscid limit for the hyperbolic Keller–Segel equations with logistic sensitivity $1-u$. We establish a common existence time and a priori estimates uniform with respect to the diffusion coefficient in the critical Besov spaces $B_{p,1}^{1+d/p}(R^d)$ with $1\le p\le\infty$. For $1<p<\infty$, we obtain explicit convergence rates in $B_{p,1}^{0}$, and we also give estimates for the case $p=1$. For initial data in $B_{\infty,1}^{1}$, we prove convergence at rate $O(\varepsilon^{1/2})$ in $L^\infty$, a bound of order $O(\varepsilon^{1/2}\log(e/\varepsilon))$ in $B_{\infty,1}^{0}$, and strong convergence in $L^\infty(0,T;B_{\infty,1}^{1})$. Additional regularity of the initial data yields improved convergence rates. We construct a fixed initial datum showing that the exponent $1/2$ in $L^\infty$ is optimal.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41670 ↗

We consider the focussing cubic Klein-Gordon equation on $\mathbb{R}^{3+1}$ in the radial setting. Foundational work by Nakanishi-Schlag gives a complete qualitative characterisation of the dynamics near the ground state $Q$, in particular revealing a threshold phenomenon, whereby a co-dimension one Lipschitz manifold separates finite time blow up from scattering solutions in forward time. Our aim in this paper is to derive an effective bound for the scattering norm $\|u\|_{L_t^3([0,\infty);L_x^6(\mathbb{R}^3))}$, for scattering energy class solutions $u$ with data sufficiently near $Q$ and close to the threshold manifold. This requires the development of an induction on energy strategy in a setting where an alternative to the usual Morawetz approach appears necessary.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41718 ↗

In this paper, we establish Morse theory for the Neumann mean field equation on a compact Riemann surface at $ρ=4πm$, $m\in\mathbb N$. To address the loss of compactness associated with bubbling, we use Bahri's theory of critical points at infinity. Assume that the reduced functionals are Morse and that the leading scale coefficient does not vanish at their critical points. We classify the critical points at infinity arising from interior, boundary, and mixed concentration, compute their Morse indices and local relative homology, and establish deformation lemmas at resonance. We prove that, for $m\geq 2$, every sufficiently negative sublevel is homotopy equivalent to the boundary-weighted barycenter space of order $m-1$. The Euler–Poincaré identity gives the Leray–Schauder degree at resonance without assuming nondegeneracy of the solutions. If all solutions are nondegenerate, we further obtain strong Morse inequalities with explicit Betti numbers over $\mathbb Z_2$. As a consequence, we obtain existence and multiplicity results, as well as criteria for solutions with prescribed Morse index.

comments
86 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41760 ↗

We study the critical Sobolev-curl quotient on $\mathbb{R}^3$ at the conformally invariant exponent $p=3/2$. Its sharp constant and optimizers were posed as open problems in the literature. We identify the toroidal and poloidal extremals. Two structural facts are used. First, the critical curl–curl equation is self-dual: the map $u\mapsto|\nabla\times u|^{-1/2}\nabla\times u$ is an involution on finite-energy solutions. It is the self-dual case of a duality between the exponents $p$ and $3p/(4p-3)$, which exchanges toroidal and poloidal fields and relates the sharp constants. Second, the toroidal minimization problem reduces exactly to a subcritical $3/2$-Laplace problem on the hyperbolic plane. Combined with symmetry and shooting results, these facts show that the two restricted sharp constants coincide. They also classify the gauge-minimal extremals: these are unique up to scaling and the Euclidean transformations realizing hyperbolic isometries, semi-explicit, and dual to each other. For nonzero one-signed profiles radial about a fixed hyperbolic centre, a gauge-canonical inverse of the duality yields an iteration by one-dimensional quadratures: the quotients are nonincreasing and converge to the restricted sharp constant, while the normalized iterates converge to the extremal pair. The first step from an explicit competitor gives the new upper bound $96\sqrt{15}π/(5\sqrt{4455π-13568})\approx11.2948$ for the restricted constants. The unrestricted sharp constant remains open. As an application, bounded axisymmetric poloidal potentials admit no $L^2$ zero mode of the Weyl-Dirac operator. An explicit conformal transformation extends this exclusion to both extremals at every real coupling.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41804 ↗

Fractional differential equations are increasingly used to model memory and anomalous transport in chemistry and related fields. However, the formal replacement of an integer-order derivative by a fractional one does not by itself justify fractional dynamics. The essential issue is whether nonlocality can be derived from a physical, stochastic, geometric, or reduction mechanism and whether it leads to experimentally testable consequences. This review, written primarily for mathematicians and specialists in fractional analysis, discusses representative applications in electrochemistry, adsorption and sorption, reaction–subdiffusion, chemical and enzyme kinetics, polymer relaxation, and atmospheric chemistry. Particular attention is paid to distinguishing mechanism-based fractional models from phenomenological fractionalizations. Mathematical questions of well-posedness, nonnegativity, conservation laws, dimensional consistency, parameter identifiability, and numerical convergence are emphasized. Experimental electrochemical examples illustrate the need to distinguish genuine fractional effects from model mismatch or non-identifiability. Several open problems connecting fractional analysis with experimentally grounded chemical modeling are formulated.

comments
31 pages, 4 figures
fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2610.41817 ↗

We consider the variational problem of minimizing the multiple integral $\int_Ωf(\nabla u) \, dx$ with a convex density $f$ being of linear growth among functions $u$: $Ω\to \mathbb{R}$ satisfying the Dirichlet boundary condition $u = u_0$ on $\partialΩ$ and for which $u \ge Ψ$ holds throughout the domain $Ω$. We prove that the dual problem has a unique solution. This result is extended to the case of interior obstacle problems, where the validity of $u \ge Ψ$ is only required on a subset $E$ of $Ω$. Moreover, we discuss the vectorial setting, i.e. the case of functions $u$: $Ω\to \mathbb{R}^N$, $N > 1$, combined with the additional condition $u(Ω) \subset A$ for a closed convex set $A \subset \mathbb{R}^N$, again showing the existence and the uniqueness of the associated dual solution.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41853 ↗

This paper is concerned with an inverse source problem for the stochastic fractional Helmholtz equation in a three-dimensional inhomogeneous medium. The source is modeled as real Gaussian spatial white noise with an unknown nonnegative strength function. For admissible frequencies, we establish well-posedness of the radiating solution in the distributional sense and show that it has a smooth representation outside the source region. For the inverse problem, by using two-point correlations of the field measured in an exterior open set at a single frequency, we prove a logarithmic stability estimate for recovering the source strength under an a priori Sobolev bound. As a uniform Hölder stability estimate is unavailable for the general admissible class, by assuming the difference between two source strengths has a fixed sign, we improve the stability to Hölder continuity with an optimal Hölder exponent. The analysis combines scattering theory and quantitative continuation for the fractional Laplacian using the Caffarelli-Silvestre extension.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41928 ↗

We prove conjectured-optimal large-time upper decay estimates for solutions of weakly coupled systems of $m\ge2$ time-fractional diffusion equations. Our result establishes the upper-bound assertion in the conjecture of Li, Liu, and Wada~\cite{LLW26}. The sharpness of these upper exponents for $m\ge3$ remains open in general; see Section~\ref{sec:discussion}. Assume that the diffusion matrices are uniformly elliptic, the coupling coefficients are bounded and may depend on time, and the matrix of $L^\infty$-norms of the coupling coefficients satisfies a spectral-radius stability condition. Let $α^*$ denote the smallest fractional order among the components with nonzero initial data. Then the $L^2(Ω)^m$-norm of the solution is $O(t^{-α^*})$ as $t\to\infty$ if $α^*<1$, whereas it is $O(t^{-(1+α_m)})$ if $α^*=1$, where $α_m$ is the smallest fractional order occurring in the system. For $m=2$, the result yields the same upper decay exponents as \cite[Theorem~1.1]{LLW26}; in the absence of self-coupling, the matrix stability condition is weaker than the corresponding small-coupling condition in that work. The case $m\ge3$ is new. Our proof uses a large-time weighted Volterra framework. If $α^*<1$, the weight is a regularly varying power weight; if $α^*=1$, it is constructed by convolution with the resolvent kernel associated with the smallest fractional order. These weights yield the required large-time bounds for all resolvent kernels in the system. In particular, the argument avoids the Laplace-transform and residue-calculus methods used in \cite{LLW26}, which become difficult to implement for systems with three or more components.

comments
22 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.41993 ↗

We consider the Lane-Emden equation associated with the operator $-Δ- γ\, (-Δ)^s$ (for positive $γ$). First we consider the equation in the entire space and we prove that no Serrin-type critical exponent arises in the $H^1(\mathbb R^n)$ framework. Secondly we consider the problem settled in an open, bounded and connected set and we prove the existence of a positive solution for sufficiently small $s$.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.42070 ↗

We establish decay estimates toward rarefaction waves for the one-dimensional fractional Burgers equation when the order of dissipation lies between one and two. The initial data are bounded, and their perturbation from the Riemann step belongs to the critical weak-Lorentz space determined by scaling; the perturbation need not be small or integrable. We obtain optimal algebraic decay exponents, improving the known decay rates, and extending convergence to every Lebesgue space above the critical exponent, including the supremum norm. We prove optimality by constructing a single solution with Riemann step initial data whose lower decay estimates have exactly the same algebraic exponents as our upper estimates. These lower bounds establish that neither the algebraic decay exponents nor the range of convergence can be improved under the stated assumptions. The upper estimates contain a logarithmic factor whose necessity remains open. The proof combines convex truncation estimates, fractional interpolation, and a one-sided slope estimate for the supremum norm.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.42125 ↗

Probability25

Many emerging communication systems increasingly involve devices that transmit packets in multiple causally dependent stages as part of an internal workflow, rather than waiting for receiver feedback. To analyze such systems, we recently introduced the $\mathrm{M^T/G/1}$ queue, in which primary customers arrive according to an exogenous Poisson process, and each triggers a secondary customer to join the queue after an independent random delay, called the \textit{triggering delay}. Our prior heuristic approximations assume identical exponential service times, and our matrix-analytic framework assumes exponential triggering delays. However, no exact analysis is available when both distributions are general. In this paper, we derive stochastic bounds, in the usual stochastic order, on the system-time distribution of each customer class under general triggering-delay and service-time distributions. The service times within each primary–secondary pair may be generally dependent. Using sample-path coupling and level-crossing arguments, we show that the stationary workload is stochastically sandwiched between those of two classical queues: a two-class $\mathrm{M/G/1}$ queue with independent Poisson arrivals (lower bound), and a batch-arrival $\mathrm{M^{[2]}/G/1}$ queue where each primary–secondary pair arrives simultaneously (upper bound). Notably, these bounds do not depend on the triggering-delay distribution, and they are asymptotically tight in the sense that the stationary workload converges in distribution to that of the lower- and upper-bound queues as the triggering delays diverge to infinity and converge to zero in probability, respectively. We translate these workload bounds into system-time bounds for each customer class. Simulation experiments support these bounds.

fields
Probability
pdf ↗ page arXiv:2610.40865 ↗

Summary statistics that measure the shapes of phylogenetic trees contributed to quantitatively understanding evolutionary processes. Among phylogenetic summary statistics, the tree balance indices have been designed to capture intuitive notions of balance or imbalance in trees. Here, we characterize the tree balance indices in terms of basis statistics, mapping the exponentially large space of rooted, binary, unlabeled trees to lower-dimensional spaces. We observe that many tree balance indices can be represented in terms of linear functions of these low-dimensional spaces, enabling us to compute exact formulas for the means, variances, and covariances of many balance indices under the Yule–Harding model. The new framework recapitulates existing formulas for the expectations of 8 indices: Sackin index, average leaf depth, total cophenetic index, $\hat{s}$-shape statistic, Colless index, corrected Colless index, quadratic Colless index, and rooted quartet index for binary trees. It produces new formulas for the expectations of 9 additional indices: equal weights Colless index, total $I$ index, total $I'$ index, Rogers $J$ index, total internal path length, total path length, average vertex depth, stairs1 index, and stairs2 index. Moreover, the framework also allows us to calculate the exact variance of all 17 indices that are represented by the bases as well as the covariance and correlation for all $\binom{17}{2} = 136$ pairs among the 17 indices.

comments
49 pages, 8 figures
fields
Probability
pdf ↗ page arXiv:2610.40894 ↗

We revisit martingale representation for counting processes using the Poisson embedding, deriving the representation in their natural filtration from that of the underlying Poisson random measure. Our approach combines Poisson Malliavin calculus with the Clark-Ocone and Mecke formulas. We first characterize measurability with respect to the counting-process filtration through the structure of the Malliavin derivative. We show that any random variable measurable in the counting-process filtration is invariant with respect to the auxiliary mark on the acceptance region obtained from the Poisson embedding method and vanishes on the rejection region. We then prove that every square-integrable martingale in this smaller filtration consequently admits a unique stochastic integral representation with respect to the compensated counting process: the integrand is explicitly characterized as the conditional expectation of the Malliavin derivative, which identifies as the conditional effect of an additional accepted event on the terminal functional. Additionally, we define the mirroring counting process which counts the rejected events in the Poisson embedding procedure and we then provide an orthogonal decomposition of any martingale in the joint filtration of the accepted and rejected counting processes as a sum of a martingale in the counting filtration obtained by Poisson embedding and its mirroring martingale part on the rejected domain. Finally, we apply these results to general Hawkes processes constructed by Poisson embedding, obtaining a martingale representation and an explicit integrand formula in their natural filtration.

fields
Probability
pdf ↗ page arXiv:2610.40980 ↗

We revisit the construction of strong solutions to singular SDEs by compactness on Wiener space. For spatially varying diffusion coefficients, assuming sufficiently small form bounds on the drift and diffusion gradient, we use basic energy estimates and an extended compactness criterion for Wiener functionals to construct an approximation-independent stochastic flow of homeomorphisms. Both forward and inverse maps have local Sobolev moment bounds, and the resulting strong solutions are pathwise unique in a class of solutions satisfying a Krylov estimate. We also show that the endpoint critical condition $b\in L^\infty_tL^d_x$, together with $\mathrm{div } b=0$ does not guarantee the existence of a strong solution. For every $d\geq2$, we construct a SDE with a divergence-free drift in this class and an additive Brownian noise, started from the origin, has a weak solution unique in law, but no strong solution. The proof uses logarithmic rescaling to characterize strong solvability by whether the sample measures of an associated stationary diffusion are Dirac masses. Positivity of the top Lyapunov exponent, combined with random SRB theory, rules out this concentration and hence strong solvability.

comments
51 pages, 2 figures
fields
Probability
pdf ↗ page arXiv:2610.41128 ↗

We design and analyze an efficient importance sampling estimator of the probability that, over a long time horizon, a Brownian particle avoids obstacles of a fixed radius placed according to an independent Poisson random field. Donsker and Varadhan identify the exponential decay rate of this probability as the optimal balance between the volume cleared of obstacles and the spectral cost of confinement, and the confinement results of Sznitman and Povel show that the particle survives by localizing its path in a ball of (roughly) deterministic radius but with a random center. This strategy makes direct Monte Carlo exponentially ineffective over long horizons, and makes a practical yet asymptotically optimal sampler non-trivial to design, since the sampler must track the random region swept out by the path. For every fixed dimension \(d\ge2\) and every fixed obstacle radius, we construct a sequential importance sampler that is unbiased and asymptotically optimal. The full algorithm has finite expected total work at every fixed horizon, and both its expected work and the coefficient of variation grow at most subexponentially with the time horizon. The method avoids repeatedly solving eigenvalue problems on the evolving explored regions and suits parallel implementation. Numerical experiments illustrate the efficiency of the method.

comments
46 pages, 1 figure, 1 table
fields
Probability
pdf ↗ page arXiv:2610.41234 ↗

We study a Markovian model for a blockchain mining process. In the limit where there is an infinite number of miners, the model satisfies a functional law of large numbers due to Barbour and Luczac (2008), leading to an infinite set of coupled ODEs characterising the blockchain heights. We show that the sequence of solutions to this system converges uniformly, to a limiting profile in an appropriate comoving frame. Utilising asymptotic arguments inspired by Brunet and Derrida (1997), we derive an explicit expression for the position of the travelling front, which exhibits a leading order logarithmic Bramson correction in time. For a finite number of miners, we further derive a correction to the travelling speed which is logarithmic in the number of miners. Finally, by application of a computation due to Ablowitz and Zeppetella (1979), we identify that, for a specific choice of the travelling speed, the Painlevé integrable, travelling wave solution of the Fisher-KPP equation with drift accurately matches our limiting profile. Our results are supported by numerical simulations.

comments
36 pages
fields
Probability, Mathematical Physics, Dynamical Systems
pdf ↗ page arXiv:2610.41312 ↗

We study the random walk on a Cayley graph of a finite group whose generators are chosen independently and uniformly at random. We show that, for every finite group, the walk exhibits cutoff with high probability whenever the number of generators is large compared with the logarithm of the order of the group but small compared with every positive power of it. The cutoff time is the time at which the expected entropy of the walk comes close to its maximum, and in general it depends on the structure of the group. This settles a conjecture of Aldous and Diaconis.

comments
20 pages
fields
Probability, Group Theory
pdf ↗ page arXiv:2610.41347 ↗

We study cyclic catalytic stochastic reaction-diffusion systems with mutation and immigration. The equations have square-root catalytic coefficients and spatially colored noise. Under a condition requiring the cyclic inward drift to dominate the corresponding noise covariance, we establish global strong well-posedness for nonnegative initial states. Our approach combines positivity-preserving approximations, uniform energy estimates, and a concave power entropy that controls the occupation of states near zero. These boundary estimates allow us to derive a Hellinger stability inequality for two synchronized weak variational solutions, yielding pathwise uniqueness directly in this solution class. We then use joint compactness of the approximating solutions and their driving noise, together with martingale identification and the Gyongy-Krylov criterion, to construct a probabilistically strong solution on the prescribed Wiener space.

fields
Probability
pdf ↗ page arXiv:2610.41400 ↗

We study sample covariance matrices formed from i.i.d., centered, isotropic random vectors $q_1, \dots, q_N \in \mathbb{R}^M$, whose coordinates are allowed to be dependent. We assume the random vectors satisfy a concentration bound for their quadratic forms $q_i^T A q_i$. Under this assumption, we establish edge universality and obtain the Tracy–Widom law for the largest eigenvalue of the sample covariance matrix after appropriate centering and scaling. In particular, our result removes the unconditionality assumption in existing edge universality results for log-concave ensembles~\cite{BX}, resolving the question that originally motivated our work. The concentration assumption also allows us to extend the result beyond the log-concave class.

fields
Probability, Mathematical Physics, Statistics Theory
pdf ↗ page arXiv:2610.41416 ↗

We consider stochastic functional differential equations with Markovian regime?switching on an (infinitely) countable state space. Via establishing the Feller prop?erty and contraction in L1-Wasserstein distance of the semigroup associated with the segment process, we show the existence and uniqueness of invariant probability measure. In order to cope with switching processes on an infinite state space, we de?velop a truncation method based on a new construction of order preserving coupling for Markov chains to generalize the estimation of exponential functionals for Markov chains with infinite states. The existing sharp estimation is only valid for Markov chains on a finite state space, as it fundamentally relies on the Perron-Frobenius theorem.

fields
Probability
pdf ↗ page arXiv:2610.41515 ↗

We study, after Coron Le Jan (2024), the genetic contribution of a single strongly advantageous mutant in a biparental Moran model under infinite selection. We prove a functional limit theorem for the entire beneficial mutation spreading period: after normalization by \(\sqrt N\), the contributions carried by the advantaged and disadvantaged subpopulations converge jointly in law to a non-degenerate random multiple of an explicit deterministic profile. Randomness is generated during the initial microscopic growth of the mutant lineage and subsequently propagated through the population along a deterministic macroscopic trajectory. At the heart of this phenomenon is an eigenmode martingale that measures the imbalance between the mean genetic weight of the founding ancestor within the two classes; its limiting value provides precisely the random amplitude of the macroscopic profile and is characterized by a smoothing equation.

fields
Probability
pdf ↗ page arXiv:2610.41650 ↗

We bridge the gap between the mixed-moment theory of càdlàg rough stochastic differential equations (short:RSDEs), introduced by Friz, Hocquet and Lê (2021) and extended to the càdlàg setting by Allan and Pieper (2026), and the theory of rough semimartingales (short:RSMs), introduced by Friz and Zorin-Kranich (2023), by investigating the intersection of these two theories for the first time. By introducing the concept of good solutions to RSDEs, solutions which are simultaneously both mixed-moment solutions and RSMs, we develop a solution concept, which allows for good control on both moment-based and pathwise properties of solutions. The well-posedness of good solutions is proven using a fixed point argument on the space of $L^{2}$-martingales by suitably leveraging previously known mixed-moment estimates. Based on these insights we further characterize a large class of good solutions of semi-linear RSDEs through a novel rough stochastic Duhamel's principle.

comments
67 pages
fields
Probability
pdf ↗ page arXiv:2610.41664 ↗

We study a discrete-time stochastic model for the spiking activity of a network of $n$ interacting neurons partitioned into $k$ communities. The interaction network is a random directed weighted graph generated by a Stochastic Block Model with $k+1$ edge types ($k$ intra-community, $1$ inter-community). Each edge is present independently with a probability that depends only on its type. When present, an edge carries a type-dependent deterministic weight scaled by $n^{-1}$, which is excitatory or inhibitory depending on an independent random sign. Given the interaction network, the spiking process is a stationary Markov chain. Our results describe how the network's community structure manifests itself in the first- and second-order statistics of this chain, as $n \to \infty$ with $k$ fixed. Representing these statistics as explicit functions of the random interaction network, we prove that, with high probability: (i) the vectors of mean firing activity and firing variances are approximated in supremum norm by deterministic vectors that are constant within each community, with an error of order $\sqrt{\log(n)/n}$; (ii) the off-diagonal part of the $0$-lagged firing covariance matrix is approximated in max norm, with an error of order $\sqrt{\log(n)/n^3}$, by a deterministic community-block matrix (constant on each block defined by a pair of communities), with block values determined by the unique solution to a $k \times k$ Stein matrix equation; (iii) in contrast, the $t$-lagged covariance matrix, for every lag $t \ge 1$, is approximated in max norm, again with an error of order $\sqrt{\log(n)/n^3}$, by a random matrix built from two components: a random matrix capturing the interaction network and a deterministic community-block matrix. Finally, we discuss conditions on the model parameters under which these limiting statistics enable the identifiability of the community structure.

fields
Probability
pdf ↗ page arXiv:2610.41706 ↗

We prove quantitative limit theorems for the edge density $Q$ in the dense edge–triangle exponential random graph model at criticality and under parameter perturbations of order $n^{-1}$. At the fixed critical parameters $(α_c,h_c),$ the centered fluctuations satisfy $n^{2/3}(Q-\mathbb{E} Q)\Rightarrow \mathcal{N}(0,σ_0^2)$, with explicit variance $σ_0^2>0$. This disproves the quartic fluctuation conjecture of Bianchi, Collet and Magnanini (2024). Gaussian limits also hold $(α_c+ λ/n,h_c + γ/n)$, except when the coefficients satisfy $-\frac{11}{16}+\frac{2}{9}(γ+\frac49λ)=0$. Under this linearity condition, the limiting laws of $\sqrt n(Q-2/3)$ have densities proportional to $\exp\{-81z^4/64+9θz^2/8\}$, where $θ=-57/16+8λ/27$. We obtain a sharp $n^{-1/3}$ rate in $1$-Wasserstein distance for the Gaussian limits and an $O(n^{-1/2})$ bound for the quartic limits. The proofs combine Stein's method for exchangeable pairs with conditional row estimates and a local Dobrushin contraction.

comments
66 pages
fields
Probability
pdf ↗ page arXiv:2610.41781 ↗

We study the integer-valued membrane model on boxes $[-L,L]^d\cap\mathbb{Z}^d$ with zero boundary conditions. We show that in dimension two the model is delocalised at sufficiently high temperature. More precisely, the variance at the origin is of order $L^2/β$, where $β$ is the inverse temperature. More generally, for every $ε>0$ and all sufficiently small $β>0$, the logarithm of each Laplace transform lies between $(1+ε)^{-1}$ and $1$ times its Gaussian counterpart. The temperature threshold is independent of $L$ and the test function. In dimensions three and four, we prove that the variance at the origin is bounded uniformly in the box size at every positive inverse temperature. Thus the integer constraint changes the order of fluctuations in these dimensions. The proofs use a modification of the Fröhlich–Spencer argument in dimension two and a comparison with integer-valued Gaussian free fields in higher dimensions.

comments
54 pages. Comments are welcome
fields
Probability
pdf ↗ page arXiv:2610.41925 ↗

We prove that the quenched pressure of the Hopfield model with $K\simαN$ i.i.d. symmetric Rademacher patterns has a thermodynamic limit at every load $α\in(0,\infty)$, finite inverse temperature and uniform external field, with no replica-symmetry, order-parameter concentration or non-condensation hypothesis required. The limit is the supremum over $r$ of limiting pressures of mixed models with $r$ binary patterns in a Gaussian bulk. Via the Hubbard–Stratonovich representation we extract the finitely many patterns whose hidden coordinates exceed $δ\sqrt N$; a collective interpolation replaces the other binary patterns by Gaussian ones with error $O(δ^2)$ uniform in the volume, and convex concentration on the discrete cube controls the disorder-dependent choice of retained patterns. For the mixed models we prove a Gaussian hidden-spin truncation bound uniform in the volume and in an arbitrary Ising potential, and convergence for Gaussian bulks with finitely many site-field types, reducing to results of Ho and of Chen, Ho and Issa on non-convex vector spin glasses with compact spins. For Gaussian patterns we give an alternative proof of the existence result announced by Camilli, Contucci and Mingione, extended to a uniform field and an extra binary pattern. In both cases the pressure has exponentially small deviation probabilities and converges a.s.; internal energy and entropy converge wherever the limit is differentiable. An extra binary Mattis pattern admits a scalar variational reduction. At zero field the Rademacher pressure dominates the Gaussian one with an extra binary pattern, which dominates the pure Gaussian one. No explicit formula is given. The proofs were generated through extensive interaction with the AI models Claude (Opus~5.5) and ChatGPT (Sol~6.1), based on the problem formulation and guidance of the authors.

fields
Probability
pdf ↗ page arXiv:2610.41938 ↗

Let $Z_N^{β_N}(x)$ denote the partition function of the two-dimensional directed polymer started at $x\in \mathbb{Z}^2$, in Gaussian environemnt, in the critical window, with parameter $β_N=\sqrt{(1-\vartheta/\log N)/R_N}$. Here $R_N\sim (\log N)/π$ is the truncated collision Green function at the origin for the simple random walk on $\mathbb{Z}^2$. Let $\mathfrak M_N$ denote the maximum of $\log Z_N(x)$ over $|x|_\infty\leq \sqrt{N}$. We prove that, for any $\vartheta\in \mathbb{R}$, in probability, $$\mathfrak M_N/\sqrt{\log N}\to 2\sqrt{2}\text{ as }N\to\infty.$$

fields
Probability
pdf ↗ page arXiv:2610.41989 ↗

Does the deepest drawdown take the longest to recover? For random walks with iid increments of positive finite mean, we find that the deepest and longest episodes are eventually distinct almost surely under light-tailed losses, but coincide with probability tending to one under heavy-tailed losses. More precisely, when losses have a finite exponential moment, the deepest episode's duration, relative to the longest among $n$ episodes, converges in probability to an explicit $ρ\in(0,1)$, while its descending duration rank is $n^{1-ρ+o_{\mathbb{P}}(1)}$. These limits hold for every fixed number of leading depths; in the Gaussian model, $ρ=1/2$, independently of drift and variance. The separation reflects the different exponential costs of reaching a large depth and sustaining a long episode. For regularly varying losses, a common large loss governs both depth and duration: the duration ratio tends to one almost surely, and for every fixed $k$, the $k$ deepest and $k$ longest episodes have the same indices in the same order with probability tending to one. For any $k_n\to\infty$ with $k_n=o(n)$, the proportion of common episodes among the $k_n$ deepest and $k_n$ longest converges in probability to zero in the light-tailed case and to one in the heavy-tailed case. These ranking laws also hold for completed drawdowns within a deterministic observation horizon.

fields
Probability
pdf ↗ page arXiv:2610.41990 ↗

In this paper we present a method to effectively and efficiently provide pathwise approximations of the Riemannian Brownian motion on a reductive homogeneous space $M\cong G/H$ endowed with a $G$-invariant metric. This setting includes spheres, hyperbolic spaces, the space of symmetric positive-definite matrices, flag manifolds and Stiefel manifolds. The simulation is based on Wong–Zakai approximations of an appropriately chosen SDE in $G$ that is projected down onto $M\cong G/H$. The effectiveness lies in the fact that well-established results on the convergence rate of the Wong–Zakai equation immediately carry over to the homogeneous space thanks to Nash's embedding. The efficiency lies in the fact that explicit representations of the Wong–Zakai approximations can be obtained in this setting using the Lie exponential map.

comments
55 pages, 10 figures. First version, all comments are welcome!
fields
Probability, Numerical Analysis
pdf ↗ page arXiv:2610.42025 ↗

Sampling from the ferromagnetic $q$-state Potts model is a fundamental problem in statistical physics and approximate counting. Although computationally hard in the worst case, sampling is expected to be efficient throughout the high-temperature uniqueness regime of the model using the natural single-site Glauber dynamics. We establish fast mixing bounds for every $q$, in the full high-temperature regimes in the two central graph families on $n$ vertices. (1) For every $Δ\ge 3$, the Potts Glauber dynamics mixes in $n\operatorname{polylog}(n)$ steps with high probability on a uniformly random $Δ$-regular graph for all $β<β_u(q,\mathbb T_Δ)$ where $β_u(q,\mathbb T_Δ)$ is the uniqueness threshold on the $Δ$-regular tree; (2) For every $d\ge 2$, the Potts Glauber dynamics has inverse spectral gap $n^{1+o(1)}$ on cubes or tori in $\mathbb Z^d$ for all $β<β_c(q,\mathbb Z^d)$ where $β_c(q,\mathbb Z^d)$ is the critical point on $\mathbb Z^d$. Both temperature thresholds are sharp as the Glauber dynamics slows down exponentially above them. These results overcome the challenge of bounding mixing times without access to correlation decay bounds in the presence of pinnings (i.e., without spectral independence or strong spatial mixing). For this purpose, our proof introduces a new dichotomy principle: throughout the high-temperature regime, any initial spin distribution is either such that its pinnings ensure good correlation decay, or Glauber dynamics escapes from that distribution quickly. We expect this kind of dichotomy to have further applications. As a demonstration, we also develop its analog for random-cluster dynamics and use it to prove that Swendsen-Wang dynamics on random $Δ$-regular graphs mixes in $O(\log n)$ time throughout uniqueness. The results in this paper were arrived at with the assistance of ChatGPT 5.6 and 6.

fields
Probability, Mathematical Physics
pdf ↗ page arXiv:2610.42059 ↗

For stochastic matrices of order $n$, the Karpelevič region is the set of all possible eigenvalues. We study matrices whose first and last $m$ rows are arbitrary probability vectors, while the remaining middle rows come from one probability vector by successive cyclic shifts to the right, moving the last entry to the first position each time. As $m$ increases, their eigenvalue regions expand from the polygon spanned by the $n$th roots of unity. For $n\ge3$, they first equal the full Karpelevič region at $m=\lceil(n-1)/4\rceil$. Reproducing all $n$ eigenvalues, counted with multiplicity, of every stochastic matrix of the same order requires $m=\lfloor n/2\rfloor$. Thus, in even order $n\ge4$, retaining two consecutive rows with the Toeplitz pattern can exclude eigenvalue lists even when every possible eigenvalue occurs individually. The boundary of the Karpelevič region consists of arcs joining roots of unity. We determine the least $m$ for which every point of each arc occurs as an eigenvalue in our family. For arcs whose nonreal points first become possible eigenvalues at order $n\ge4$, and for $m\ge1$ below this minimum, no point on the arc other than its endpoints occurs in the family. At the minimum, every point on the arc becomes attainable. At the smallest order $n\ge4$ at which a nonreal point on a boundary arc away from its endpoints can occur as an eigenvalue, that point determines the characteristic polynomial of every stochastic matrix of order $n$ having it as an eigenvalue. Further results describe the smaller regions and the angles at which their boundaries approach roots of unity.

comments
41 pages, 4 figures, 16 references
fields
Probability, Rings and Algebras
pdf ↗ page arXiv:2610.42072 ↗

The thin-film equation describes the evolution of a thin liquid film driven by capillarity and limited by viscosity. For sufficiently small film heights, thermal fluctuations may not be neglected. Since the thin-film equation as an overdamped evolution equation has a gradient flow structure, the informal implementation of the fluctuation-dissipation principle is straightforward. In this paper, we work out that the concept of Dirichlet forms and the related construction of Markov processes are tailor-made for this type of overdamped continuum evolution with thermal noise. However, the concept of Dirichlet forms relies on the availability of a Gibbsian equilibrium measure, which, in case of the thin-film equation, limits the treatment to the $(1+1)$-dimensional case. By means of Dirichlet form theory, we obtain the existence and uniqueness of the evolution of laws on configuration space. We show that the regularity of the realizations of this Markov process matches the heuristics. In a simplified situation with a Gaussian reference measure and a non-degenerate mobility function, we characterize it as what is called an energy solution. The latter provides a pathwise decomposition of the driving forces into a deterministic and a thermal one, which was the starting point. Applying the same techniques in the spatially discretized setting proposed in [Math. Comp. 92 (2023), 1931–1976], we obtain the same results uniformly in the discretization parameter. In particular, Galerkin as well as spectral approximations of the stochastic thin-film equation are shown to be uniformly tight in optimal topologies. In the simplified setting, we also prove Mosco convergence of the associated Dirichlet forms and thereby obtain convergence of these approximations.

comments
50 pages, 1 table
fields
Probability, Analysis of PDEs, Numerical Analysis
pdf ↗ page arXiv:2610.42086 ↗

We consider bipartite Erdős-Rényi graphs $\mathbb{G}(n, p, d/\sqrt{np})$ with two parts of sizes $n$ and $p$, respectively, and edge probability $d/\sqrt{pn}$. We show that for $d > b_* \log (n+p)$ all eigenvectors of its adjacency matrix, with eigenvalues away from zero, are delocalized with high probability. The threshold $b_*$ is an explicit function of the size ratio $n/p$ and optimal for eigenvector delocalization near the spectral edges. Our results extend to bulk eigenvectors down to the optimal threshold $d\geq \mathcal{D}\sqrt{\log (n+p)}$ for a constant $\mathcal D \geq 1$. In the supercritical scale $d \gg \log (n+p)$ we show that all eigenvectors are delocalized. Our results also hold for sparse rectangular random matrices with independent entries.

comments
20 pages, 3 figures
fields
Probability, Mathematical Physics, Combinatorics
pdf ↗ page arXiv:2610.42088 ↗

Let $G(n,p,m)$ be the random subgraph of the cycle power $C_n^m$ obtained by retaining each edge independently with probability $p$. We study connectivity, component geometry, and random walks throughout the regime $\log n\ll m<n/2$. We prove that $p\sim(\log n)/(2m)$ is a sharp threshold for connectivity. If $n(1-p)^{2m}\toλ\in(0,\infty)$, then the number of isolated vertices converges to $\mathrm{Poisson}(λ)$, and with high probability every other vertex lies in a single component $C_1$. In particular, the connectivity probability converges to $e^{-λ}$. The component $C_1$ contains a cyclic chain of connected sets occupying all but an $o(1)$ fraction of each consecutive block of length $\asymp m$, with $Θ(m\log n)$ edges between consecutive sets. Its diameter has order $n/m+\log n/\log\log n$ with high probability. For every fixed $c>1$, uniformly over $p\geq c\log n/(2m)$, the lazy-walk spectral gap is asymptotic in probability to that of $C_n^m$. In the same range, the mean hitting time of simple random walk satisfies, uniformly over distinct $x,y$, $$ H_G(x,y)=\frac{2|E(G)|}{d_G(y)} +\frac{r_{xy}(n-r_{xy})}{κ_m} +o_{\mathrm{Prob}}\left(n+\frac{n^2}{m^2}\right), $$ where $r_{xy}$ is cyclic distance between vertices $x$ and $y$, $d_G(z)$ is the degree of vertex $z$ and $κ_m=(m+1)(2m+1)/6$. Thus, for this random graph model, connectivity has the Erdős–Rényi asymptotics with $2m$ possible neighbours, while distances, relaxation, and hitting times retain the geometry of the underlying cycle.

comments
55 pages
fields
Probability, Combinatorics
pdf ↗ page arXiv:2610.42107 ↗

We prove that in rotationally invariant first-passage percolation on $\mathbb{R}^d$ with $d\geq2$, the passage time $T_n$ over a distance $n$ satisfies $\mathbb{E}[\exp(|T_n-μn|/n^{\frac{1}{2}(1-\varepsilon)})]\leq C$ for some $\varepsilon>0$, where $μ$ is the time constant. This implies that its variance is at most a constant times $n^{1-\varepsilon}$, the bound that Basu, Sidoravicius and Sly proved in the planar case. As in their work, our argument is an induction on the scale, but in place of their proof that the geodesic is chaotic, we use the averaging argument of Benjamini, Kalai and Schramm together with a logarithmic improvement of the Efron-Stein inequality due to Benaïm and Rossignol.

code
github.com/nitromannitol/FPP-variance ∎ Lean project (lakefile.lean, lean-toolchain) builds with 20 sorry warnings at cfe58d1, checked 2026-11-02 build details
comments
19 pages, 4 figures; Lean 4 formalization at https://github.com/nitromannitol/FPP-variance
fields
Probability
pdf ↗ page arXiv:2610.42124 ↗

Information Theory23

In this work, we introduce a state-space distance calibrated by stabilizer states based on the dual structure of the robustness of magic (RoM), and show that RoM is exactly represented as the distance to the free-state set. We then extend this geometry to state conversion by defining the directed conversion deficiency $C_{\mathcal F}$ as the distance between a target state and the set of states freely reachable from a given input state. The quantity $C_{\mathcal F}$ characterizes free convertibility and is related to the minimum stabilizer quasiprobability weight required to represent the conversion residual through the relation $1+2C_{\mathcal F}$. It also contains the conventional magic state resource lower bounds based on RoM and can detect stronger conversion obstructions. For single-qubit systems, we analytically construct a continuous family of mixed states with identical RoM but asymmetric convertibility.

comments
25 pages
fields
Information Theory, Quantum Algebra
pdf ↗ page arXiv:2610.38401 ↗

We study adaptive UAV sampling for radio environment map (REM) construction from non-Gaussian IoT observations. Integrated nested Laplace approximation(INLA) updates a Matérn GMRF posterior, with waypoints selected by entropy pre-screening and Sherman–Morrison lookahead over the reachable set. We show that the likelihood-induced Skew-Normal correction has a sextic entropy effect, explaining its limited influence on waypoint selection. Across 19 retained ray-traced windows,Random and lawnmower sampling cost a median 18.0\% and 14.0\% more residual NMSE than posterior-informed sampling, equivalently reductions of 15.3\% and 12.3\% when referred to those baselines, while differences among informative rules remain unresolved. Widening the prior shadowing standard deviation from 6.0 to 10.025\,dB improves NMSE and credible-interval coverage, with monotone improvement over a 4–16\,dB sweep on this corpus and no observed turning point, although overconfidence persists. Online width estimation from the UAV's own counts yields about 6.5\,dB without recovering the gain. Across 30 measured LoRaWAN propagation windows paired with simulated counts and scored only on received cells, the margin over non-adaptive sampling reproduces, the training-derived width minimizes the tested interval score, and GP-UCB has nominally $3$--$5\%$ lower NMSE. The results show substantial, corpus-dependent sensitivity to prior specification, while fine differences among acquisition rules have limited or unresolved effects, the preferred width depends on the corpus, scored cells and model mismatch.

fields
Information Theory
pdf ↗ page arXiv:2610.40723 ↗

Estimating biased finite-memory sources from a single trajectory is a fundamental problem with practical impact, for example in estimating entropy for cryptography. We study order-$d$ binary sources whose conditional probabilities have bias at most $ε$, and establish *sharp* sample complexity bounds: relative accuracy $γ$ with confidence $1-δ$ needs $γ^{-2}\,(\tfrac12-ε)^{-d}\,(\log d+\log\tfrac1δ)$ samples up to constants depending on $ε$, with matching upper and lower bounds and no dependence on mixing time or spectral gap. The proof rests on a sharp bound on expected hitting times of the sliding-window (de Bruijn) chain, which stay of order $1/π(w)$ uniformly in $d$ exactly when $ε<\tfrac3{10}$, and on majorization of the window masses by those of independent biased coins. Applied to published entropy-source validations, the bounds show that their trace lengths may be underpowered: $10^6$ bits cannot resolve all transitions of an $8$-bit model to better than about $4.5\%$ relative error.

fields
Information Theory, Probability
pdf ↗ page arXiv:2610.40845 ↗

Consider distributed databases that contain both sensitive (private) and non-sensitive (public) attributes. A user seeks to compute a function of public attributes and side information by interacting with the databases and downloading their responses. In this paper, we study the trade-offs among download cost, information leakage about the private attributes, and distortion between the computed function and its reconstruction, referred to as the capacity region. Our main contributions are as follows. For a single database, we provide a complete characterization of the capacity region for general discrete sources and characterize the capacity region of two different setups for Gaussian sources when the computed function is defined as the sum of the public attribute and side information. For distributed databases, we establish inner and outer bounds on the capacity region of discrete sources and we characterize the capacity region when the databases and the side information are mutually independent. As a special instance, we study lossless function computation, and results analogous to the lossy case are obtained.

comments
A preliminary version of this paper was presented at IEEE ISIT 2025
fields
Information Theory
pdf ↗ page arXiv:2610.41034 ↗

Integrated sensing and communications (ISAC) has been extensively studied from the perspective of communication-sensing performance tradeoffs. In MIMO communication systems, the diversity-multiplexing tradeoff (DMT) fundamentally characterizes the tradeoff between communication reliability and multiplexing gain, and DMT-optimal space-time codes achieve this optimal tradeoff without instantaneous channel state information at the transmitter (CSIT), making them particularly attractive for ISAC scenarios where accurate CSIT is difficult to acquire or maintain. In this work, we propose an analytical framework for characterizing the sensing behavior and communication-sensing gain tradeoff of DMT-optimal space-time codes in ISAC systems. Under a single-target detection model, we propose the sensing diversity order as the high-SNR sensing reliability metric. We further show that a class of DMT-optimal space-time codes, exemplified by Perfect codes, eliminates sensing blind spots for all nonzero codewords, thereby achieving joint DMT-sensing-diversity (DMTS) optimality. Furthermore, using the Perfect code as a representative DMT-optimal construction, we show that its isotropic average transmit covariance limits the achievable sensing gain when the target angular region is narrow. We therefore develop a full-rank spatial precoding framework and a low-complexity surrogate precoder that preserve joint DMTS optimality while shaping the transmit beampattern. Simulation results show that, for a target angular region of $θ\in[-π/10,π/10]$, the proposed design achieves approximately 3 dB sensing-gain improvement over the unprecoded Perfect code.

comments
Submitted to IEEE ICC 2027
fields
Information Theory
pdf ↗ page arXiv:2610.41144 ↗

LLM-guided program evolution can produce mathematical constructions, but validity, generality, and reproducibility require distinct evidence. We develop a researcher-in-the-loop discovery pipeline that combines program evolution, exact evaluation, mathematical generalization, certified nonexistence checks, and formal proof checking. We apply it to entanglement-assisted quantum error-correcting codes (EAQECCs). Starting from tasks that did not specify a family formula, evolved programs produced finite qubit constructions whose structure researchers generalized to the family $[[n,1,n-1;n-q-1]]_q$ for prime powers $q\ge3$ at $n\ge2q+1$ and for $q=2$ at odd $n\ge5$. Lean checks the algebraic certificate underlying this family. Together with entanglement lifting, the family closes 54 gaps in archived EAQECC tables. The released evidence includes 115 verified code records, ten nonexistence results with checked DRAT proofs, family implementations, and preserved search programs for exact checks and seeded replay. In a controlled cohort of eight paired generation blocks, validation-selected iterative programs achieve success in 46.1% and 31.6% of held-out executions at unseen lengths $n=13,15$, compared with 19.5% and 3.1% for independently generated programs. In this setting, program evolution and researcher analysis turn finite computational discoveries into a proved family with independently checkable evidence.

code
github.com/LeGenAI/llm_guided_EAQECC_search ∎ Lean project (formal/eaqecc/lakefile.lean, formal/eaqecc/lean-toolchain) does not build at 23904e5, checked 2026-11-02 build details
comments
25 pages, 6 figures. Code, data and Lean proofs: https://github.com/LeGenAI/llm_guided_EAQECC_search
fields
Information Theory
pdf ↗ page arXiv:2610.41183 ↗

Point clouds have become a fundamental data format for emerging three dimensional (3D) applications, while their wireless delivery remains challenging due to the large data volume and irregular geometric structure. Existing semantic point cloud transmission methods are mainly developed for single-input single-output systems, making them difficult to apply directly to multiple-input multiple-output (MIMO) systems. To address this issue, we propose a partition-oriented semantic transmission framework for MIMO (POST-MIMO) point cloud delivery. POST-MIMO partitions the input point cloud into spatially coherent blocks and aligns them with parallel MIMO streams. Based on this block-aligned representation, the transmitter employs a hierarchical encoder to capture both intra-block local geometry and inter-block global dependency, while the auxiliary side information is transmitted jointly to assist reconstruction. At the receiver, a progressive decoder reconstructs the block geometry in a progressive manner and restores the final point cloud with block-aware adjustment. Experiments demonstrate improved reconstruction quality and lower computational resource consumption than representative reference schemes under the considered MIMO setting.

comments
6 pages, 6 figures, 1 table. Accepted for IEEE GLOBECOM 2026
fields
Information Theory
pdf ↗ page arXiv:2610.41253 ↗

Quantifying correlations is important for understanding how quantum states and channels support information processing. Quantum umlaut information $U$ connects such correlations to reliability in hypothesis testing and certain communication settings, but its known formulas require classical descriptions of the underlying states. To estimate correlations without these descriptions, we introduce measured quantum umlaut information $U^M$ by replacing Umegaki relative entropy with measured relative entropy. Its variational formulation leads to a quantum generative adversarial framework for estimating $U^M$ from measurement statistics of unknown bipartite states and classical-quantum channels. This framework optimizes reference states and measurements without reconstructing the target states or channel outputs. For classical-quantum channels, it also optimizes the input probabilities. We further analyze the relation between $U^M$ and $U$, deriving a state-dependent gap bound and characterizing equality for full-rank bipartite states. Numerical simulations show trained estimates close on average to independently computed numerical references for $U^M$ across the tested states and channels. These estimates could help assess how well quantum memories preserve correlations and how noise or drift affects input-output dependence in quantum transmitters and state-preparation devices.

comments
22 pages, 7 figures, 1 table
fields
Information Theory
pdf ↗ page arXiv:2610.41282 ↗

Thresholding methods provide a simple iterative framework for sparse signal recovery by combining a search direction with a thresholding operation. Gradient-based search directions are computationally efficient because they require only first-order information for the error metric. We incorporate componentwise gradient scaling into this framework by preserving large-magnitude components and attenuating the remaining components of the gradient of error metric. Combining this strategy with hard thresholding yields scaled-gradient hard thresholding (SGHT) and its pursuit variant SGHTP. Under suitable restricted isometry conditions, we establish recovery error bounds for both methods that account for measurement noise and deviations from exact sparsity. Numerical experiments indicate that SGHTP achieves recovery performance comparable to established pursuit methods on the tested ECG signals.

fields
Information Theory
pdf ↗ page arXiv:2610.41302 ↗

Traditional phased arrays (PAs) achieve near-field (NF) beamfocusing through spherical wavefront curvature, but lose this range selectivity in the far field (FF). Frequency diverse arrays (FDAs) instead introduce frequency offsets across antenna elements. These offsets synthesize a range-dependent beam whose beamdepth is governed by the aggregate frequency-offset span rather than the physical aperture, thereby extending finite beamdepth into the FF. Although FDAs have been extensively studied for radar, their broader potential for wireless communications remains largely unexplored. Accordingly, this article examines how this capability can support wireless communications and integrated sensing and communication (ISAC). We compare PA and FDA focusing mechanisms, explain FDA coupling and periodicities, and assess coexistence with orthogonal frequency-division multiplexing (OFDM). Integer-multiple offsets preserve alignment with the OFDM subcarrier grid and avoid offset-induced intercarrier interference under ideal synchronization, but require suitable symbol mapping and receiver processing. We also distinguish FDA beamdepth from waveform-based radar range resolution and discuss their complementary roles in sensing. For the configurations studied, FDAs offer no additional single-user spatial degrees of freedom. Interestingly, their principal benefit is multiuser separation through location division multiple access (LDMA). Numerical case studies under an idealized FDA channel model show that FDA-LDMA achieves higher sum-rates than PA-based schemes for users with small angular separation in the FF, even after accounting for the bandwidth reserved for frequency offsets.

comments
8 pages, 5 figures. Interactive figures and code: https://akadircelik.com/demos/fda-ofdm-isac/
fields
Information Theory
pdf ↗ page arXiv:2610.41415 ↗

Low Earth Orbit (LEO) constellations are a promising way to connect Internet of Things (IoT) devices beyond terrestrial coverage, yet the intermittent visibility of their satellites imposes hard constraints on system and protocol design. Coverage probability and other snapshot statistics, which are often computed without accounting for the terminal's latitude, are insufficient for this purpose. Designers also need the temporal structure of coverage: the rate of coverage gaps, their duration distribution and mean, and quantities such as the probability that a retransmission succeeds after an initial transmission has failed. We develop an ergodic framework for Walker Star constellations in which these quantities reduce to spatial integrals over a two-dimensional torus that can be evaluated efficiently. We validate the framework against a seven-day Simplified General Perturbations 4 (SGP4) simulation of the Iridium NEXT constellation and find close agreement for the coverage probability, the revisit-time distribution, and the retransmission success probability. A comparison of Walker Star configurations further shows that constellations with similar satellite counts can yield substantially different packet-delivery probabilities with at most one retransmission, depending on the number of orbital planes. The results show that the ergodic model accurately gives the long-run values that finite time-domain simulations can only approximate, providing researchers with a tractable basis for the analytical modelling of IoT connectivity over non-terrestrial networks (NTN).

fields
Information Theory
pdf ↗ page arXiv:2610.41440 ↗

Identification differs from classical transmission in that the receiver determines whether a particular identity was sent rather than recovering a message. We study secure deterministic identification over state-dependent discrete memoryless wiretap channels with private noiseless feedback and a sensing distortion constraint. The encoder has no local randomness and uses only the identity and past feedback. A symbol-wise input restriction guarantees the sensing constraint. We show that the secure deterministic identification capacity equals its non-secure feedback counterpart when the admissible main channel has positive ordinary transmission capacity and H(Y|Z,X=x)>0 for some fixed admissible input x. This equality does not require positive classical wiretap secrecy capacity. For achievability, noiseless feedback generates common randomness for identification and enables secret-key extraction by privacy amplification. The extracted key one-time-pad encrypts a short identification tag, whose reliable transmission requires only a sublinear number of additional channel uses. We also present an alternative construction in which feedback-generated randomness replaces the local randomization of a stochastic wiretap encoder. We further prove that secure identification capacity vanishes whenever non-secure identification capacity is zero. We also identify a class of channels with input-independent wiretapper observations that achieves full non-secure identification capacity even when the private conditional feedback entropy is zero, illustrated by a binary state-dependent wiretap channel. Thus, feedback can support both identification and secrecy without local encoder randomization while satisfying sensing constraints.

fields
Information Theory
pdf ↗ page arXiv:2610.41646 ↗

We study finite $2\times2$ Alamouti-type space–time codes associated with classical maximal orders in definite quaternion algebras over $\mathbb Q$. For the family indexed by primes $q\equiv3\pmod4$, we prove that the Eisenstein case $q=3$ uniquely minimises the finite quotient-alphabet energy at fixed minimum lattice distance and alphabet size. This ordering holds for every integer $p\geq2$, including when the representatives and their common translation are jointly optimised, with an explicit positive gap for every $q>3$. For uniform finite inputs, we derive a controlled high-SNR expansion of the information variance and determine its limiting ratio to the mutual information deficit. The mutual information expansion specialises arbitrary input fading results and gives an explicit remainder. Standard random-coding and second-order achievability tools relate these quantities to reliability over independently faded Alamouti blocks. All alphabets preserve symbol-wise maximum-likelihood detection at the Alamouti level. Numerical energy, mutual information, and error-rate comparisons support the analysis. Centred comparisons with classical Alamouti, Golden, and Silver codes illustrate the trade-off between reliability, symbol rate, and decoding effort.

fields
Information Theory
pdf ↗ page arXiv:2610.41693 ↗

Reusing the Rate-Splitting Multiple Access (RSMA) common stream for sensing eliminates dedicated probing signals in Integrated Sensing and Communication (ISAC), but creates a feasibility bottleneck: the common-power fraction must simultaneously support common decoding, private decoding, and target detection. This paper investigates this tradeoff in downlink zero-forcing RSMA with imperfect channel state information at the transmitter (CSIT) and imperfect successive interference cancellation (SIC). With the common beam fixed over a sensing dwell through sensing geometry, the common-power fraction becomes the sole design variable. We prove that the communication-feasible set is either a closed interval or empty. Analytical outage expressions localize these endpoints, revealing a high-signal-to-noise ratio (SNR) floor-ceiling effect: common decoding dictates a common-power floor, while imperfect CSIT and SIC impose a private-decoding ceiling. For sensing, the conditional detection constraint is affine, preserving the interval structure. A geometry-matched common beam ensures monotonic detection improvement, reducing joint-feasibility check to the communication-feasible upper endpoint. Simulations validate our analysis, identifying regimes where common-stream-only sensing is viable.

comments
Accepted at IEEE GLOBECOM 2026. 6 pages, 5 figures
fields
Information Theory
pdf ↗ page arXiv:2610.41744 ↗

Bent partitions play a significant role in the theory of bent functions. In this paper, we study weakly regular bent partitions, that is, all generated bent functions are weakly regular (called in class $\mathcal{WRBP}$). Let $Γ=\{A_{i}, i \in V_{m}\}$ be a bent partition of $V_{n}$ in class $\mathcal{WRBP}$, where $V_{n}$ denotes an $n$-dimensional vector space over the prime field $\mathbb{F}_{p}$, $n$ is even, $m \leq \frac{n}{2}$, and $m \geq 2$ if $p=2$. We first characterize bent partition $Γ$ in terms of the duals of component functions of vectorial bent functions. In particular, we prove that if $p \neq 3$, then all bent functions generated by $Γ$ are regular. Building on the above result, we give characterizations of bent partition $Γ$ in terms of two classes of linear codes. When $n \geq 4$ and $m=\frac{n}{2}$, except for the case where $(p, n)=(2, 4)$ and $|A_{i_{0}}|=1$ for some $i_{0}$, we show that $Γ$ is obtained from a spread (after a shift). Particularly, any normal bent partition of $V_{n}$ of depth $p^{\frac{n}{2}}$ in class $\mathcal{WRBP}$ must be obtained from a spread (after a shift), where $n \geq 4$. We establish the Macwilliams-like identities for bent partitions in class $\mathcal{WRBP}$. Furthermore, if $p$ is odd, we present some nonexistence results on bent partitions in class $\mathcal{WRBP}$.

fields
Information Theory
pdf ↗ page arXiv:2610.41779 ↗

We study channel interconversion under non-signalling superchannels, where the asymptotic conversion rate is given by the ratio of the channel capacities. Our main result determines the second-order expansion of the optimal target length. Namely, when each channel has a single capacity-achieving dispersion, the coefficient is given by the inverse sesquinormal distribution, with parameters determined by the capacities and dispersions. The result extends state-dichotomy theory to channels: one superchannel approximates the target uniformly over inputs and maps every reference channel whose output is independent of its input to another such channel. As in state dichotomies, there are finite-error resonances where the second-order term vanishes, leaving an $O(\log n)$ remainder. Crucially, our new characterization cannot be obtained by concatenating channel coding and simulation protocols [Cao {\it et al.}~IEEE Trans.~Inf.~Theory (2024)]. Instead, the proof combines comparisons of binary hypothesis testing experiments with uniform Gaussian estimates and a construction that uses symmetry within type classes to satisfy the non-signalling constraints. Shared randomness assistance already suffices at first order, but we show that its second-order behaviour can differ strictly from the non-signalling case.

comments
34 pages, 4 figures
fields
Information Theory
pdf ↗ page arXiv:2610.41858 ↗

Modern language models motivate renewed interest in universal prediction problems in which a predictor can exploit previously observed data before predicting a fresh sequence. We study an idealized version of this setting by introducing \emph{batch regret}: a conditional analogue of classical average regret in which a predictor observes a batch of $n$ training sequences and is then evaluated on an independent test sequence generated by the same unknown source. Additive smoothing estimators, which add a bias to the observed token frequencies before predicting the next token, played a fundamental role as a simple prediction strategy with almost-optimal average regret for memoryless and Markov sources. Their equivalent under batch regret are the conditional additive smoothing estimators, which estimate the conditional distribution of the test sequence given the training batch. In this work, we study their asymptotical batch regret for binary memoryless and first-order Markov sources. For binary memoryless sources, we show how the optimal bias of the estimator, and consequently the corresponding batch regret, change as a function of the training batch size $n$, by extending the Poisson-based analysis of Krichevsky. For binary Markov sources with positive transitions, we characterize the leading coefficient of batch regret and its dependence on whether the initial law is known, free, or stationary. Finally, we introduce a Rényi analogue of batch regret and identify the corresponding optimal predictors as conditional $α$-NML estimators, thereby giving an operational interpretation to a conditional version of Sibson's $α$-mutual information.

comments
Submitted to the IEEE Transactions on Information Theory
fields
Information Theory
pdf ↗ page arXiv:2610.41859 ↗

The problem of transmitting both data and state over a memoryless state-dependent multi-access channel with causal state information is addressed. The state comprises three independent components: a common component, which is revealed to both en- coders, and two private components, each of which is revealed to a different encoder. The data are to be recovered with arbitrarily-small probability of error, and the state sequence with arbitrarily-small symbol-error rate. Under the assumption that the capacity region in the absence of a state recon- struction constraint is nonempty, the capacity region is derived and is shown to be achievable using a Block-Markov coding scheme employing binning and superposition coding.

comments
31 pages, 2 figures
fields
Information Theory
pdf ↗ page arXiv:2610.41860 ↗

Reinforcement learning (RL) embodies a foundational paradigm for sequential decision-making in the face of uncertainty. The rapid growth of model complexity in emerging applications has significantly amplified the challenges of achieving data-efficient RL, particularly in sample-starved scenarios where data collection may be costly, time-consuming, or even high-stakes. Consequently, understanding and improving the data efficiency of RL algorithms has become a central problem of both fundamental theoretical interest and practical importance. Despite decades of research, the statistical underpinnings of RL remained remarkably underdeveloped until relatively recently, particularly in the finite-sample regimes most relevant to modern practice. Over the past few years, there has been a surge of activity advancing the frontiers of non-asymptotic statistical RL theory, drawing on ideas from high-dimensional statistics and probability, information theory, and statistical learning theory. In this monograph, we present a non-asymptotic treatment for RL based on tabular Markov decision processes, focusing on three canonical sampling mechanisms --- RL with a generative model, online RL, and offline RL --- together with two algorithmic paradigms: model-based and model-free approaches. Centered around non-asymptotic sample complexity analysis, our treatment emphasizes minimax-optimal algorithm design, particularly methods that attain statistical optimality without significant burn-in costs, as well as sharp statistical characterizations of several widely used RL algorithms. Whenever possible, we provide largely self-contained analyses together with intuitive explanations. We also briefly discuss several extensions, such as robust RL, competitive multi-agent RL, federated RL, uncertainty quantification, and RL from human feedback.

fields
Information Theory
pdf ↗ page arXiv:2610.41903 ↗

$K$ users each observe a fixed linear function of common data over a finite field of characteristic $p$, and a server must recover another fixed linear function with zero error from one message sent by each user. The capacity, the minimum total number of symbols the users send per computed instance, depends only on the configuration of subspaces formed by these linear functions. We study the instance in which three users observe $\boldsymbol{a}$, $\boldsymbol{b}$ and $\boldsymbol{a}+\boldsymbol{b}$ and the server wants $\boldsymbol{a}+λ\boldsymbol{b}$. The four functions span four lines in a plane with cross-ratio $λ$, the smallest instance of the four subspace problem of representation theory. It is the smallest configuration whose capacity is not determined by the dimensions of the subspaces formed by sums and intersections: these dimensions are the same for every $λ$, but the capacity depends on the cross-ratio and on the prime subfield. For the harmonic cross-ratios, $λ=-1$, $λ=2$ or $2λ=1$, it is $2$: every zero-error code reveals $(\boldsymbol{a},\boldsymbol{b})$ to the server, so no code beats having two users send their observations. For $λ$ outside the prime subfield it is the cut-set bound $3/2$, achieved by codes linear over the prime subfield, while codes linear over the field need $2$. For the other $λ$ in the prime subfield it lies in $[3p/(2p-1),2]$, so no code at any block length approaches the cut-set bound. The qualitative separation persists under vanishing error for uniform data: the cut-set bound is tight exactly for $λ$ outside the prime subfield. The four-line instance is also exact repair of one node of a $[4,2]$ MDS storage code, and the results translate to that setting.

fields
Information Theory
pdf ↗ page arXiv:2610.42023 ↗

This paper introduces embodied movable antennas (MAs), a new MA architecture that rethinks the source of antenna mobility. Unlike conventional MAs that rely on dedicated antenna actuators, embodied MAs harness the mechanical degrees of freedom already present in physical agents to reconfigure antenna position and orientation. We distinguish two modes of embodied MAs: task-coupled and task-preserving. Focusing on the latter, we show how wireless channels can be reconfigured by exploiting task-equivalent host configurations without altering the prescribed physical task. Using a redundant robotic manipulator as a canonical realization, we characterize the intrinsically coupled antenna translation and orientation induced by task-preserving self-motion and derive an exact Fourier representation of the resulting multipath channel. Our analysis reveals that the wireless value of embodied mobility depends not merely on how far an antenna can move, but on which configurations remain feasible under the physical task and how their coupled motion interacts with polarization and multipath propagation. Building on this insight, we develop a cross-task antenna mounting design that enables wireless adaptation with a fixed physical installation. Numerical results demonstrate substantial communication gains through task-preserving motion. These findings establish mechanical redundancy as a new, quantifiable resource for wireless channel reconfiguration.

fields
Information Theory
pdf ↗ page arXiv:2610.42078 ↗

The Pareto secretary problem asks for online selection of a Pareto-maximal observation from an i.i.d. sample: no other observation is at least as good in every coordinate and strictly better in one. For every fixed dimension $d\ge2$ and independent continuous coordinates, the optimal failure probability is of order $1/[(\log n)^{d-1}\log\log n]$, both with full information and with coordinate ranks alone. A deterministic rank rule attains the optimal leading term uniformly over the marginals. For every fixed positive definite Gaussian covariance, the same failure order holds, with a leading constant depending on the correlation matrix, and a rank-based rule attains the optimal leading failure probability without knowing the Gaussian parameters. In both models, we determine the optimal fixed cutoff and its leading failure constant; its failure probability exceeds the unrestricted optimum by a factor $Θ_d((\log\log n)^2)$. Under arbitrary dependence, a fixed joint density with uniform marginals can instead have optimal full-information success tending to approximately $0.58$. The proof uses the scarcity of observations with very small domination probability to bound every stopping rule. A matching rule selects a currently maximal point when this probability is below a threshold that increases as time runs out. Coordinate ranks from an initial sample provide sufficiently accurate estimates to preserve the optimal leading failure probability. For singular Gaussian distributions with known correlations, we characterize the optimal failure asymptotics attainable using ranks. Singular dependence can make every observation acceptable or recover the difficulty of selecting a single best point.

comments
48 pages, 8 figures. A preliminary version has been accepted to WINE 2026
fields
Information Theory
pdf ↗ page arXiv:2610.42101 ↗

Instance-dependent limits for bandit regret, sequential tests, best-arm identification and Markov decision processes are proved by one chain. It runs from a divergence decomposition through data processing to a statistic, a testing step that turns a class guarantee into a requirement on the history, and optimization over alternatives and allocations. Only the first step is an identity. This work writes the chain as an exact ledger. By the chain rule, data processing loses exactly a conditional divergence, which vanishes for arms in natural exponential families at the per-arm counts and reward sums. A one-parameter family of martingale identities extends the decomposition to every tilt between two instances. The classical limits follow by relaxations with named defects. All defects but the asymptotic step's can be computed from an algorithm's own runs, the uniform quantification's only when the class guarantee is a number. On seven published sources, from movie ratings to engine replacement, the ledger locates the difference between standard algorithms and the classical limits. For identification among trained models and for Wald's test the largest term is usually the uniform quantification's.

fields
Information Theory, Statistics Theory
pdf ↗ page arXiv:2610.42133 ↗

Mathematical Physics22

Action variables are adiabatic invariants: their values are preserved during adiabatic time evolutions in completely integrable classical Hamiltonian systems. Nevertheless, an adiabatic cycle can alter how an action variable depends on phase-space variables that are independent of adiabatic parameters. Such a change, which we call action variable holonomy, induces adiabatic torus pumping in phase space. For a family of completely integrable kicked-top systems, a homotopy classification of adiabatic cycles identifies which adiabatic cycles induce action variable holonomy and adiabatic torus pumping. A physical example is also provided.

comments
8 pages, 3 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.40671 ↗

Constructing accurate and physically interpretable tight binding (TB) Hamiltonians from first principles calculations remains challenging, particularly for materials exhibiting complex orbital hybridization, spin orbit coupling (SOC), and magnetic order. SKFitter is an open-source Python framework for constructing and optimizing symmetry resolved Slater Koster TB models directly from density functional theory (DFT) calculations performed using Quantum ESPRESSO or VASP. The framework supports s to f orbitals, incorporates on site SOC and site resolved magnetic exchange interactions, and accommodates both orthogonal and nonorthogonal basis. A central feature is the symmetry resolved parameterization of orbital manifolds, which allows independent two centre hopping integrals between symmetry allowed orbital groups and across distinct neighbour shells. SKFitter also identifies missing orbital contributions and guide systematic basis set refinement and optimizes parameters through weights. The resulting Hamiltonian can be used to evaluate anomalous (AHC) and spin Hall conductivities (SHC), and topological invariants. We demonstrate the framework using eight representative materials spanning semimetals, semiconductors, transition metals, f-electron compounds, and magnetic systems. Symmetry resolved models achieve band energy errors of 9 to 41 meV within 1 eV of the Fermi level. The fitted Hamiltonians reproduce characteristic features of the SHC response of Pt, the SOC dependence of the AHC of Fe, and the momentum dependent altermagnetic spin splitting of MnTe without any altermagnet specific interaction. For EuMn2Bi2, projection guided orbital refinement reduces the near Fermi level fitting error from 120 to 32 meV. Together, these results establish SKFitter as a flexible and reproducible framework for developing symmetry consistent, physically interpretable TB models of complex materials.

comments
Open source software available at https://github.com/sundeep-kapila/skfitter
fields
Mathematical Physics
pdf ↗ page arXiv:2610.40729 ↗

High-fidelity CFD is essential for coupled hypersonic aerodynamic and aerothermal phenomena, but its cost precludes multi-objective design optimization. We develop three DeepONet surrogates for two hypersonic configurations, unified by one device: the trunk receives coordinates in which the varying feature is held fixed. The first is a prescribed family of 37 spherically blunted tangent ogives, varied through ogive radius of curvature to span nose-tip ablation forebody change; no ablation is computed. A body-fitted canonical transformation maps every domain onto a common (s, eta) space, retaining physical coordinates for forces and heat transfer. A branch network predicts reduced coefficients and a Transolver trunk with Physics-Attention learns the spatial basis; a two-step strategy optimizes both jointly, orthonormalizes the basis by QR factorization, and trains a second-stage branch for unseen geometries. Surrogates cover density, pressure, velocity, temperatures, conductivities and species mass fractions. In leave-one-out tests, drag, mean and peak wall heat flux are recovered to 1.16%, 1.62% and 5.29%; a sweep of 1001 geometries gives a 200-design Pareto front trading 6.68% in drag for 4.42% in peak heating (a diagnostic, not an objective). The second is the 3D AEDC waverider at 29 angles of attack: static mesh, moving leading-edge shock. For surface quantities, the same construction predicts integrated heat load within 0.22% and force coefficients to errors below 2.6x10^-5; Pareto fronts identify angles of attack trading drag, lift and heating. For the volume field, a Cartesian trunk smears the shock; a frozen ShockNet supplies a shock-aligned coordinate and a hotspot-aware mixture of experts corrects the residual, recovering density to 12.28% relative L2 error at a held-out incidence (5.08x10^7 nodes). The 1001-design sweep takes about three minutes on a single CPU.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.40731 ↗

We prove an entropic $c$-theorem in two dimensions by constructing a $c$-function from the mutual information between spacelike separated wedges, defined through the relative entropy. The proof follows Casini and Huerta's, with strong subadditivity of entanglement entropy replaced by increasing differences of the mutual information, which we prove directly in the continuum from the split property and standard properties of relative entropy. The benefit of relative entropy is that every quantity we use is finite and no cutoff ever needs to be introduced. The assumptions of the theorem are only the standard axioms of algebraic quantum field theory, the split property with finite mutual information across a gap (both implied by modular $p$-nuclearity), and a finite ultraviolet bound on the resulting $c$-functions. We also note that, formally, the $c$-function is proportional to the minimal averaged null energy of a state that reproduces the vacuum correlations between the two wedges.

comments
8 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2610.40874 ↗

We analyze scaling limits in algebraic quantum field theory using continuous bundles of C*-algebras. Different conditions of continuity on such bundles encode different strengths of approximation of the theory by its scaling limit. As a proof of concept, we consider the small-distance scaling limit of the theory of a massive free scalar field, which corresponds to a massless free scalar field. We provide a number of examples of local algebras of observables for which the corresponding bundle of C*-algebras representing the scaling limit is merely lower semicontinuous. Further, we construct an algebra of nonlocal observables for which the corresponding bundle of C*-algebras representing the scaling limit is uniformly continuous.

fields
Mathematical Physics, Operator Algebras
pdf ↗ page arXiv:2610.40924 ↗

Weyl points are quasimomentum-energy pairs in the band structure of three-dimensional periodic media at which two dispersion surfaces touch conically. They first arose in a model introduced by H. Weyl in 1929, but were first observed only recently in condensed matter physics and photonics experiments. Weyl points have been explored theoretically via numerical computation and via tight binding (discrete) models. In this article we present the first analytical treatment of Weyl points in a continuum PDE-type model. Starting with a Schrödinger operator whose electric potential is periodic with respect to the body-centered cubic (BCC) lattice, we introduce a small BCC-periodic magnetic vector potential that is equivariant with respect to a specific abelian group of symmetries. For magnetic Schrödinger operators of this type, we prove (i) that a pair of Weyl points bifurcates from a threefold degeneracy in the band structure of the underlying non-magnetic operator, and (ii) that these Weyl points are stable against small self-adjoint BCC-periodic perturbations of both the electric and magnetic vector potentials.

comments
55 pages, 4 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.40963 ↗

In this paper, we study the Connes spectral distance as a quantum distance measure and its applications in discriminating quantum states and calculating quantum entanglement. Based on finite matrix algebra, we construct a spectral triple for finite dimensional quantum states. The Connes spectral distance is a generalization to the quantum trace distance and Hilbert-Schmidt distance. The corresponding optimal elements can always be chosen to commute with the difference matrix $Δ$, and then the optimal element and $Δ$ can be simultaneously diagonalized. We obtain the formulas of Connes spectral distances in three different parameter regimes which depending on the spectrum of difference matrix. As applications, we show how to use spectral distances to distinguish state pairs that have the same trace distance and Hilbert-Schmidt distance. We also define a new geometric measure of entanglement based on spectral distance. These results can help us study the relationships among Connes spectral distance, quantum trace distance and other quantum distance measures. These concrete constructions are significant for studies of quantum states in the framework of noncommutative geometry.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.40978 ↗

We obtain exact solutions of the nonlinear Dirac (NLD) equation in $1+1$ dimensions of the form $ψ(x,t) = e^{-iωt} ψ(x)$ for all possible combinations of scalar-scalar (SS), vector-vector (VV) and pseudoscalar-pseudoscalar (PP) interactions at nonlinearity parameter $κ= 1$, which corresponds to the most general 4-Fermion interactions. We look for solutions where $0 < ω< m$ where $ω, m$ are frequency and mass respectively. Remarkably, we show that at $κ= 1$, all the NLD models with SS, VV and PP interactions can be reduced to suitable combinations of SS and VV interactions and hence all solitary waves in these models must also be bound states. We show that two of the ten possible combinations are equivalent to $SS -(1/p) VV$ interactions while two other combinations are equivalent to $VV -(1/q) SS$ interactions for which we already know the results. For the remaining six combinations of SS, VV and PP, we show that depending on the parameters, whereas in some cases entire range of $0 < ω< m$ is allowed, in some other cases $ω$ is restricted to a smaller range $ω_{min} < ω< m$. Further, depending on the parameters, while the solitary waves in some cases have only one hump, in other cases the solitary waves have both single and double hump depending on the value of $ω$. We compute the charge $Q$ and the energy $E$ for each of the solitary wave solutions and show that while $E$ and $Q$ depend on coupling constant $g$, their ratio $E/Q$ is independent of $g$. When $κ= 1$, all the NLD models with SS, VV and PP interactions can be reduced to suitable combinations of SS and VV interactions. As a result all solitary waves in these models are bound states with $E/Q<m$.

comments
30 pages, 9 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.41084 ↗

Subregions in gauge theories give rise to edge modes and corner charges supported on their boundaries. While this is well-established classically and in quantum lattice gauge theories, the fate of corner modes remains poorly understood in continuum quantum field theories owing to their distributional nature. In this paper, we revisit this challenge and propose a new corner operator for a regional Cauchy slice algebra generated by local gauge-invariant observables of scalar QED in Minkowski spacetime. The corner mode is given by the longitudinal electric field sourced by the matter outside the subregion and thus constitutes a centre for the regional algebra. Our main goal is to show that there exists a perturbative GNS state on this algebra that quantises this mode. Pulling back the Fock vacuum to the slice, we construct this state by renormalising the divergences arising from the lack of temporal smearing. We clarify that the inclusion of the new centre mode necessitates a further renormalisation on the corner of the region, thereby physically distinguishing it and breaking the translation invariance of the state. Our work shows how the requirements of locality, renormalisability, and positivity together put strong constraints on the existence of a continuum quantum field theory which supports corner modes as genuine quantum operators. We finish with comments about the von Neumann completion of the regional algebra, as well as possible physical interpretations of the renormalised vacuum state.

comments
70 pages (50 + appendix)
fields
Mathematical Physics
pdf ↗ page arXiv:2610.41240 ↗

We consider the RP$^2$ model on the two-dimensional square lattice with nearest-neighbor interactions. It is obtained by replacing the spin in the standard classical Heisenberg model with a director, which does not distinguish two opposite directions. Unlike the standard classical Heisenberg model or the O(3) nonlinear sigma model, the RP$^2$ (and its cousin RP$^3$) model allows point vortex defects classified by $π_1(RP^2)= Z_2$ in two dimensions. It is tempting to expect the $Z_2$ vortices to induce a phase transition analogous to the well-established Berezinskii-Kosterlitz-Thouless transition in the O(2) model (XY model) in two dimensions. Indeed, the possibilities of a critical phase or a critical point with power-law correlations of the order parameter in the RP$^2$ and RP$^3$ models have been discussed in the literature. However, the challenging nature of the problem has prevented a definitive conclusion on the phase diagram so far. In this paper, we report a rigorous mathematical proof by Artificial Intelligence (ChatGPT 6 Astra) that the absolute value of the two-point correlation function of order parameters at distance $r$ in the RP$^2$ and RP$^3$ models on the two-dimensional square lattice is bounded above by $C_β\exp{(-c_β[\log(2+r)]^2)}$ for any finite inverse temperature $β\ge0$, with positive constants $C_β$ and $c_β$ depending only on $β$. Namely, the correlation function decays faster than any power of $r$. The proof is formalized and verified in Lean; the formalization is publicly available at doi:10.5281/zenodo.23548847 . A heuristic argument based on a one-parameter effective theory ansatz at large distances combined with the rigorous correlation bound suggests (but does not prove) that the $Z_2$ vortices are always unbound at any nonzero temperature.

comments
35 pages, no figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.41256 ↗

We study the ground states of the antiferromagnetic three-state Potts model in which the third state carries a single-ion energy cost $D_3$, on arbitrary lattices, and show that the difficulty of the ground-state problem is controlled by the maximum coordination $Δ$ alone; on planar lattices with $Δ\le5$ it is classified completely. On lattices with coordination at most three the third state heals frustrated bonds one for one: every graph of maximum degree three other than $K_4$ has an independent set of sites that meets every odd cycle and whose size equals the number $\operatorname{fr}$ of frustrated bonds of the Ising ground state. This identity also follows from known results on the chromatic vertex stability of graphs; we give a constructive proof. Hence $E_0 (D_3)=\min(D_3,1) \operatorname{fr}$ for all $D_3$, computable in polynomial time on planar lattices. In graph-theoretic terms, independent odd cycle transversal is solvable in polynomial time on planar subcubic graphs. On planar lattices with maximum coordination four, in contrast, deciding the ground-state energy is NP-complete for every fixed $0\le D_3<2$ and polynomial for $D_3\ge2$. The boundary is set by a local principle: a site of coordination $d$ in the third state relieves at most $\lfloor d/2\rfloor$ frustrated bonds, so the third state is useless once $D_3\ge\lfloorΔ/2\rfloor$; this threshold is sharp for every $Δ\ge3$ and reproduces the triangular-lattice point $D_3^*=3$ of the equivalent spin-1 Blume–Capel antiferromagnet. From a single Ising ground state the proof builds an exact ground state in polynomial time, which for quasicrystalline, foam-like and defective lattices with thousands of sites takes seconds to minutes.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.41277 ↗

Neural networks provide a promising approach to learning and generating the stochastic dynamics of nonequilibrium systems. In this study, we propose a stochastic dynamics generative method based on a two-channel convolutional neural network, using the (1+1)-dimensional directed bond percolation model to investigate its ability to learn local stochastic transition rules and reproduce stochastic dynamics across different system sizes. The proposed method takes the current particle configuration and the control parameter as inputs, learns the conditional activation probabilities at the next time step through local convolutions, and generates stochastic temporal evolution through iterative Bernoulli sampling. The results show that the network-generated dynamics agree well with Monte Carlo simulations across different control parameters. Furthermore, applying the trained network to a substantially larger system with $L=10000$ yields an estimated critical probability of $p_c=0.643$ and an effective density decay exponent of $δ\approx0.1592$, close to the established theoretical values. These findings demonstrate that the proposed method can reproduce the stochastic dynamics of directed percolation and capture its macroscopic dynamical and statistical properties across different spatial sizes.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.41299 ↗

We study the root-of-unity limit of the multi-component spin solution of the star-star relation whose Boltzmann weights are products of lens elliptic gamma functions, and whose spins take values in $\mathbb{R}^{n-1}\times(\mathbb{Z}_r)^{n-1}$. When one of the elliptic nomes tends to a primitive $2rN$-th root of unity, the quasi-classical expansion gives at leading order the classical star-star relation of the Bazhanov-Sergeev multi-spin model; at the next order, under the standard assumptions on the saddle points, a $(\mathbb{Z}_{rN})^{n-1}$-state solution of the star-star relation, which depends on $r$ and $N$ only through $rN$. Real $\mathbb{Z}_n$-symmetric configurations of the continuous spins give elliptic $\mathfrak{sl}(n)$ Kashiwara-Miwa type models labeled by $n$-torsion points. Extended to general primitive torsion points, these include the broken $\mathbb{Z}_{rN}$ model of Quano and Fujii, for which the star-star relation follows from their Yang-Baxter equation when $\gcd(n,rN)=1$. In the trigonometric limit the saddle point equations single out the spectral curve of the $\mathfrak{sl}(n)$ chiral Potts model of Bazhanov, Kashaev, Mangazeev and Stroganov, and the Boltzmann weights become those of that model.

comments
50 pages, 4 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.41438 ↗

Accurate magnetic-field evaluation on and near current-carrying surfaces is a recurring computational bottleneck in magnetic-confinement fusion. As targets approach the source surface, Biot–Savart and virtual-casing kernels become singular, causing standard Gauss–Legendre and upsampled trapezoidal quadratures to lose accuracy. We introduce AMPERe, a quadrature-by-expansion (QBX) evaluator for vector magnetostatic kernels on general three-dimensional, non-axisymmetric tokamak and stellarator geometries. The method develops a spherical-harmonic QBX formulation for the vector Biot–Savart and virtual-casing (Stratton–Chu) fields and expresses the resulting magnetic field as the analytic curl of a regular-solid-harmonic vector potential. This construction yields a divergence-free discrete field at every expansion order, provides analytic derivatives, and enables an a priori separation of truncation, quadrature, and geometric errors. On the rotating-ellipse benchmark of Malhotra et al. [1], QBX maintains a relative error of 2–5 x 10^{-6} throughout the near-surface regime at distance-independent cost, whereas upsampled trapezoidal quadrature exhibits O(1) errors and requires node counts scaling as (h/d)^2. AMPERe is implemented as a header-only, performance-portable Kokkos library supporting CPU and GPU backends, with target-specific and local-correction accelerations validated against the baseline. The method is further demonstrated on realistic Wendelstein 7-X configurations, including sheet-current and virtual-casing field evaluations.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.41544 ↗

Although nonconservative forces and extended phase space are individually common topics in Hamiltonian dynamics, their combination is seldom developed explicitly. This note derives the full Hamiltonian equations of motion for nonconservative and nonautonomous dynamical systems formulated on extended phase space. Particular attention is given to the momentum coordinate and generalized force associated with time. The classic analytical dynamics framework is used throughout, with all developments formulated in terms of arbitrary symplectic coordinates (i.e., canonical coordinates) or functions thereof.

fields
Mathematical Physics, Symplectic Geometry
pdf ↗ page arXiv:2610.41635 ↗

These lecture notes originate from a one-semester course on the mathematical foundations of second quantization, first given at the Department of Physics of the Friedrich-Alexander-Universität Erlangen-Nürnberg in the Winter Semester 2025/26 to a mixed audience of physics and mathematics students. Their guiding thread is a class of physical models: spin-boson models, describing a two-level system coupled to a boson field. After summarizing the necessary background on unbounded operators, up to the Kato-Rellich and KLMN theorems, and developing direct sums and tensor products of Hilbert spaces and of unbounded operators, we treat single-mode boson fields, constructing the ladder and number operators and characterizing their domains, before moving to Fock spaces, second quantization, bosonic and fermionic annihilation and creation operators, and the canonical (anti)commutation relations. Throughout, spin-boson models provide both the motivation and the test bench for the formalism: we rigorously establish their self-adjointness, solve the pure dephasing and Jaynes-Cummings models explicitly, deriving their spectra and the qubit channels they induce, show how their multimode counterparts display irreversible decoherence and decay, and, in an appendix, prove the rotating-wave approximation for the quantum Rabi model with an explicit error bound. Finally, we use the KLMN theorem to construct spin-boson models with non-normalizable form factors, and briefly discuss stronger ultraviolet divergences and the need for renormalization. Exercises with full solutions are included.

comments
131 pages. Lecture notes with exercises
fields
Mathematical Physics
pdf ↗ page arXiv:2610.41657 ↗

We review rigorous results concerning the Area Law for the entanglement entropy of free fermions whose one-body Hamiltonian is an ergodic Schrödinger operator with a strongly localised pure point spectrum. Particular attention is paid to operators with deterministic potentials (quasi-periodic ones and those generated by subshifts of finite type). Our approach relies on reducing the problem to obtaining exponential estimates for the eigenfunction correlator. This provides an effective route from localisation to the Area Law and allows for the identification of additional spectral estimates that need to be proven for various classes of ergodic operators. We realise this approach by using various localisation techniques, including dynamical localisation and the uniformly localised eigenfunction property for bounded and unbounded ergodic operators (e.g. the Maryland model), as well as a new method for obtaining the corresponding exponential bounds for Schrödinger operators generated by subshifts of finite type.

comments
25 pages
fields
Mathematical Physics, Spectral Theory
pdf ↗ page arXiv:2610.41845 ↗

We construct a finite-range, translation-invariant counterexample to the Weak Gibbsianity property for quantum spin systems on $\mathbb{Z}^2$. The proof relies on showing how a local potential can induce macroscopically large deviations from an equilibrium configuration after a perturbation is introduced at the boundary of the lattice.

comments
11 pages, 2 figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.41871 ↗

Anomalies of global symmetries can arise in long-range entangled quantum many-body systems and provide useful constraints on their dynamics. In this paper, we study anomalies of $d$-dimensional topological quantum field theories (TQFTs) coupled to parameter spaces and their implications for ground-state degeneracies. Specifically, we formulate anomalies as obstructions to constructing parametrized families of TQFTs. We discuss how the classification of these anomalies depends on the choice of topology on the universal $\mathrm{U}(1)$ $(d-1)$-form symmetry. In this framework, we use the projective ground-state bundle obtained by compactifying the TQFT on a spatial manifold to derive a general constraint relating its ground-state degeneracy $N$ to its torsion anomaly $ω$, expressed as $N\int_Wω\in\mathbb{Z}$ for bundles $W$ of spatial manifolds over closed surfaces. We apply this constraint to ground-state degeneracies of (2+1)-dimensional TQFTs with framing and 1-form symmetry anomalies and to obstructions to symmetry-preserving TQFTs. We also discuss implications for bordism generators and propose a conjecture on anomaly detection by manifolds fibered over surfaces.

comments
41 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2610.42007 ↗

We determine exactly which gap labels are realized by open spectral gaps for the unitary almost Mathieu operator at every irrational frequency and all coupling constants ($0<λ_1,λ_2\leq1$). When both couplings are strictly less than one, every permitted gap label is realized by an open gap, including in the critical case ($λ_1=λ_2$). When at least one coupling equals one, both lifts of every odd winding label are realized by open gaps, whereas all even winding labels are collapsed. The proof combines quantitative almost reducibility and Aubry duality with a monotonicity principle for gap heights.

comments
29 + 11 pages, no figures
fields
Mathematical Physics, Spectral Theory
pdf ↗ page arXiv:2610.42120 ↗

We investigate the dynamics of quantum systems whose evolution is governed by linear generators of fixed unitarily invariant norm. For this broad class of evolutions, we derive a universal time-optimal generator of the dynamics converting one given state into another. It provides a time-optimal non-Hermitian Hamiltonian for pure state conversion. We then use this general result to derive quantum speed limits for mixed-state evolutions governed by Lindbladians and non-Hermitian Hamiltonians. Our findings follow from the geometric properties of the group $GL(n,\mathbb{C})$, in particular the geodesic distance. The derived bounds hold for an arbitrary system size $n$, depend only on the norms of the evolution generator and the initial and final states, and do not require performing any time-averaging.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.42131 ↗

As observed by Kienzle and the second author, rank-two $\mathcal{O}(3)$-twisted Higgs bundles on the projective line encode all hyperbolic crystals of genus two: the spectral curves are the unit cells, and the Hitchin fibers are the abelian Brillouin zones. In a uniform magnetic field of two flux quanta, the lowest Landau level is one-dimensional at every nonzero momentum, and, as found by Iengo and Li, its dimension jumps to two at the zero momentum. We identify the states of the lowest Landau level with the Baker–Akhiezer sections of the Higgs bundles, and their zeros, the vortices, with the Baker–Akhiezer divisors. A family of nonzero momenta limits to the zero momentum exactly when each of the two vortices approaches the image of the other under the spatial inversion, and the states of the jumped level are labeled by the directions of limiting. On the Higgs bundle side, the zero momentum is the point where the Hitchin fiber intersects the Hitchin section. Such a family of Higgs bundles limits to the Hitchin section, thereby crossing to a lower stratum of the \BB stratification, an example of the phenomenon investigated in terms of Baker–Akhiezer divisors by the first author with Figiel and Teschner.

comments
First version, 26 pages. Comments welcome!
fields
Mathematical Physics, Algebraic Geometry
pdf ↗ page arXiv:2610.42134 ↗

Number Theory22

Let $f$ be a Rademacher random multiplicative function supported on the squarefree integers. We investigate the almost sure fluctuations of the restricted sum $M^{>}(x)=\sum_{n\le x;~ P(n)>x^{1/2}}f(n)$, where $P(n)$ is the largest prime factor of $n$. With $Q(x)$ being the conditional variance of $M^{>}(x)$ given the values of $f$ at primes not exceeding $x^{1/2}$, we prove that, almost surely, the upper and lower limits of $M^{>}(x)(2Q(x)\log\log Q(x))^{-1/2}$ as $x\to\infty$ are $1$ and $-1$, respectively. Furthermore, it is shown that the almost sure cluster set of the normalized sum is $[-1,1]$.

fields
Number Theory, Probability
pdf ↗ page arXiv:2610.40707 ↗

We study the integral Hasse principle for the affine surfaces given by $x^4+y^4-w^2=m^2$, where $m>1$ is odd and squarefree. We give a criterion to determine if there is an integral Brauer–Manin obstruction, which depends only on simple congruence conditions on the prime factors of $m$ modulo $8$. We use this to show that for $100\%$ of such surfaces there is no integral Brauer–Manin obstruction to the Hasse principle for integral points.

comments
15 pages, comments welcome!
fields
Number Theory, Algebraic Geometry
pdf ↗ page arXiv:2610.40815 ↗

Let $E/F$ be an unramified quadratic extension of non-archimedean local fields of characteristic zero and odd residual characteristic. Offen proved that, on the full spherical space of nondegenerate Hermitian matrices, equality of the Hironaka and Satake transforms implies Jacquet–Ye matching. We give a purely local proof of this theorem. The argument combines Jacquet's compatibility of matching with Fourier transform, the Godement–Jacquet functional equation, and Hironaka's scalar functional equation for Hermitian zeta integrals. The local proof recovers Jacquet's spherical-Hecke fundamental lemma from the unit case.

fields
Number Theory
pdf ↗ page arXiv:2610.40819 ↗

We examine modular inversion of strict Grid certificates in a ceiling continued-fraction construction for the Erdős–Straus equation. The six Mordell classes modulo $840$ form four inversion orbits. Prime-input pattern frequencies in the pairs $121 \leftrightarrow361 $ and $289 \leftrightarrow 529$ have equal asymptotic limits for every fixed cutoff.

comments
8 pages
fields
Number Theory
pdf ↗ page arXiv:2610.41027 ↗

We revisit the generalized Rayleigh quotient optimization problem related to the Maynard-Tao sieve with the goal of finding a more efficient function basis. We demonstrate that a family based on products of reciprocals, previously used in asymptotic estimates for $H_m = \liminf_{n \to \infty}(p_{n+m}-p_n)$, provides results that are close to the theoretical limits of the sieve across many regimes. In particular, we establish $H_2 \leq 31 \, 140$, $H_3 \leq 2 \, 560 \, 022$, $H_4 \leq 193 \, 672 \, 056$, and $H_5 \leq 13 \, 659 \, 045 \, 840$. We also show that, as $k \to \infty$, the related variational quantity $M_k$, introduced by Maynard, satisfies the lower bound $M_k \geq \log k - 0.273572 + o(1)$.

comments
17 pages
fields
Number Theory
pdf ↗ page arXiv:2610.41113 ↗

We study Hilbert spaces constructed from the Weil quadratic form associated with the zeta function of a curve over a finite field, together with self-adjoint extensions of the differentiation operator on these spaces. For a distinguished family of these extensions, we prove that the eigenvalues converge to the ordinates of the zeros of the zeta function and that the corresponding entire spectral functions, after normalization, converge locally uniformly to an infinite product determined by these zeros. These limits occur as the interval parameter approaches a finite endpoint, rather than infinity as in the setting of the Riemann zeta function. We also realize the global Hilbert space through the Rosati trace pairing and show that translation by $\log q$ is unitarily equivalent to multiplication by normalized Frobenius. This provides a geometric model for the analogous analytic constructions associated with the Riemann zeta function.

comments
31 pages
fields
Number Theory
pdf ↗ page arXiv:2610.41115 ↗

Let $A_ρ$ be the coordinate ring of the projective line over $\Fq$ with a closed point of degree $N$ removed. We give finite formulas for higher Weil operators in a basis of Riemann–Roch spaces, which recover the formulas of Hu–Ou–Yang when $N=1$. We identifythese operators with representatives of the residue Casimir tensors and obtain a determinant formula for the Weil pairing, relative to a chosen determinant realization. For a normalized rank-one Drinfeld module, we compute the remainder of its Anderson generating function modulo any nonconstant $p\in A_ρ$. The formula includes multiple zeros and a zero at the ordinary point at infinity. Its last coefficient is a nonzero scalar multiple of an exponential division value. When $(p)$ is a finite principal prime, a value of the associated special function at one zero of $p$ generates the torsion field after adjoining that zero. We give an explicit $q$-linear formula recovering a torsion generator from this value. We also specialize the evaluation theorem of Gazda–Maurischat and compute the different as the principal ideal $(\partial_xp)$.

comments
38 pages
fields
Number Theory
pdf ↗ page arXiv:2610.41131 ↗

Let Q be a real indefinite quadratic form of type (1, 4) with determinant D \neq 0. We prove that for every c \in \mathbb{R}^5 there exists x \in \mathbb{Z}^5 such that 0 < Q(x + c) \le (8|D|)^{1/5}, and that the constant 8 is optimal. This establishes a conjecture of Bambah, Dumir and Hans-Gill and, together with earlier work, completes the determination of the sharp constants Γ_{r,s} for positive inhomogeneous values of nondegenerate indefinite quadratic forms. The proof combines classical reduction and gap theorems with uniform bounds for circular gaps of quadratic residues. A central step identifies the exceptional parameters in certain resonant reduced families: an explicit ladder of integer witnesses resolves their punctured neighbourhoods, leaving only three rational parameter points, subject to stated coefficient conditions. These are treated by retaining the coupling in a preceding, less reduced form. The remaining covering statements are verified by finite certificates in exact rational arithmetic.

fields
Number Theory
pdf ↗ page arXiv:2610.41223 ↗

We introduce a version of the Katok-Sarnak formula linking twisted traces of scalar valued Maass cusp forms for $Γ_0(N)$ with $N$ square-free to Fourier coefficients of vector valued Maass cusp forms for the metaplectic group $\operatorname{Mp}_2(\mathbb{Z})$ and the Weil representation associated to a lattice of signature $(2,1)$. We prove this by computing the Fourier coefficients of the vector valued Siegel and Millson theta lift of a scalar valued newform by two different methods and comparing the results.

fields
Number Theory
pdf ↗ page arXiv:2610.41298 ↗

We establish nontrivial bounds for Kloosterman sums over $p$-adic fields associated with the longest Weyl element on arbitrary split reductive groups, at arbitrary level. As an application, we obtain the regularity of Bessel distributions (of Whittaker models) for all such groups.

fields
Number Theory, Representation Theory
pdf ↗ page arXiv:2610.41332 ↗

We prove an asymptotic formula for the number of primes $p\leq X$ of the form $n^2+h$ for almost all $h \leq X^{1/2-o(1)}$. We prove that, for almost all $h \leq X^{2/5}$, the number of primes satisfies a correct-order lower bound. We also show cancellation of the Möbius function over $n^2+h$ for almost all $h < X^{1/2-o(1)}$.

fields
Number Theory
pdf ↗ page arXiv:2610.41356 ↗

We give an asymptotic formula for the twisted second moment of a fixed Dirichlet $L$-function in the $t$-aspect. This is a generalization of Selberg's asymptotic formula for the twisted second moment of the Riemann zeta function. As an application, we observe the change and the stability of the weighted one-level density of zeros of a fixed Dirichlet $L$-function.

fields
Number Theory
pdf ↗ page arXiv:2610.41394 ↗

Let $χ$ be a non-trivial real Dirichlet character and $s$ be a sufficiently large integer. In this paper, we prove that among the first $s$ non-critical special values of the Dirichlet $L$-function associated with $χ$ at least $c\sqrt{\frac{s}{\log s}}$ are irrational, where $c$ is a constant independent of $s$. This is a substantial improvement on work of Fischler when $4$ divides the conductor of $χ$. Otherwise, our constant improves on work of Mistiaen. The proof adapts some techniques of Lai and Yu for odd zeta values to the case of $L$-functions. The lack of positivity arguments is overcome by applying an irrationality criterion by Lai and Sprang originating from the $p$-adic setting.

comments
31 pages. Comments welcome!
fields
Number Theory
pdf ↗ page arXiv:2610.41458 ↗

Motivated by conjectures of Schlosser and Zhou on sign patterns of Fourier coefficients of infinite products, we study two families of theta quotients and their positive integral powers. We determine asymptotic formulas for their Fourier coefficients and use them to obtain explicit eventual periodic sign patterns. Our results provide new instances of the sign phenomena observed by Schlosser and Zhou and illustrate how dominant cusps determine the resulting sign patterns. These results were formalized and verified in Lean by AxiomProver.

fields
Number Theory
pdf ↗ page arXiv:2610.41569 ↗

We study torsors under tori and groups of multiplicative type together with the obstruction they define to the existence of rational points in the setting of geometric global fields. Our methods give two new arithmetic applications: the iterated toric descent obstruction coincides with the toric decent obstruction and the toric obstruction is the only obstruction to the existence of rational points for homogeneous spaces of connected linear groups with either connected or abelian geometric stabilizers. These applications generalize some known results for classical global fields, due to Borovoi, Colliot-Thélène, and Sansuc, while exhibiting novel behaviors.

comments
31 pages, Comments are welcome :)
fields
Number Theory, Algebraic Geometry
pdf ↗ page arXiv:2610.41573 ↗

We prove that the shifts $ζ(s+id_jτ)$, $1\le j\le n$, are jointly universal for any nonzero real parameters $d_j$ with pairwise distinct absolute values. These restrictions are best possible. As an immediate consequence, we obtain generalized strong recurrence for arbitrary finite families of nonzero real parameters. Our proof uses Emsalem's generalization of the six exponentials theorem and a new variant of Pecherskii's rearrangement theorem.

fields
Number Theory
pdf ↗ page arXiv:2610.41600 ↗

Special Fibonacci and Lucas type expansions proved useful to ``correct'' some theorems on the number of expansions in non-integer bases \cite{BaiKomLor2021, KomLorPed2024, KomLorPed2025}. The proof of a key lemma was non-constructive. It was made constructive in the Fibonacci case by giving a precise estimate of the growth of the partial sums of the series of the inverses of the Fibonacci numbers \cite{KomPed2025}. In this paper we extend this result to the general case.

comments
18 pages
fields
Number Theory
pdf ↗ page arXiv:2610.41658 ↗

We use a new version of the Taylor–Wiles method, based on a tame analogue of the techniques introduced by Skinner–Wiles, to prove new modularity theorems for residually reducible Galois representations of rank 2 over totally real fields.

comments
76 pages
fields
Number Theory
pdf ↗ page arXiv:2610.41764 ↗

Let $p$ be a prime number and let $K$ be a field such that $p$ is prime to the absolute Galois group $\operatorname{Gal}(K)$, in the sense that $p$ does not divide the degree of any finite Galois extension of $K$. We conjecture that $K$ is large in the sense of Pop: any smooth algebraic curve $C$ over $K$ with a $K$-rational point has infinitely many $K$-rational points. This conjecture would have some consequences in mathematical logic. For instance, it would imply the expected classification of stable fields of positive characteristic. We give some weak evidence for this conjecture. Specifically, suppose that $K$ is a perfect field and $\operatorname{Gal}(K)$ is prime to $p$. Then every abelian variety $A$ over $K$ has Zariski dense $K$-rational points, and if $C$ is a smooth curve of genus less than $7-p$ with at least one $K$-rational point, then $C$ has infinitely many $K$-rational points.

comments
15 pages
fields
Number Theory, Logic
pdf ↗ page arXiv:2610.41843 ↗

We study series $L(x,s):=\sum_n ρ(n)e^{2πi n x}/n^s$, where $ρ(n)$ are Fourier coefficients of an automorphic form $φ$ for a discrete cofinite subgroup $Γ\subset \mathrm{SL}(2,\mathbb{R})$ and $x\in\mathbb{R}$. Let $\mathfrak{b}\in\mathbb{R}\cup\{\infty\}$ represent a cusp of $Γ$. Ordering the orbit $Γ\mathfrak{b}$ by increasing denominators, we prove that the values $L(x,1/2)$ for $x\inΓ\mathfrak{b}$, suitably normalized, become equidistributed according to a standard complex Gaussian law, under natural assumptions on the Laplace eigenvalue of $φ$. Previously this was known either when $Γ=\mathrm{SL}(2,\mathbb{Z})$ or when $φ$ is associated to a holomorphic cusp form. This applies in particular to central values at $s=1/2$ of additive twists of Maass form $L$-functions $\sum_{n\geq1}λ(n)e^{2πi n x}/n^s$, or to the Dirichlet-theta series $\sum_{n\geq1}χ(n)e^{2πi n^2 x}/n^s$.

comments
58 pages, 5 figures
fields
Number Theory
pdf ↗ page arXiv:2610.42008 ↗

We establish a spectral Beurling-Selberg approximation theorem on compact connected Riemannian manifolds for sets whose boundaries have finite upper Minkowski content. Our majorants and minorants have controlled spectral coefficients and explicit stability under Hausdorff perturbations, uniformly over families satisfying a common Minkowski content bound. We extend the construction to singular spaces arising from Riemannian submetries with basic mean curvature, and apply it to equidistribution on quaternionic Shimura varieties. Assuming the conjectural formalism of the specified motivic L-functions, including the motivic Riemann hypothesis, we develop effective Chebotarev-Sato-Tate estimates for fixed sets and varying families with controlled boundary geometry and Hausdorff variation, and describe how additional symmetries improve the error exponents. The proofs combine classical analytic methods with new methods from differential geometry, analytic geometry, and Riemannian spectral analysis. Applications include an asymptotic formula for a motivic analogue of Atkin-Serre estimates and tubular estimates for Picard-rank jumps of K3 surfaces via their transcendental motives.

fields
Number Theory, Algebraic Geometry, Differential Geometry, Representation Theory
pdf ↗ page arXiv:2610.42022 ↗

We study the Iwasawa $μ$- and $λ$-invariants of quadratic twists of rational elliptic curves at Eisenstein primes $p\geq 3$. We verify Greenberg's $μ=0$ conjecture for all such twists, which gives an Euler system free proof of Mazur's cotorsion conjecture, and study the $λ$-invariants when the twist is admissible in the sense of [GV00]. As a consequence, we prove the cyclotomic Iwasawa Main Conjecture and obtain the $p$-part BSD formulae in rank $0$ for admissible twists under mild conditions, covering many elliptic curves of potentially ordinary reduction.

comments
29 pages. Comments welcome!
fields
Number Theory
pdf ↗ page arXiv:2610.42137 ↗

Numerical Analysis22

We prove the Kung–Traub conjecture: a deterministic stationary root-finding method without memory, using at most $n$ real scalar function or derivative evaluations per iteration, cannot have local order greater than $2^{n-1}$ on every real analytic function with a simple zero. For $n\geq2$, the bound is attained by the methods of Kung and Traub. The local estimate is required at every sufficiently close starting point, with constants that may depend on the function. Evaluation points and derivative orders may be chosen adaptively, and the rules for choosing them and processing the data are unrestricted. The proof uses a localisation property of polynomial Wronskians to obtain a lower bound at each fixed scale. A diagonal construction then gives one entire function, with derivative bounded above and below on the real line, for which every larger order fails. We also obtain a sharper bound when evaluations are made in prescribed groups, and prove the corresponding result for complex scalar evaluations. The results of this paper are also formally verified in Lean.

code
github.com/MColbrook/Kung-Traub_conjecture ∎ Lean project (lakefile.toml, lean-toolchain) builds, no sorry at 218c345, checked 2026-11-02 build details
fields
Numerical Analysis, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.40660 ↗

We present for the first time an approach for identifying the geometric and material characteristics of layered composite structures through an inverse wave and finite element approach. More specifically, this Non-Destructive Evaluation (NDE) approach is able to recover the thickness, density, as well as all independent mechanical characteristics such as the tensile and shear moduli for each layer of the composite structure under investigation. This is achieved through multi-frequency single shot measurements. It is emphasized that the success of the approach is independent of the employed excitation frequency regime, meaning that both structural dynamics and ultrasound frequency spectra can be employed. It is demonstrated that more efficient convergence of the identification process is attained closer to the bending-to-shear transition range of the layered structure. Since a full FE description is employed for the periodic composite, the presented approach is able to account for structures of arbitrary complexity. The procedure is applied to a sandwich panel with composite facesheets and results are compared with two wave-based characterization techniques: the Inhomogeneous Wave Correlation method and the Transition Frequency Characterization method. Numerical simulations and experimental results are presented to verify the robustness of the proposed method.

journal
Composite Structures, 2017, 182, pp.566-578
fields
Numerical Analysis
pdf ↗ page arXiv:2610.40689 ↗

Numerical approximations of transport problems, even when stabilised, often violate the bounds respected by the exact solution. Depending on the problem being solved, this can produce non-physical results, such as negative densities or higher-than-unity concentrations. Various bound-preserving numerical methods exist, yet they are usually either conditionally stable or only first-order accurate. In this context, the present work extends the nodally bound-preserving finite element method to the linear transport equation. As a baseline discretisation we use the least-squares finite element method, which is instrumental in achieving unconditional stability for second-order time stepping. It also leads to symmetric algebraic systems, thereby enabling simple, efficient solvers, which we further enhance through Anderson acceleration. Unconditional stability is proved for the backward Euler and Crank–Nicolson discretisations. We also derive error estimates for the fully discrete second-order scheme and provide various numerical examples, including two-phase incompressible flows.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.40841 ↗

The bottleneck of attention mechanisms, which are the fundamental building block of modern language models, is their quadratic complexity in terms of context size. There has thus been an increasing interest in the compression of the context, also called KV-cache compression. In this article, we consider the problem of compressing the context of attention mechanisms by first formulating it in the Hilbert space associated with its kernel. In this formalism, attention can be represented as the ratio of two linear maps whose interpolant is also the ratio of interpolants, this allows us to consider the interpolation of the linear maps directly. We thus introduce two novel KV-cache compression methods which we call GAIM and Q-DAIM, based on the Generalized Empirical Interpolation Method (GEIM) and the Q-Discrete Empirical Interpolation Method (Q-DEIM), respectively. As the cost of these methods is quadratic in terms of context length, we propose surrogate models based on the Taylor series, the Performer and the Random Fourier Features method which allow the compression to be done in linear time. Finally, we extend this result to the streaming attention setting, providing new methods which allow one to penalize the selection of interpolation nodes based on the previously interpolated context and to update the previous interpolation based on new information. We then provide numerical results on the convergence of the residuals and the output error in future queries and preliminary results on the LongBench-E benchmark showing performance comparable to the state-of-the-art methods. These experiments suggest that the range revealed by our Q-DAIM method could help identify where information is concentrated in the KV cache, thus enabling an efficient compression.

comments
44 pages, 17 figures, 8 tables
fields
Numerical Analysis
pdf ↗ page arXiv:2610.40943 ↗

Latent dynamics learning has emerged as a prominent approach for Model Order Reduction (MOR) of time-dependent, parameterized Partial Differential Equations (PDEs). Traditional autoencoder-based approaches are intrinsically tied to fixed spatial discretizations during both training and inference, motivating the development of meshless methodologies. In this context, conditioning provides a natural mechanism for representing temporal evolution through spatially continuous functions, enabling mesh-agnostic Latent Dynamics Models (LDMs). In this work, we introduce a unified mathematical framework for mesh-agnostic LDMs based on conditioning in Neural Fields (NeFs). Leveraging a taxonomy of conditioning mechanisms (weak, strong, and pure), we formulate meshless LDMs as NeFs conditioned on a latent state. We establish a Universal Approximation Theorem (UAT) for this class of models and analyze their stability by deriving Lipschitz bounds for different conditioning mechanisms. Our analysis reveals an expressivity–stability trade-off, providing insight into why more expressive conditioning, despite its higher representational capacity, can potentially degrade performance during temporal extrapolation. Finally, we provide a comprehensive comparison of meshless and mesh-based architectures in the continuous limit of the spatial discretization, showing that meshless models guarantee grid-independent Lipschitz bounds, unlike mesh-based counterparts. Our experiments across Burgers' equation, Navier–Stokes flow past a cylinder, and linear advection show that meshless LDMs can achieve predictive accuracy comparable to or better than their mesh-based counterparts while using only 2% of the trainable parameters and exhibiting nearly constant error across spatial discretizations.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41043 ↗

We develop a novel hybrid Schwarz method, termed the robust edge multiscale space based hybrid Schwarz method (REMs-HS), for the efficient solution of high-frequency Helmholtz problems. The method combines a one-level restricted additive Schwarz preconditioner with impedance transmission conditions (RAS-imp) and an edge multiscale (EMs) coarse solver in a multiplicative way. A key ingredient of our construction is a modified impedance boundary condition (mIBC), which is imposed both on the subdomain problems and on the local problems that generate the EMs coarse space. Compared with the original impedance boundary condition, the mIBC makes the proposed method both stable and inexpensive theoretically. We prove that the resulting REMs-HS preconditioner gives rise to a uniform convergent fixed-point iteration for large wavenumbers. Moreover, the required resolution conditions are substantially relaxed. Only a mild condition on the level parameter $\ell$ of the hierarchical coarse basis $\ell>\frac{1}{2}\log_2 k$ is needed, which is markedly weaker than the conditions in previous analyses. Extensive numerical experiments demonstrate the robustness of the proposed method for Helmholtz problems with large wavenumbers.

comments
21 pages, 2 tables
fields
Numerical Analysis
pdf ↗ page arXiv:2610.41095 ↗

In this paper, we study a general frequency principle under gradient descent training for the finite linear approximation classes. Under suitable assumptions, we prove that the learner amplitude relative to the target amplitude vanishes at high frequencies, uniformly throughout training. We also establish sufficient conditions for low-frequency priority, covering single-frequency targets trained separately and two frequency components trained in the same run. This general frequency principle does not imply that lower frequencies always converge faster. Instead, for two frequencies that satisfy certain comparison conditions, one component will take longer to learn than the other; and in common cases, the longer component has higher frequency. Numerical experiments with Legendre and trigonometric bases validate the theoretical results.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41179 ↗

We establish sharp convergence results for a sequential semi-implicit good-vorticity integrator for the two-dimensional inviscid Oldroyd–B system with stress diffusion on the torus. For exact solutions satisfying $u\in C_tB^{1+α}_{\infty,1}$ and $τ\in C_tB^α_{\infty,1}\cap L^1_tB^{2+α}_{\infty,1}$, $0\leα\le2$, we prove convergence in the critical norms $B^1_{\infty,1}\times B^0_{\infty,1}$, uniformly over the time nodes on every finite interval where these regularity assumptions hold, including the critical endpoint $α=0$. For $0<α\le2$, we obtain the rate $O(h^{α/2})$ without any additional time Hölder assumption, reaching first order at $α=2$. These algebraic exponents are optimal for this Eulerian scheme over the stated solution classes, as shown by fixed-data lower bounds in the case $a=b=0$ with nonzero velocity mean. For a fixed shear solution at critical regularity, we establish matching logarithmic upper and lower bounds on the numerical error, ruling out every positive algebraic convergence rate. We also prove that the gain of one stress derivative at the time nodes is optimal when the velocity is controlled in $B^1_{\infty,1}$, even for smooth initial stress.

comments
37 pages, 2 figures
fields
Numerical Analysis, Analysis of PDEs
pdf ↗ page arXiv:2610.41250 ↗

We present a geometric interface reconstruction method for Volume-of-Fluid (VOF) simulations that captures both curved interfaces and sharp features using only volume fraction data. We represent interfaces with linear facets, circular arcs, and corner facets formed by two lines or arcs joined within a single cell. We first propagate local orientation constraints through a graph and merge cells to resolve ambiguities, treating topological consistency as a first-class constraint. An optional conservative refinement improves $C^0$ continuity while preserving local volume constraints. Users can independently enable circular facets, corner facets, cell merging, and continuity refinement to balance accuracy and computational cost. In its simplest configuration, our method provides a fast local line-fitting scheme competitive with state-of-the-art PLIC methods. On standard static benchmarks, the full configuration reconstructs identifiable lines, circles, and sufficiently resolved corners to numerical precision on coarse, perturbed Cartesian meshes. On varying-curvature interfaces, it achieves third-order convergence in facet-gap error and approximately first-order convergence in curvature error. These results demonstrate global volume conservation and accurate reconstruction of curved and corner features.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41305 ↗

We propose a high-order well-balanced path-conservative discontinuous Galerkin method for one-dimensional balance laws with non-conservative products. Generalized hydrostatic reconstruction at cell interfaces couples effective numerical fluxes with Dal Maso-LeFloch-Murat path integrals, while a cellwise reference-equilibrium decomposition cancels projection-induced residuals in the volume discretization. Under explicit assumptions on equilibrium recovery, admissibility, paths, and boundary data, the spatial discretization is path conservative, and prescribed rest and moving equilibria are exactly preserved by both the semi-discrete scheme and its fully discrete SSP-RK approximation. The framework is applied to the two-layer shallow water equations and a blood-flow model with spatially varying vessel properties. Numerical experiments confirm the designed order of accuracy and round-off-level preservation of smooth and discontinuous equilibria, demonstrate accurate resolution of small perturbations, and show that the interface treatment captures shocks and stationary material contacts without visible spurious oscillations.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41484 ↗

Stochastic differential equations (SDEs) driven by subdiffusion see great applications in anomalous transport and asset price dynamics. Trusted numerical methods are essential for their application in practice. In this paper, for SDEs driven by subdiffusion with superlinear coefficients, we propose a brand-new explicit numerical method. The sharp convergence rate is proved with new techniques. The potential application of the proposed method in numerically solving a generalized Fokker-Planck equation is also illustrated. Numerical simulations are provided to demonstrate the theoretical results.

comments
23 pages, 2 figures, 3 tables
fields
Numerical Analysis
pdf ↗ page arXiv:2610.41550 ↗

We introduce randomized algorithms for computing the hierarchical Tucker decomposition of high-order tensors. Unlike traditional algorithms performed in a sequential leaves-to-root direction, our algorithms traverse the dimension tree in a root-to-leaves direction, leveraging randomized sequential projections to accelerate the computation of intermediate subspaces. This strategy can reduce the computation time by an order of magnitude relative to prior randomized algorithms using the leaves-to-root direction, while guaranteeing quasi-optimal approximation. Although the theoretical error bound is slightly larger than for leaves-to-root methods, in all our experiments the practical error adheres to the classic bound. Our results highlight the method's potential for scalable tensor decompositions in high-dimensional data analysis.

comments
28 pages, 8 figures, 3 tables, presented at GAMM ANLA 2025 and METT XI
fields
Numerical Analysis
pdf ↗ page arXiv:2610.41554 ↗

We develop two sampling algorithms for constructing extensible designs with mesh ratio at most two on finite unions of full-dimensional rational convex polytopes. Starting from a rational initial point, both algorithms produce rational coordinates and perform the geometric comparisons governing point selection exactly. This ensures that distance ties and geometric degeneracies are handled without rounding ambiguity, preserving the prescribed selection rules and mesh-ratio guarantee at every stage. The first algorithm, Local Exact Farthest-Point Sampling (LEFPS), implements a prescribed farthest-point rule by local updates of the restricted Voronoi diagram. The second algorithm, Arithmetic-Aware Admissible Sampling (AAS), minimizes the representation length of reduced rational coordinates over admissible fragment vertices and edge midpoints. Since the cost of exact arithmetic depends on coordinate lengths, we analyze both operation counts and bit complexity. For fixed dimension and input, each sampling rule admits an $O(N\log N)$ unit-cost implementation and a separate implementation with $\widetilde O(L_N)$ bit complexity, where $L_N$ is the total length of the output coordinates. On rational interval covers, every admissible sequence from the vertex-midpoint candidate set has logarithmic coordinate height. We also identify the ordered sequences explicitly for LEFPS on interval unions and four-dimensional coordinate-order domains, and for AAS on a rectangle and a nonconvex cross. These descriptions yield direct generation with $O(N)$ unit-cost work and optimal $Θ(N\log N)$ bit work, including exact quality measures at every prefix. Numerical experiments compare local Voronoi maintenance, coordinate growth, and direct generation.

comments
37 pages, 6 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.41592 ↗

Deriving mesh-robust error bounds is an open problem for low-rank methods, including dynamical low-rank (DLR) and step-and-truncate (SAT) approaches, as highlighted in a recent review. Existing bounds can depend on Lipschitz constants that grow with inverse mesh size, making them impractical, particularly for kinetic equations, a major application of low-rank methods. We develop a unified framework for implicit SAT that provides a discrete counterpart of energy-localization analysis for nonlinear partial differential equations. By restricting numerical solutions to energy-localized regions, we obtain one-sided Lipschitz stability with constants independent of spatial resolution. The resulting error estimates and time-step restrictions are mesh robust, answering this open problem for the class of implicit SAT schemes covered by our assumptions. We emphasize that energy localization provides a general mechanism for mesh-independent robust error estimates in low-rank approximation, extending beyond the SAT methods considered here. Analogous results for DLR methods will be presented in subsequent work. Returning to implicit SAT, we demonstrate the broad applicability of the framework through error estimates for different classes of equations. Specifically, for semilinear parabolic and Vlasov–Poisson equations, we establish robust first-order error bounds for backward-Euler SAT under suitable regularity assumptions. Energy localization also has practical implications. We design an energy-constrained implicit SAT algorithm. Implicit SAT methods without energy constraints can select incorrect solution branches even when their algebraic residuals are small. Moreover, time-step reduction can be essential for continuation into larger energy regions. Numerical experiments support these conclusions.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41606 ↗

For computing the dominant eigenvalue and the corresponding eigenvector of a real and symmetric matrix, inspired by the parameterized power method and multistep iteration, we propose a two-sweep parameterized power method, and establish its convergence theory. To further improve the convergence speed of the new method, we use Gaussian process regression (GPR) to predict the quasi-optimal parameter and give a version of the proposed method based on GPR. Numerical experiments show that the two-sweep parameterized power method with quasi-optimal parameters outperforms some existing methods in terms of computing efficiency.

comments
28 pages, 3 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2610.41627 ↗

We consider here computational homogenization of a periodic two-phase conduction problem. Quantum linear-system algorithms for its finite element discretization are limited by the condition number of the stiffness matrix. This condition number grows quadratically with the number of elements per side. Classically, preconditioning by the homogeneous (single-phase) operator bounds the condition number by the conductivity contrast, for every mesh size. On a quantum computer, however, a preconditioner encoded as a separate factor multiplies the subnormalizations, and the quadratic dependency returns. In this paper, we absorb the closed-form inverse square root of the homogeneous operator into the finite element factors. The preconditioned operator then becomes the product of a norm-preserving gradient operator $\mat{V}$, a diagonal matrix of scaled conductivities, and $\mat{V}^T$. Each factor has norm at most one. The block encoding therefore has subnormalization one, and its condition number is bounded by just the conductivity contrast, for every mesh size and in every dimension. The operator $\mat{V}$ is applied as a quantum Fourier transform, a small rotation at each wavenumber, and an inverse transform. Quantum singular value transformation then block-encodes the inverse of the preconditioned operator. We test the method numerically on three periodic cells with up to $10^6$ unknowns. The results confirm that the condition number does not depend on the mesh, and the energy-norm error of the inverse is below the polynomial tolerance at every mesh. On meshes of $4 \times 4$ and $8 \times 8$ elements, we also build the entire inverse as one Qiskit circuit and simulate it by state vector, with the rotation at each wavenumber built from uniformly controlled rotations. The output state agrees with the finite element solution to within the polynomial tolerance.

fields
Numerical Analysis, Analysis of PDEs
pdf ↗ page arXiv:2610.41690 ↗

Solving systems of linear equations, $Ax=b$, is a ubiquitous task in computational science. The Harrow-Hassidim-Lloyd algorithm requires phase estimation, controlled rotations, and post-selection, impractical on noisy intermediate-scale quantum hardware. We extend sample-based quantum diagonalization (SQD), developed for quantum chemistry, to linear systems. Following the sample-based Krylov variant, the Hermitian matrix $A$ is decomposed in the Pauli basis, the Trotterized time evolution of $|b\rangle$ is sampled at several times to select computational-basis configurations, and the system is solved classically by a Galerkin projection onto the selected subspace. We prove that the subspace solution is exact when the sampled configurations contain the Krylov support of $b$ and optimal in the energy norm for symmetric positive-definite matrices, so that noise-induced configurations cannot degrade it as long as they do not displace the ideal ones. On IBM Quantum hardware, the sampled distribution of a $4\times4$ example deviates from the exact one by eleven times the shot-noise level, yet the solution is recovered exactly. For a localized problem of dimension $N=1024$, SQD selects 33 configurations, reaches a relative error of $10^{-5}$, and its truncated subspaces coincide with the optimal choice; readout errors enlarge the subspace without degrading the accuracy. The tests also show the limitations: the selected subspace coincides with that of inexpensive classical selection, banded matrices require exponentially many Pauli terms in the binary encoding, and Trotter errors populate spurious configurations unless the circuits are deep. An MIT-licensed Python package and notebooks that reproduce the numerical results are provided.

comments
13 pages, 5 figures, 6 tables. Code: Python package sqd_linsolve (MIT License) with two notebooks in the ancillary files: reproduce_paper.ipynb reproduces the numerical results of the paper; sqd_linear_systems_selfcontained.ipynb gives a self-contained, step-by-step implementation with explanations and a worked example
fields
Numerical Analysis
pdf ↗ page arXiv:2610.41708 ↗

In this paper, we analyze the standard and the mass-lumped linear finite element methods for the Gross–Pitaevskii eigenvalue problem. For the standard method, we give a new proof of the uniqueness of the discrete ground state up to sign for sufficiently small mesh sizes, which only requires the $H^1$-convergence of the discrete ground states. For both methods, we establish optimal convergence rates in the $L^p$- and $W^{1,p}$-norms for $1 \le p < \infty$. Under additional regularity assumptions on the potential, we further obtain an optimal convergence rate in the $W^{1,\infty}$-norm and an almost optimal convergence rate, up to a logarithmic factor, in the $L^\infty$-norm. Moreover, we prove that the discrete ground states superconverge to the Ritz projection of the exact ground state in the $W^{1,p}$-seminorm. The analysis is based on a perturbation argument. Numerical experiments confirm the theoretical results.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41715 ↗

This paper analyzes a symmetric interior penalty discontinuous Galerkin (IPDG) method, a hybrid mixed DG method, and two representative hybridizable discontinuous Galerkin (HDG) methods for the Poisson problem on convex polyhedral domains. For any fixed $ε\in (0,\frac{1}{2})$, we first derive optimal discrete $W^{1,p}$ error estimates for $2\leq p<\frac{1}ε$ by combining the interface Green's functions developed in \cite{lq2026} with the Fubini-type weighted estimates established in this paper. These error estimates lead to discrete $W^{1,q}$-$W^{1,p}$ inf-sup stability for $2\leq q<\frac{1}ε$, where $q$ is the conjugate number of $p$ such that $\frac{1}{p}+\frac{1}{q}=1$. Using this stability result and a suitably constructed dual problem, we extend the optimal error estimates to $\frac{1}{1-ε}<p<2$ and subsequently establish the corresponding inf-sup stability for $\frac{1}{1-ε}< q<2$. Thus, both optimal discrete $W^{1,p}$ error estimates and discrete $W^{1,q}$-$W^{1,p}$ inf-sup stability hold throughout the range $\frac{1}{1-ε}<p<\frac{1}ε$, with $q=p/(p-1)$. Note that all of the aforementioned results are free of logarithmic factors and also hold on polyhedral meshes (except for the hybrid mixed DG method). Compared with the conforming finite element method (FEM), establishing such error estimates and inf-sup stability results for DG methods on convex polyhedral domains is more delicate because their analysis involves mesh-dependent norms. Consequently, the operator interpolation theory employed for conforming FEMs cannot be directly applied to DG methods. Finally, a numerical example is presented to illustrate the theoretical results.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41747 ↗

We introduce a numerical quadrature algorithm with stretched exponential convergence to evaluate one-loop integrals in dimensional regularisation in the causal Loop Tree Dual (cLTD) representation. The key feature that makes this possible is the fact that for Euclidean kinematics all singularities of the cLTD integrand lie on a compact set in loop momentum space. We prove that our algorithm admits a stretched exponential convergence with the number of integrand evaluations, and we illustrate it by evaluating triangle, box and pentagon integrals at selected Euclidean kinematics points. We benchmark our implementation against analytic results as well as results obtained via Monte Carlo integration. We also provide a proof of concept that our algorithm can be extended to physical kinematics after subtraction of threshold singularities.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41872 ↗

We present a high-order, positivity preserving, and asymptotic preserving approximation of the gray radiative transfer equations. The time and space approximation is done using a diagonally-implicit Runge-Kutta method with stabilized, continuous finite elements, and the angular approximation uses the method of discrete ordinates. We detail a simple and robust inner/outer iteration strategy to solve these equations and prove that the proposed scheme is exactly conservative, independent of iterative tolerances. A maximum and minimum principle is enforced using a conservative redistribution limiting technique, and the method is demonstrated to be robust in the thick diffusion limit.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.41967 ↗

Risk management of worst-of autocallables needs price sensitivities (Greeks) up to third order and the potential future exposure, a tail quantile of the note's future value. In Monte Carlo, each order of differentiation multiplies the sampling noise by a factor that grows with the observation frequency, and nested simulation turns the noise of its inner paths into a bias of the quantile. We instead solve the backward pricing recursion, storing the continuation value as an $hp$ functional tensor train (hp-FTT) whose piecewise-polynomial cores are split at the barriers, so that worst-of barrier events act exactly, with rank one. For Gaussian transitions we show that the polynomial degree sufficient for an error $\varepsilon$ grows only as $\sqrt{\log(1/\varepsilon)}$. One solve then gives every Greek at any spot without sampling noise, by integrating the likelihood-ratio weight against the stored value, and the values it stores at future dates replace inner simulation. Against the strongest Monte Carlo estimators we could build or find published, on worst-of autocallables observed monthly on three to five assets, the order of the Greek and the accuracy decide which method is cheaper. At $1\%$ accuracy Monte Carlo is more than an order of magnitude cheaper for deltas, second-order Greeks go either way, and hp-FTT is cheaper for every third-order Greek, by up to two orders of magnitude. At $0.1\%$ hp-FTT is cheaper for every Greek we could check, under geometric Brownian motion and local volatility alike. For the exposure alone Monte Carlo is cheaper at $1\%$ and hp-FTT at $0.1\%$, and a solve run for the Greeks yields it for little more than scenario sampling. Sampling pays in variance for each derivative and hp-FTT in rank for each asset, so Monte Carlo keeps low-order Greeks at percent accuracy, rarely observed contracts and many assets.

fields
Numerical Analysis
pdf ↗ page arXiv:2610.42119 ↗

Algebraic Geometry21

We present a coarse classification scheme for the boundary data of massive QFTs based on transitions of boundary states via soliton / kink profiles. Starting from the lattice of boundary states, we introduce the boundary screening group $\mathbb{B}$: the equivalence classes of boundary states which cannot be connected by such transitions. In string theory backgrounds which geometrically engineer an interacting QFT in the target space, $\mathbb{B}$ computes the worldsheet analog of the target space defect group $\mathbb{D}$. When $\mathbb{B}$ is non-zero, it provides a diagnostic for having a target space with non-normalizable modes. For interfaces between QFTs, similar considerations imply the existence of a non-trivial conserved charge associated with soliton absorption / transmission, which means that the interface cannot fully dissolve / disappear. Similar considerations hold for junctions of QFTs, which in the special case of string worldsheet CFTs translate to junctions of string theories. We illustrate these general considerations with a number of non-geometric 2D $\mathcal{N} = (2,2)$ Landau-Ginzburg models, as well as their natural lifts to 4D $\mathcal{N} = 1$ Wess-Zumino models.

comments
42 pages, 4 figures
fields
Algebraic Geometry, Representation Theory
pdf ↗ page arXiv:2610.40777 ↗

We compute the Balmer spectra of geometric Tate and Artin–Tate motives with $\mathbb{F}_p$-coefficients over any field whose characteristic differs from $p$. For Tate motives, we find the bigraded comparison map to be a homeomorphism to the homogeneous spectrum of the motivic cohomology of the base field. For Artin–Tate motives, we describe the spectrum as a gluing of the spectrum of Artin motives computed by Balmer–Gallauer with the punctured Tate spectrum. We determine all specialisations between these pieces and the resulting topology. Under the hypothesis that the spectrum of rational Tate motives is a singleton, we also compute the integral Tate spectrum after inverting the exponential characteristic of the base field. As a consequence of our results, we identify the punctured Tate spectrum as an open subspace of the spectrum of the category of geometric Voevodsky motives, and draw some conclusions concerning the latter. Finally, we include an appendix where we define generalised periodicity in the context of tensor-triangular geometry, and explore its applications in motivic settings.

comments
61 pages
fields
Algebraic Geometry, Algebraic Topology, Category Theory, K-Theory and Homology
pdf ↗ page arXiv:2610.40806 ↗

Following the recent work by OpenAI on the irrationality of very general cubic fourfolds in Hassett divisors, we prove some compatibility of Iritani's blow-up formula with the classical blow-up decomposition of integral Hodge structures. As a consequence, we express OpenAI's birational invariant in terms of the theory of atoms developped by Katzarkov–Kontsevich–Pantev–Yu. We also show that, for every smooth complex rational cubic fourfold, the primitive cohomology is isometric as an integral Hodge structure to the twisted middle cohomology of a K3 surface and we prove irrationality of the very general Verra fourfold.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.40811 ↗

The goal of this paper is to develop algebraic and combinatorial techniques for studying the geometric properties of truncated Grassmannians, a class of algebraic varieties obtained via certain projections of classical Grassmannians within their Plücker embeddings. We introduce a class of cyclic representations of maximal parabolic subalgebras of the general linear Lie algebra, which are proved to serve as duals of the homogeneous components of coordinate rings of truncated Grassmannians. We prove that the characters of these representations are computed via truncated Cauchy formulas and establish a connection with coordinate rings of complete collineations. We also construct monomial bases of these representations in terms of certain generalized chain polytopes and use the RSK correspondence to establish connections with Cauchy-type formulas and with Gelfand–Tsetlin patterns.

comments
27 pages
fields
Algebraic Geometry, Combinatorics, Representation Theory
pdf ↗ page arXiv:2610.40891 ↗

Given two smooth compact Kähler threefolds $X$ and $Y$ whose universal covers are biholomorphic, we show that $κ(X) = 0$ if and only if $κ(Y) =0$.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.40950 ↗

Bryan, Oberdieck, Pandharipande and Yin predicted the number of hyperelliptic curves on general abelian surfaces under a transversality condition. Some of the numbers have already been confirmed. In this paper we construct singular hyperelliptic curves on $(1,d)$ polarised abelian surfaces for $d\leq 15$ confirming a few predicted numbers, namely of (geometric) genus $g=d$ and of geometric genus $k+3$ on $(1,2k)$ and $(1,2k+1)$ polarised surfaces. Moreover, we have found that all genus 7 hyperelliptic curves on general $(1,8)$ polarised surfaces are double coverings of curves locally embedded into $(1,4)$ polarised surfaces. More interestingly, there is a unique genus 9 hyperelliptic curve (up to translation) locally embedded into a general $(1,12)$ polarised surface and it is a Klein covering of a curve locally embedded into a $(1,3)$ polarised surface.

comments
19 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.40989 ↗

On a complex toric surface, oriented real rational curves tangent to the toric boundary at real points are distinguished by their value of quantum index. Using variants of usual Welschinger multiplicities, G. Mikhalkin defined enumerative invariants $W_{\tildeΔ}^κ$ that are provided by a signed count of oriented real rational curves with given quantum index $κ$ and fixed tangency points on the boundary of the positive quadrant (the tangency orders being encoded by $\tildeΔ$). We prove that, for most toric surfaces and tangency orders, the invariant $W_{\tildeΔ}^κ$ is zero in the case of the highest nonmaximal value of $κ$. This is done by enumerating tropical curves with Block–Göttsche multiplicities. Additionally, we prove that, in many situations, the lower bound, given by the absolute value $|W_{\tildeΔ}^κ|$ of $W_{\tildeΔ}^κ$, for the number of oriented real rational curves is not sharp close to the tropical limit in the case of the highest nonmaximal value of $κ$.

comments
24 pages, 7 figures
fields
Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.41031 ↗

In this paper, we apply a weaker version of Hodge atoms to show that a standard conic bundle over a smooth complex projective rational surface $S$ is irrational if $|2K_S+Δ|\ne\emptyset$, where $Δ$ is the discriminant divisor. This proves Shokurov's conjecture on rationality of conic bundle threefolds. As an application, we deduce precisely which nodal quartic double solids are rational.

comments
25 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.41174 ↗

Let $F=G/B$ be the complete flag variety of a connected semisimple complex linear algebraic group. We prove that if a smooth Fano variety $X$ has the same dimension and Picard number as $F$, then the existence of a continuous map $X\to F$ of odd topological degree implies $X$ is isomorphic to $F$. In particular, a smooth Fano variety homotopy equivalent to $F$ is isomorphic to $F$. The algebraic input is a numerical characterization of complete flags: a positive-dimensional smooth Fano variety $X$ is a complete flag variety if and only if its pseudoindex $ι_X$ is at least two and ${\rm ker}({\rm Sym}^2N^1(X)_{\mathbb R}\to N^2(X)_{\mathbb R})$ contains a positive-definite quadratic form. The pseudoindex condition can equivalently be replaced by the divisibility of $-K_X$ by two in ${\rm Pic}(X)$. The key observation is that a single definite quadratic relation excludes surfaces in every fiber of any elementary contraction; standard contraction theorems then yield smooth ${\mathbb P}^1$-fibrations and the characterization of complete flags applies. For the odd-degree theorem, Poincaré duality and a mod-two Wu-class argument show that the anticanonical class is divisible by two; comparison of the integral coroot lattices then identifies the flag. Most proofs were developed through extensive discussions with OpenAI's ChatGPT (GPT-5.6 Sol and GPT-6), based on mathematical directions supplied by the author and arguments suggested by ChatGPT; see the AI Disclosure.

comments
17 pages
fields
Algebraic Geometry, Algebraic Topology, Complex Variables, Differential Geometry, Representation Theory
pdf ↗ page arXiv:2610.41237 ↗

We show that if the projectivised stratum $\mathbb{P}{H}_g(μ)$ associated to a partition $μ=(m_1,\ldots,m_n)$ of $2g-2$ contains a complete curve, then \[ κ_μ:=\sum_{i=1}^n\frac{m_i(m_i+2)}{m_i+1}>12. \] The weak inequality follows from semipositivity applied to a quotient of the Hodge bundle, while equality is excluded using the second Fujita decomposition and the geometry of fibred surfaces with maximal unitary rank. Consequently, no projectivised holomorphic stratum in genus $g\leq5$ contains a positive-dimensional complete subvariety. In genus $g=6$ the same conclusion holds for \[ (10),\ (9,1),\ (8,2),\ (8,1,1),\ (7,3),\ (6,4),\ (5,5). \]

comments
7 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.41287 ↗

This paper proves that the Bridgeland stability manifold $\mathrm{Stab}(X)$ has infinitely many connected components for general quintic threefolds, other general complete intersection Calabi–Yau threefolds, certain odd-dimensional Calabi–Yau varieties, and certain rational weak Fano varieties. Moreover, we show that the geometric locus $\mathrm{Stab}^{\mathrm{geo}}(X)$ meets distinct connected components of $\mathrm{Stab}(X)$ for general quintic threefolds and their products with varieties with finite Albanese morphisms, which gives examples in any dimension at least three. For the weak Fano examples, we also construct semiorthogonal decompositions in infinitely many braid group orbits. The proof is an extension of Nordskova's argument to families, using semilinear autoequivalences and absolute Galois actions on geometric generic fibers to permute mutually orthogonal spherical objects. Then, we specialize the separation to closed fibers. In our examples, the spherical objects are constructed from rigid rational curves or spinor bundles on quadrics.

comments
48 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.41344 ↗

We construct weighted and polyhedral completions of the Weyl algebra over a complete nonarchimedean field and organize them into a structure sheaf on a polyhedral $G$-site. For discretely valued fields, we prove two-sided flatness for $Γ$-rational polyhedral completions, yielding exact base change for finitely generated $D$-modules. For compact integral polyhedra, we also construct finite-type integral models whose special fibers are toric face rings and describe the reduction of Gauss points in terms of toric strata.

comments
27 pages
fields
Algebraic Geometry, Rings and Algebras
pdf ↗ page arXiv:2610.41578 ↗

For a holonomic $D$-module, we consider the extended tropicalization of the reduced $L$-characteristic fiber along the family $L_s=F+(s-1)V_Y$. For $A$-hypergeometric systems, this tropical fiber determines the complete $(A,L)$-umbrella and hence recovers the classical slopes. We extend this result to holonomic binomial $D$-modules, for which the tropical jump parameters are exactly the classical slopes at the origin. As an application, a holonomic binomial $D$-module is regular holonomic if and only if there are no tropical jumps along any coordinate subspace.

comments
10 pages
fields
Algebraic Geometry, Commutative Algebra, Rings and Algebras
pdf ↗ page arXiv:2610.41587 ↗

We extend the notion of the right relative Du Bois complex to a smooth base of arbitrary dimension. We further show that this complex satisfies base change in the sense of Kovács–Taji, for flat families with strict higher Du Bois fibers, extending the result by Ning over a smooth curve.

comments
14 pages
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.41719 ↗

Recent work of Brantner and Taelman shows that derived deformation theory and period maps can recover strong mixed-characteristic lifting results. In this paper, we develop a prismatic version of this method for Calabi–Yau varieties over transversal or crystalline prisms, constructing compatible deformations over $A/I^n$ for every $n$ under suitable hypotheses. We also apply our method to $F$-split Calabi–Yau varieties over a field $k$ of characteristic $p>2$. In particular, we consider an $F$-split smooth proper variety $X$ of dimension $d$ over an algebraically closed field $k$ of characteristic $p>2$, with $H^{d-1}(X,O_X)=0$ and trivial canonical bundle. We show that $X$ admits a lift to $W_3(k)$.

fields
Algebraic Geometry
pdf ↗ page arXiv:2610.41727 ↗

We endow a tropical curve with the structure of a ringed space and establish analogues of Riemann-Roch and Serre duality for these ringed spaces. As a corollary, we give a cohomological proof of the tropical Riemann-Roch theorem.

comments
45 pages. Comments welcome!
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.41737 ↗

We give a tropical perspective on the recent arguments announced by OpenAI for Nagata's conjecture and the Nagata–Biran–Szemberg conjecture. By degenerating to both the normal cone and a semistable model whose dual graph is a cycle, we replace sections on a surface by Laurent polynomials in two variables. In place of theta functions, tropical methods bound the exponents of these Laurent polynomials, and a lift of the tropical Abel–Jacobi class of the normal bundle controls the shape of the resulting exponent diagram. Combining these support bounds with interpolation on a torus proves the Nagata–Biran–Szemberg conjecture with its predicted threshold whenever the corresponding linear series contains an integral curve whose only singularity is an ordinary node that is not a basepoint.

comments
55 pages, 5 figures, comments very welcome
fields
Algebraic Geometry, Commutative Algebra
pdf ↗ page arXiv:2610.41931 ↗

Recently, OpenAI announced a counterexample to the Zariski multiplicity conjecture in dimension 4. Here, we prove that the metric Zariski multiplicity conjecture is true in dimension 4. More precisely, let $f,g\colon(\mathbb{C}^n,0)\to(\mathbb{C},0)$ be germs of reduced holomorphic functions, and let $\varphi\colon(\mathbb{C}^n,0)\to(\mathbb{C}^n,0)$ be a bi-Lipschitz homeomorphism such that $\varphi(V(f))=V(g)$. We prove that if every irreducible component of the tangent cone of $V(f)$ at the origin has a singular locus of dimension at most two, then $V(f)$ and $V(g)$ have the same multiplicity at the origin. In particular, the multiplicity of complex analytic hypersurface singularities in $\mathbb{C}^n$, $n\leq 4$, is an embedded bi-Lipschitz invariant; the cases $n=2$ and $n=3$ were known. This extends a previous result of the author for tangent cones whose components have singular loci of dimension at most one. We also prove that two irreducible homogeneous hypersurfaces whose singular loci have dimension at most two, and which are mapped onto each other by a homeomorphism of $\mathbb{C}^n$ fixing the origin, have the same degree.

comments
This article firstly appeared on ResearchGate on October 25th, 2026. 26 pages
fields
Algebraic Geometry, Complex Variables, Metric Geometry
pdf ↗ page arXiv:2610.41954 ↗

We prove that birational projective irreducible holomorphic symplectic manifolds have equivalent integral cohomological motives as commutative algebra objects in $\mathrm{SH}(\mathbb C)$. Consequently, their algebraic K-theory spectra are equivalent as $E_\infty$-algebras, and their integral graded K-theory rings are isomorphic.

comments
9 pages, comments are very welcome
fields
Algebraic Geometry, Algebraic Topology
pdf ↗ page arXiv:2610.42045 ↗

We introduce Grassmannians of absolute Banach–Colmez spaces and prove the $k$-Grassmannian is perfectoid if and only if the space has slope $< 1/k$. For $d=1$, this proves a conjecture of Hansen–Mann. The key ingredient is a Plücker morphism, which realizes the $k$-Grassmannian of a slope $d/n$ space as a Zariski closed subspace of the projectivization of a slope $kd/n$ space. We also introduce a general method to construct invariant elements in perfectoid Tate–Huber pairs, and use this to prove absolute Banach–Colmez spaces of slope $<1$ have perfectoid projectivizations. As applications, we establish vanishing bounds for Scholze's mod $p$ Jacquet–Langlands functor on parabolic inductions, and also prove its vanishing in the top degree on every infinite-dimensional non-supersingular irreducible representation.

comments
comments welcome!
fields
Algebraic Geometry, Representation Theory
pdf ↗ page arXiv:2610.42110 ↗

When $X_G=\mathbb{C}^n/\!/G$ is a GIT quotient of a vector space, replacing $G$ with a maximal torus $T \subset G$ produces a closely related toric variety $X_T:=\mathbb{C}^N/\!/T$, the \emph{Abelianization} of $X_G$. Abelian/non-Abelian correspondences connect structures on $X_G$ with structures on $X_T$. In this paper, we study the interaction between the toric moment map $μ$ of $X_T$ and $X_G$, when $X_G=\mathrm{Gr}(r,n)$. We do this by looking at the restriction of $μ$ to an intermediate space $Z/T \subset X_T$, where $Z/G=X_G$. We show that the generic fibres of this map are co-Lagrangians that project to Lagrangians on $X_G$. We completely describe the image in the case where $X_G=\mathrm{Gr}(2,n)$, and show that in the general $\mathrm{Gr}(r,n)$ case the image can be sliced so that each piece is determined by a finite set of points, related to matroid circuits.

fields
Algebraic Geometry, Combinatorics, Symplectic Geometry
pdf ↗ page arXiv:2610.42123 ↗

Representation Theory12

Equivariant linear maps cannot mix non-isomorphic irreducible representations, whereas nonlinear activations can. We develop a degree-resolved version of this nonlinear interaction picture and use the resulting block inventory to construct Specht Triple, a residual on unordered graph triples. On a synthetic structural benchmark with a frozen GIN backbone, Specht Triple reduces aggregate standardised test MSE by 11.05% relative to a parameter-matched Projector Control using the same irreducible projectors and complete linear equivariant space. The results support irreducible decomposition as an inductive bias for nonlinear higher-order equivariant processing.

comments
18 pages; comments are welcome
fields
Representation Theory
pdf ↗ page arXiv:2610.41059 ↗

Let $L_n$ be the finite-dimensional irreducible $\mathfrak{sl}_2$-module of highest weight $n$. We study the space of highest-weight vectors in $L_n\otimes L_m$ for the diagonal $\mathfrak{sl}_2$-action. The extremal projector constructs these singular vectors and gives operators connecting the highest-weight lines in the Clebsch-Gordan decomposition. We interpret these operators in the Mickelsson-Zhelobenko algebra of the diagonal pair $(\mathfrak{sl}_2\oplus\mathfrak{sl}_2, \mathfrak{sl}_2^{\mathrm{diag}})$. A rescaling of their quadratic relation yields the cubic Smith relation $[E,F]=2(H^3-cH)$. This cubic algebra is also known as a Higgs algebra in the physics literature. We express both the central parameter $c$ and the Smith Casimir in terms of the quadratic $\mathfrak{sl}_2$ Casimir elements of the two tensor factors. The singular subspace becomes an irreducible Smith module of dimension $\min\{n,m\}+1$, with explicit formulas for the actions of the raising and lowering operators $E$ and $F$. Finally, comparison with finite-dimensional simple Smith modules gives a necessary and sufficient condition for a module to arise from this tensor-product construction.

comments
28 pages, 1 figure
fields
Representation Theory, Mathematical Physics, Quantum Algebra
pdf ↗ page arXiv:2610.41150 ↗

An Artin algebra is called a composition factor algebra if its indecomposable finitely generated modules are determined, up to isomorphism, by their composition factors. We prove that every composition factor algebra is representation-finite. This settles a conjecture of Auslander, Reiten and Smalø. We also give criteria for comma categories and factorization categories to be of finite representation type.

comments
25 pages
fields
Representation Theory, Category Theory, Rings and Algebras
pdf ↗ page arXiv:2610.41527 ↗

In this paper we study analytic properties of the Chevalley map $p:\mathfrak{u} \to \mathfrak{c}$ where $\mathfrak{u}$ is the Lie algebra of a unitary group over a local field of odd characteristics. This map essentially sends a matrix to its characteristic polynomial. Our main result is that smooth compactly supported measures on $\mathfrak{u}$ are pushed by the Chevalley map to measures on $\mathfrak{c}$ with density functions in $L^q(\mathfrak{c})$ for all $q < \infty$. We also provide similar results for the Chevalley map attached to the unitary group. An important consequence is an odd characteristic analogue to results of Harish-Chandra stating the local integrability of the orbital integrals on the unitary group and its Lie algebra. Further possible applications to local integrability of characters of admissible representations of unitary groups over local fields of odd characteristic are also discussed.

comments
28p
fields
Representation Theory
pdf ↗ page arXiv:2610.41630 ↗

We classify the connected components of support $τ$-tilting graphs of surface rigid algebras associated with partial tagged triangulations of punctured marked surfaces with nonempty boundary. For a connected algebra, the graph is disconnected if and only if the algebra is weakly symmetric, has at least two simple modules, and has Cartan rank one. In this case, it has exactly two connected components. For every nonzero $τ$-rigid pair over a connected surface rigid algebra, the graph of $τ$-tilting completions is connected. More generally, we classify the completion components for an arbitrary surface rigid algebra and a fixed $τ$-rigid pair. For partial tagged triangulations with all endpoints at punctures, we express the Cartan matrix and $g$-vectors by signed endpoint data. We prove that $τ$-tilting finiteness is equivalent to positive definiteness of the Cartan matrix. In the finite case, we determine the primitive $g$-vector ray generators in endpoint coordinates. Trees give type A, and a criterion on the original partial tagged triangulation distinguishes types D and C in the unicyclic case.

comments
46 pages, 7 figures
fields
Representation Theory, Combinatorics
pdf ↗ page arXiv:2610.41703 ↗

We study rational wavefront sets of irreducible representations of finite special orthogonal groups over $\mathbb F_q$. We assume that $q$ is odd and that both $q$ and its characteristic are sufficiently large. Using {the second author's} results for full orthogonal groups, we determine the maximal rational wavefront sets and the dimensions of the corresponding generalized Whittaker models. We show that each rational wavefront set is the rational closure of its maximal elements. For unipotent representations, we describe the maximal rational wavefront sets in terms of Lusztig's canonical quotient and their behavior under Alvis–Curtis duality.

fields
Representation Theory
pdf ↗ page arXiv:2610.41741 ↗

Let $G$ be $\textrm{Sp}_{2n}(F)$ or split $\textrm{SO}_{2n+1}(F)$ over a non-archimedean local field $F$ of characteristic zero. We classify the irreducible unitary representations of $G$ of purely bad parity: such a representation is unitary if and only if it is of Arthur type, except for one additional family when $G=\textrm{Sp}_{2n}(F)$. In particular, we confirm the remaining case of Ciubotaru's conjecture for $\textrm{SO}_{2n+1}(F)$, namely that every unitary unipotent representation with integral infinitesimal parameter is of Arthur type.

comments
28 pages, comments are welcome
fields
Representation Theory
pdf ↗ page arXiv:2610.41742 ↗

We prove that a finite-dimensional connected monomial algebra over an algebraically closed field is derived tame if and only if every finite connected convex subcategory of its universal cover has non-negative Euler form. Derived tameness is also equivalent to tameness of its repetitive category and trivial extension. For one-cycle algebras, this condition is equivalent to the non-negativity on $\mathbb R$ of $|Q_0|$ explicitly computable polynomials in $\mathbb Q[t]$. This gives a finite algorithm for deciding derived tameness. We also construct an infinite family of derived-tame one-cycle string algebras with minimal cubic relations, no two of which are derived equivalent.

comments
17pages
fields
Representation Theory, Rings and Algebras
pdf ↗ page arXiv:2610.41759 ↗

Jiang's conjecture predicts that if $π$ is a generic representation of a quasi-split $p$-adic group, then the $L$-parameter of the Aubert-Zelevinsky involution of $π$ must be trivial on $\mathrm{SL}_2$. In this paper, we prove this statement for a general quasi-split group under certain assumptions on the conjectural local Langlands correspondence.

comments
10 pages, comments are welcome
fields
Representation Theory
pdf ↗ page arXiv:2610.41765 ↗

I introduce the notion of a family of super unitary representations depending upon odd parameters as well as the notion of a direct Berezin integral of such representations. I use this to decompose the (left-)regular representation of the super Heisenberg group and of a super Heisenberg-like group into smaller (irreducible) subrepresentations. I compare the obtained subrepresentations to the representations obtained by applying (super) geometric quantization of the coadjoint orbits of the corresponding groups. It turns out that for the super Heisenberg group we need coadjoint orbits with an odd symplectic form in order to describe all subrepresentations found in the decomposition and for the Heisenberg-like super group we even need orbits with a non-homogeneous symplectic form.

comments
71 pages
fields
Representation Theory
pdf ↗ page arXiv:2610.41829 ↗

In the Borel case $P=B$, we endow Haines' set of maximal-dimensional positively folded alcove walks (arXiv:2405.17174) with a crystal structure. We construct a bijection with the LS-gallery model of Gaussent and Littelmann (arXiv:math/0307122) and prove that it intertwines the root operators. This answers a question raised by Haines. We also extend the Gaussent-Littelmann stratification to the parabolic setting, prove the Levi branching rule for LS galleries, and compare the resulting gallery model with Haines' maximal-dimensional positively folded alcove walks in the parabolic setting.

comments
38 pages, 5 figures
fields
Representation Theory, Algebraic Geometry, Combinatorics
pdf ↗ page arXiv:2610.41929 ↗

Over an arbitrary ordered set we consider the class of filter modules, i.e. indicator modules of filters in the ordered set. We show that the induced exact structure of filter modules, the so-called filter exact structure, which has filter modules as relative projective modules, admits an explicit projectivization by the distributive lattice of filters. This allows the reduction of the projective-homological properties of the induced exact structure to the standard homological algebra of $k{\operatorname{Fil}(Q)}$, the incidence algebra of $k$-representations on the lattice of filters. In the case of finite join-semilattices, we leverage the combinatorial structure of the distributive lattice of filters to determine the standard global dimension of $k{\operatorname{Fil}(Q)}$. By virtue of projectivization we provide lower and upper bounds for the global dimension of the filter exact structure. Moreover, we call an ordered set filter-tiny if its relative injective modules one-to-one-correspond to the standard injective modules. We exploit relative derived functors to show that filter-tiny ordered sets do not admit relative injective prehulls if the relative global dimension is larger than 1. It is open to which classes of ordered sets $Q$ the filter-tininess condition applies, and what the relative injective modules are in the case of non-filter-tiny ordered sets.

comments
33 pages, 4 figures. Comments welcome
fields
Representation Theory
pdf ↗ page arXiv:2610.42080 ↗

Differential Geometry11

The Nash Embedding theorem asserts that every closed Riemannian manifold can be isometrically embedded into a Euclidean space of sufficiently high dimension. In this paper, we address the asymptotically hyperbolic version of the problem. We prove that an asymptotically hyperbolic manifold $\left(M,g\right)$ can be realized as a submanifold of a sufficiently high-dimensional hyperbolic space $\mathrm{H}^{N+1}$, intersecting the sphere at infinity transversely, if and only if $g$ is asymptotically Poincaré-Einstein to second order.

comments
12 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.40962 ↗

We classify complete, non-compact, simply connected, irreducible four-dimensional steady gradient Ricci solitons that are globally conformally Kähler and have an isoparametric potential. The steady Ricci solitons are $U(2)$-invariant nuts on $\mathbb R^4$, including the Bryant and Cao solitons, and bolts on $\mathbb{S}^2 \times \mathbb R^2$ and on $\mathcal O(-m)\to\mathbb{CP}^1$. For each $m\ge3$ there is exactly one non-collapsed bolt. A key step, valid for Ricci solitons of every sign, is that away from the locally conformally flat case, the conformal factor is aligned with the potential function, which yields cohomogeneity-one symmetry.

comments
35 pages, 1 figure
fields
Differential Geometry
pdf ↗ page arXiv:2610.40994 ↗

A conjecture of Berger says that on a simply connected manifold all of whose geodesics are closed, all unit speed geodesics have the same shortest period. In this paper we will show that the S^1-action induced by the geodesic flow is free. We use symplectic geometry to investigate the topology of the quotient of the unit tangent bundle T^1M/S^1 by the S^1 action that comes from the geodesic flow. We show that this quotient admits the structure of a smooth symplectic manifold and that this symplectic manifold is symplectomorphic to S^2 \times S^2 equipped with a split form. This will enable a computation involving the Euler class of the orbibundle T^1M\rightarrow T^1M/S^1 which will lead to a solution of Berger's conjecture for three-dimensional manifolds.

comments
29 pages
fields
Differential Geometry, Symplectic Geometry
pdf ↗ page arXiv:2610.41006 ↗

We study the relation between conformal geodesics and conformal Killing-Yano (CKY) 2-forms on anti-self-dual conformal manifolds of dimension 4. In the flat case, twistor quadrics determine self-dual CKY forms whose zero loci contain conformal geodesics. To extend this construction to curved manifolds, we consider self-dual 2-forms with values in a line bundle and satisfying the gauged CKY equation, coupled to connections with anti-self-dual curvature. In this way we provide a local ambient characterization of non-null conformal geodesics as zero curves of gauged CKY forms, with the curvature vanishing along these curves.

fields
Differential Geometry
pdf ↗ page arXiv:2610.41008 ↗

For immersed surfaces in the homogeneous spaces $E(κ,τ)$ with $τ\ne0$, we prove that the original Abresch–Rosenberg differential $[2(H+iτ)p-(κ-4τ^2)A^2]\,dζ^2$ is holomorphic if and only if the mean curvature $H$ is constant; here $p\,dζ^2$ is the Hopf differential and $A=\langleξ,\partial_ζ\rangle$. The result is local: no completeness, compactness or analyticity is assumed, and there are no exceptional surfaces. Where $dH\ne0$, holomorphy forces the shape operator into the normal form $\bigl(\begin{smallmatrix}0&τ\\τ&2H\end{smallmatrix}\bigr)$ in the frame $(\nabla H/|\nabla H|,J\nabla H/|\nabla H|)$. The round case $κ=4τ^2$ is settled by an elementary argument; for $κ\ne4τ^2$ the structure equations become a rational system whose solutions are algebraic, and an embedding into a field of Puiseux series excludes every branch. In $M^2(κ)\times\mathbb{R}$, $κ\ne0$, where the statement fails, we show that a non-CMC surface with holomorphic original differential has nowhere vanishing $dH$ and constant angle function, and belongs to an explicit family; in particular, closed surfaces with holomorphic original differential are CMC there. We also show that the Hopf-cylinder alternative in a theorem of Dorfmeister, Inoguchi and Kobayashi contains no non-CMC surface.

comments
35 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.41197 ↗

For $1\leq p\leq n/2$, we derive sharp eigenvalue estimates for the Hodge Laplacian on differential $p$-forms under the curvature condition introduced by Petersen and Wink, and study their equality cases. Equality recovers the stronger curvature operator lower bound and, in the closed case, leaves only round or Sasakian metrics on the universal cover. For manifolds with boundary, weak $p$-convexity of the boundary makes the unit hemisphere the only equality case for the absolute boundary problem.

comments
13 pages, comments are welcome!
fields
Differential Geometry, Spectral Theory
pdf ↗ page arXiv:2610.41297 ↗

We prove strong uniqueness of tangent flows at the first singular time of a closed Ricci flow when one tangent flow is generated by a complete smooth asymptotically conical Ricci shrinker. The modified flow converges to this shrinker in a fixed gauge, with quantitative estimates on expanding domains. The main ingredients are a Łojasiewicz-Simon inequality for Perelman's entropy, based on the global weighted divergence-free gauge constructed in the preceding paper, and estimates controlling the geometric radius under Ricci flow.

fields
Differential Geometry
pdf ↗ page arXiv:2610.41345 ↗

Let $(X,d,p)$ be a nontrivial pointed Gromov–Hausdorff limit of complete connected Riemannian $n$-manifolds satisfying $\operatorname{Ric}\geq-(n-1)g$. For $1\leq j\leq n$, set \[ \mathcal{R}_j(X) = \bigl\{x\in X:\text{ every tangent cone at } x \text{ is isometric to } \mathbb{R}^j\bigr\}, \qquad \mathcal{R}(X)=\bigsqcup_{j=1}^n\mathcal{R}_j(X). \] We prove that there is a unique $k$ such that $\mathcal{R}(X)=\mathcal{R}_k(X)$, answering the pointwise regular-dimension question originating in Cheeger and Colding's work~\cite[p408]{CC97}. We construct a volume-corrected heat entropy on RCD space and identify the distributional logarithmic-density defect $\operatorname{tr}(\operatorname{Hess} f)-Δf$ by Green-volume regularization.

comments
31 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.41449 ↗

We give an explicit construction of all ALF and ALE toric gravitational instantons in terms of certain rational matrix functions of one complex variable on the Riemann sphere; these provide patching matrices for their corresponding twistor holomorphic vector bundles. The reconstruction of the instantons arises from the solution to a Riemann–Hilbert splitting problem and we show how this can be solved via linear algebra, resembling but not equivalent to the Belinski–Zakharov ansatz. This construction gives a complete classification together with an explicit form for all toric gravitational instantons up to the resolutions of conical and orbifold singularities. We illustrate the formalism by computer–assisted numerical construction of two new ALF (but not AF) gravitational instantons with five nodes.

fields
Differential Geometry
pdf ↗ page arXiv:2610.41867 ↗

Geroch conjectured that the local solutions of the vacuum Einstein equations with two commuting Killing fields and vanishing twist constants are generated from flat space by an infinite-dimensional group of transformations, and Hauser and Ernst proved this for stationary axisymmetric spacetimes near the axis. The solutions generated from flat space by the transformations with finitely many poles in the spectral parameter are the solitons of Belinski and Zakharov. We prove a solitonic form of the conjecture in Riemannian signature, for toric gravitational instantons. The proof combines Weinstein's uniqueness argument, a closed formula for the soliton metrics in Hankel determinants and a degree argument. It also applies to the toric Ricci-flat metrics with conical singularities constructed by Li and Sun.

comments
88 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.41875 ↗

Let $M^{n+1}$ be a complete Riemannian manifold with $\sec_M\ge-K_0$, $K_0\ge0$, and let $\mathfrak{F}$ be a codimension one foliation of $M$ by hypersurfaces of constant mean curvature, the constant depending on the leaf. We prove that the mean curvature of every leaf satisfies $|H|\le(n+1)\sqrt{K_0}$. In particular, the mean curvature of the leaves of a CMC foliation of a complete manifold with $|\sec_M|\le1$ is bounded by $n+1$, which answers positively a conjecture of Meeks III, Pérez, and Ros.

comments
36 pages
fields
Differential Geometry
pdf ↗ page arXiv:2610.42093 ↗

Rings and Algebras11

Positive cyclic quantum mutations at a third root of unity induce an explicit weight on polygon flip graphs. We study the resulting discrepancy between minimum flip count and minimum accumulated normalization loss. Every pentagon three-to-two replacement strictly decreases weight, and in connected types $A_1$, $A_2$, and $A_3$ every minimum-weight path is a flip geodesic. In type $A_4$, however, a six-flip path has smaller weight than the unique five-flip geodesic. A compatible positive projective gluing construction amplifies this example to an exponential success-probability separation in every connected type $A_n$, $n\ge4$. We also give a quantitative perturbation bound preserving the separation. The low-rank minimality statement has an exhaustive finite certificate; the pentagon inequality and the all-rank construction are analytic. In the associated sequential postselection model, differences of path weights determine success-probability ratios independently of the input state. The comparison holds for separately normalized exact mutation steps.

comments
18 page
fields
Rings and Algebras, Algebraic Geometry, Combinatorics, Quantum Algebra
pdf ↗ page arXiv:2610.40661 ↗

In this paper, we investigate the generalized core-EP inverse of block operator matrices over a Banach algebra. As a result, novel representations and explicit formulae for the core and core-EP inverses of the $2\times 2$ block operator matrices are established.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.40670 ↗

We revisit a canonical filtration of simplicial algebras coming from the Moore-Postnikov system and provide several applications in studying the structure and classification of simplicial algebras and simplicial modules.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.40672 ↗

We define an invariant of central simple algebras of degree 8 and exponent 2 over fields of characteristic 2. It is built from the discriminant Pfister forms of symplectic involutions, takes values in a quotient of the third part of the graded group associated to the fundamental filtration of the Witt group of quadratic forms, and vanishes on tensor products of three quaternion algebras, thus giving a necessary condition for decomposability. Variants of the consctuction yield necessary and sufficient conditions for a biquaternion algebra to descend from a quadratic extension of the base field, and for a quaternion algebra to descent from a biquadratic extension that is not purely inseparable. Our results extend known results both in characteristic 2 and in characteristic different from 2, see the introduction. Using residue maps for quadratic forms, we show that these invariants do not vanish in general, and we give an explicit central simple algebra of degree 8 and exponent 2 with nonzero invariant, which is therefore indecomposable. As explained in the appendix, the purely inseparable case is quite different; in particular, a quaternion algebra descents if and only if its Brauer class also does.

fields
Rings and Algebras, K-Theory and Homology, Number Theory
pdf ↗ page arXiv:2610.40673 ↗

We study balanced extended $\mathcal O$-operators on Lie–Yamaguti algebras. The construction links operator equations, algebra splittings and Yang–Baxter equations. We build a double from a representation and a balanced equivariant map. The graphs of extended $\mathcal O$-operators are exactly its subalgebras complementary to the first summand. We define extended post-Lie–Yamaguti algebras and prove their two-way relation with extended $\mathcal O$-operators. The tensor form gives a balanced extended Lie–Yamaguti Yang–Baxter equation. We construct a curved $L_\infty[1]$-algebra whose Maurer–Cartan elements are extended $\mathcal O$-operators. The normal representation of a graph gives a cohomology theory for infinitesimal deformations, rigidity and obstructions. We include both cyclic conditions in the lowest paired degree. We also map operator infinitesimals to the deformation cohomology of the induced algebra. Examples give obstructed infinitesimals, rigid scalar operators and nontrivial formal deformations.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.40686 ↗

In this paper, we first construct the dual of a representation of a pre-Leibniz algebra and study infinitesimal deformations of a pre-Leibniz algebra with a representation, which lead to Nijenhuis pairs, characterizing trivial deformations, and to dual-Nijenhuis pairs. Building on these notions, we introduce Kupershmidt-(dual-)Nijenhuis structures and show that they give rise to hierarchies of pairwise compatible Kupershmidt operators, the converse holding under an invertibility condition. Finally, we show that Rota–Baxter–Nijenhuis structures, $r$-$n$ structures and BN-structures all arise as particular cases, and that the first two correspond to each other on quadratic pre-Leibniz algebras.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.40837 ↗

We classify linear maps $φ$ between possibly different rectangular matrix spaces over a field $\mathbb F$ with $|\mathbb F|\ge5$ such that $\operatorname{rank} φ(A)\le2$ whenever $\operatorname{rank} A\le2$. Every such map has one of six forms. If its image contains a matrix of rank greater than two, then $φ(A)=MAN$ or $φ(A)=MA^{\top}N$ for fixed matrices $M,N$. The remaining forms have images consisting entirely of matrices of rank at most two, including an exceptional form arising from $3\times3$ alternating matrices. As a consequence, we characterize linear maps preserving rank exactly two. We also prove that rank-$k$ preservation implies rank-$k$ nonincrease when $|\mathbb F|>k+2$, and that rank-$k$ nonincrease implies rank-$\ell$ nonincrease for every admissible $\ell\ge k$ when $|\mathbb F|>k+1$. The classification uses the fact that the Chow rank of the generic $3\times3$ determinant is five, connecting the rank preserver problem with product decompositions of polynomials.

comments
24 pages
fields
Rings and Algebras, Functional Analysis
pdf ↗ page arXiv:2610.40995 ↗

We show that the Lie algebra of algebraic vector fields for any connected, semi-simple matrix group over $\mathbb{F}$, for $\mathbb{F}=\mathbb{R},\mathbb{C}$, can be generated by three $\mathbb{F}$-complete algebraic vector fields.

comments
18 pages. A merged and extended version of arXiv:2608.21670 and arXiv:2609.02144, with new results
fields
Rings and Algebras, Complex Variables, Optimization and Control
pdf ↗ page arXiv:2610.41201 ↗

An anti-quadratic nearly associative algebra $(A, \bullet, ω)$ is a nearly associative algebra equipped with a non-degenerate skew-symmetric bilinear form $ω$ satisfying the anti-associativity condition $ω(u \bullet v, w) = -ω(u, v \bullet w)$ for all $u, v, w \in A.$ In this paper, we investigate the structural properties and extension theory of these algebras. We prove that any anti-quadratic nearly associative algebra is inherently nilpotent of index at most $6$. Furthermore, we introduce a novel $T T^*$-extension method and show that every anti-quadratic nearly associative algebra arises from this construction. We also introduce the notion of a double extension and prove that every such algebra can be constructed by iterated double extensions. Finally, as an application, we give a complete classification of anti-quadratic nearly associative algebras of dimension at most $4$.

comments
24 pages
fields
Rings and Algebras
pdf ↗ page arXiv:2610.41465 ↗

In this paper, we study quasi-projective dimensions of modules over rings connected by homological ring epimorphisms. We establish several inequalities relating the quasi-projective dimensions over the source and target rings. As applications, we investigate the quasi-global dimension of triangular matrix rings and provide examples showing that the class of modules with finite quasi-projective dimension does not satisfy the 2-out-of-3 property in general.

fields
Rings and Algebras, Representation Theory
pdf ↗ page arXiv:2610.41614 ↗

This paper introduces the ${\rm P}$-generalized determinant, a more general class of quaternion determinants. It unifies several classical quaternion determinants proposed between 1922 and 2008, including the Moore determinant, the Xie determinant, the Chen determinant, and the Kyrchei row–column determinants, thereby revealing the essential nature of quaternion determinants as multivariate polynomial-type functions with a fixed ordering of products. We provide equivalent characterizations of the expansion of the $\mathrm{P}$-generalized determinant along a given row or column, and establish conversion relations between the $\mathrm{P}_{l}$- and $\mathrm{P}_{r}$-generalized determinants. We prove that the $\mathrm{P}$-generalized determinant of a quaternion Hermitian matrix is a real number independent of the choice of cycle representation rule. We further investigate the behavior of the $\mathrm{P}$-generalized determinant under elementary row and column operations on quaternion Hermitian matrices, as well as its linearity in rows and columns. Finally, using the $\mathrm{P}$-generalized determinant, we derive determinantal representations of the inverse of an invertible quaternion Hermitian matrix and of the unique solution to the quaternion matrix equation $AXB=C$. In particular, this yields Cramer's rule for the left and right linear systems $Ax=b$ and $xA=b$.

fields
Rings and Algebras
pdf ↗ page arXiv:2610.41644 ↗

Statistics Theory11

Learning a shared subspace in multitask learning is an important statistical problem, yet exact rates remain poorly understood in many of the formal settings considered in the literature. To address this gap, which arises from limitations of existing upper-bound techniques, we propose a general multitask lemma. The lemma provides sufficient conditions for fast subspace estimation in generic multitask settings, in the form of geometric conditions on the task regression functions and their estimators. We subsequently show that the multitask lemma yields sharp minimax rates—previously unknown—for linear regression and classification in the regimes considered, and rates matching up to logarithmic factors for additive-noise regression.

comments
81 pages, 2 figures
fields
Statistics Theory
pdf ↗ page arXiv:2610.40713 ↗

We propose a test of the global null hypothesis of zero means in correlated multivariate Gaussian data by combining marginal and conditional evidence at each coordinate. The conditional tests are based on the distribution of each coordinate given the remaining coordinates. At each coordinate, the corresponding two-sided $p$-values are combined using the Simes method, followed by a Bonferroni correction across coordinates. Because conditioning can either strengthen or weaken evidence against the null, neither marginal nor conditional testing is uniformly preferable. The proposed procedure adapts to their relative strengths while retaining the simplicity of minimum-type global testing. We establish finite-sample control of the global Type-I error under both known covariance and covariance estimated from an independent sample. The latter setting requires a careful analysis of the dependence between studentized marginal and conditional statistics. We evaluate the power of the proposed test through simulations, compare its performance with existing global tests, and illustrate its practical utility using a TCGA carcinoma dataset.

fields
Statistics Theory
pdf ↗ page arXiv:2610.40899 ↗

For multinomial counts $X\sim\operatorname{Mult}(N;\mathbf p)$ on $n$ ordered bins and a reference law $\mathbf r$ we study the expectation of the reference-centered discrete total variation $d_{\mathbf r}(X)=\sum_{j=1}^{n-1}|(X_{j+1}-Nr_{j+1})-(X_j-Nr_j)|$, the natural extension of the comb-test statistic to a non-uniform null. For a non-uniform null the raw total variation of the counts carries a deterministic staircase of order $N$ that no centering of the statistic removes; the counts are therefore centered at the reference law, and under the null ($\mathbf p=\mathbf r$) the statistic is the matched $d_{\mathbf p}$. Exact finite-sample expectations are obtained by conditioning on the total mass in each adjacent pair, which reduces every edge to a translated binomial absolute-deviation problem. For fixed $n\ge3$ and a sampling law bounded away from the boundary of the simplex, the matched statistic admits a three-term asymptotic expansion in powers of $1/N$ whose coefficients are explicit continuous periodic functions of $N(p_{j+1}-p_j)$ (a generalized, non-Poincaré asymptotic expansion), obtained by a weighted local Edgeworth estimate and a corner Euler–Maclaurin summation; all half-integer orders vanish by parity. Each correction separates into a smooth Edgeworth part and an oscillatory lattice part, the latter an Edgeworth perturbation of one formal Euler–Maclaurin series of a Gaussian lattice sum. Under a sampling law $\mathbf p\neq\mathbf r$ the leading expectation is the folded-normal profile implied by the central limit theorem; under local alternatives we identify the $O(1)$ correction (a skewness term, plus a variance-gradient term for Pitman alternatives), and fixed alternatives give a law-of-large-numbers limit, the $\ell_1$ distance between the adjacent-gradient profiles of $\mathbf p$ and $\mathbf r$, on which threshold tests are consistent.

comments
43 pages, 2, tables
fields
Statistics Theory
pdf ↗ page arXiv:2610.41255 ↗

Nearest-neighbor two-sample tests are often analyzed with dimension fixed. We study the power of the generalized edge-count test on a directed 1-nearest-neighbor graph. This statistic combines a weighted sum and a difference of the two within-sample edge counts. For balanced isotropic Gaussian samples with $d/\log N\toγ>0$, we derive the connection probabilities under local location and scale alternatives. Under location alternatives, the two counts have equal mean shifts, which cancel in their difference. Under scale alternatives, the two counts instead have mean shifts of opposite signs. We also derive the growth rate of the null expectation $\mathbb{E}_0C_3$, where $C_3$ counts pairs of directed edges sharing exactly one vertex. On an explicit range of $γ$, we establish local-power limits strictly above the test level for scale alternatives in the difference component at the scale $|\varepsilon_N|\asymp (\mathbb{E}_0C_3)^{1/2}/(Nd)$. Here $\mathbb{E}_0C_3$ gives the variance order of the difference count under these local alternatives. At this scale, the standardized drift of the weighted sum vanishes. When $d=1$, the weighted component instead detects local Gaussian scale alternatives while the difference component has no limiting drift. Thus dimension changes both the magnitude of a detectable scale difference and the component of the test that detects it.

fields
Statistics Theory
pdf ↗ page arXiv:2610.41271 ↗

Statistical contracts reward experimenters according to the evidence produced by a costly study, so the reward rule influences both who participates and what they report. Principal–agent testing theory explains how to prevent ineffective proposals from being profitable. We ask when such contracts can also elicit honest parameter reports if the decision concerns a mean effect but the variance or distributional shape is unknown. Under risk neutrality and a fixed participation cost, we construct menus of e-values --- nonnegative normalized rewards with expected value at most one under every null --- such that (i) agents who know they are null have no incentive to participate, (ii) every alternative type earns positive expected profit and is thus incentivized to participate, (iii) truthful reporting uniquely maximizes expected their reward, (iv) and each contract is admissible: its payouts cannot be pointwise improved while retaining null validity. For Gaussian observations, we treat separately agents who know both mean and variance and agents who know only their mean; in the latter case, truthfulness holds at every variance without requiring a nuisance prior. Multi-sample extensions preserve these incentives and can reward accumulating evidence. Beyond Gaussian models, we show that two independent nonnegative observations suffice to elicit any finite mean above a positive null threshold, without any further distributional knowledge, but one observation cannot combine strict mean elicitation with individual admissibility on this class. Finally, we distinguish the existence of useful tests from the identification needed for truthful contracts. Boundedness is not required, and finitely many null constraints do not by themselves guarantee admissible elicitation.

comments
30 pages
fields
Statistics Theory
pdf ↗ page arXiv:2610.41283 ↗

This paper develops methods for simultaneous inference on component functions in additive models. We first propose bias-corrected smooth backfitting estimators inspired by least squares linear regression and establish their asymptotic properties. We then show that the distributions of the uniform errors of these estimators converge to a Gumbel-type distribution. Based on this distributional approximation, we construct simultaneous confidence bands and develop a practical residual-based procedure for testing the additive components. Simulation studies and empirical data analyses demonstrate that the proposed estimators reduce bias and improve estimation accuracy compared with several existing estimators. Moreover, the empirical coverage probabilities of the proposed simultaneous confidence bands are found to be comparable to those of their oracle counterparts.

fields
Statistics Theory
pdf ↗ page arXiv:2610.41442 ↗

We present a unified framework for uncertainty quantification over filtration scale, extending the topological average treatment effect (TATE) analysis of \citet{kim-lee} to conditional mean confidence and individual effect prediction. For the conditional topological average treatment effect (CTATE), we derive multiplier confidence bands at a fixed covariate value, simultaneous over filtration scale, and prove validity in a correctly specified finite-dimensional regression regime, with explicit control of smoothing bias and nuisance-estimation error. For the individual topological treatment effect (ITTE), we integrate existing weighted and functional split-conformal methods to obtain finite-sample marginal prediction coverage under known target-to-calibration likelihood ratios, with a total-variation bound for estimated ratios. We also extend the current TATE silhouette stability analysis to the diagram-independent bound $\|Φ_r(D)-Φ_r(D')\|_\infty \le (1+2r)W_1(D,D')$ for $r\geq1$, recovering the known $r=1$ constant, and propagate diagram approximation error to prediction bands. The installable Python library \texttt{tcda\_uq} integrates mean confidence, individual prediction, and topological preprocessing, with tutorials and reproduction materials. Simulations and a PACCE lesion-measurement application illustrate the distinct coverage targets and the importance of declaring the target population.

comments
57 pages, 9 figures
fields
Statistics Theory
pdf ↗ page arXiv:2610.41593 ↗

We show that outlier eigenvectors of the Bethe Hessian achieve weak recovery in the stochastic block model down to the Kesten-Stigum threshold, for every fixed common expected degree $d>1$. This extends previous recovery guarantees, which required $d$ to be sufficiently large. This threshold is information-theoretically optimal for the symmetric two-community model and is conjectured to mark the limit of efficient recovery in broader symmetric models. Our result matches the recovery guarantees of non-backtracking spectral methods while using Hermitian $n\times n$ matrices. To do this, we identify recursive formulas for the negative outlier eigenvalues and show that their eigenspaces have nonvanishing overlap with the corresponding planted eigenspaces. The proof constructs approximate eigenvectors from the planted labels on the limiting Poisson tree, with weights given by its Green's function, and transfers these approximations to the finite graph.

comments
28 pages
fields
Statistics Theory, Combinatorics, Probability
pdf ↗ page arXiv:2610.41802 ↗

We consider estimation of entropy of a multivariate variate normal distribution with unknown mean vector and unknown covariance matrix. Using a decision-theoretic approach, we study the problem under a general location-invariant loss function that is strictly convex and minimized at zero. We first characterize the class of affine equivariant estimators and obtain the best affine equivariant estimator. We further show that the best affine equivariant estimator is inadmissible under the general loss. Exploiting information in the sample mean, we derive Stein-type improved estimators, including an extension to the case of a mean matrix. Under the linex loss, we obtain a smooth Brewster–Zidek-type estimator that dominates the best affine equivariant estimator, and we show that this estimator is generalized Bayes under a suitable non-informative prior.

fields
Statistics Theory
pdf ↗ page arXiv:2610.41849 ↗

We study high-dimensional graph combinatorial hypothesis testing problems in which a signal is supported on the edge set of an unknown subgraph drawn from a class invariant under vertex relabeling. For Gaussian observations, we characterize sharp statistical and computational thresholds (at polynomial order) for the signal magnitude required for strong detection across various combinatorial graph alternatives. Our analysis is based on bounding the second moments of the full and low-degree likelihood ratios in terms of the overlap of two independent latent structures and comparing the resulting lower bounds with scan, average, and coordinatewise maximum tests. Under an appropriate low-degree conjecture for computational hardness, we obtain planted-clique thresholds matching those known for Gaussian submatrix detection and identify new statistical-computational gaps for multiple planted cliques and hypercliques, thick cycles and thick paths. We also establish a novel transfer principle across natural exponential family observation models: matching the one-coordinate chi-square divergence preserves full likelihood-ratio second moments exactly and makes low-degree second moments equivalent up to constant factors under a degree condition. Under local regularity, we obtain corresponding testing upper bounds in terms of an effective signal strength. When these upper and lower bounds match, they yield universal threshold scales in the local alternative regime. As a consequence, statistical-computational gaps for planted-clique detection in Erdős–Rényi settings can be deduced from the Gaussian model.

comments
106 pages, 6 figures, 1 table
fields
Statistics Theory
pdf ↗ page arXiv:2610.41995 ↗

Consider the classical statistical problem of sub-Gaussian mean estimation under martingale dependence: given $n$ observations with conditional mean $μ$ and a known conditional sub-Gaussian proxy $s>0$, we seek a nonasymptotically valid confidence interval for $μ$. The usual Chernoff interval has half-width $s\sqrt{2\log(2/α)/n}$; it neither adapts to the actual standard deviation $σ\le s$ nor attains the Gaussian quantile $z_{1-α/2}$. We construct an analytically closed-form nonasymptotically-valid interval under martingale dependence whose endpoints, at every iid law with $σ> 0$, agree almost surely to first order with $\bar X_n\pmσz_{1-α/2}/\sqrt n$. The two new building blocks are a sub-Gaussian exponential supermartingale with an \emph{empirical} quadratic variance and a cubic correction term, and a sub-Gaussian continuation supermartingale, the former delivering variance adaptivity ("Empirical-Bernstein'' for unbounded data), the latter also providing the Gaussian quantile ("Empirical-Berry-Esseen'' for unbounded data). The same constructions yield a closed-form time-uniform confidence sequence with asymptotically unimprovable half-width $σ\sqrt{2\log\log n/n}$. We also give closed-form multivariate ellipsoids with the true-covariance chi-square limit, and extend the scalar Gaussian and LIL constructions to handle heavy-tailed settings where one only knows a bound on any absolute central moment of order greater than two. Experiments show substantial width reductions when the true variance is smaller than the implied variance bound, at small loss in the boundary cases. The surprising (to us) takeaway message is: given any more than two moments, one can construct fully data dependent nonasymptotic confidence sets for the mean that are asymptotically efficient.

fields
Statistics Theory
pdf ↗ page arXiv:2610.42058 ↗

Dynamical Systems10

We study equivariant gradient degree comparisons for strongly indefinite Hamiltonian systems through Clarke duality. Under suitable convexity and compactness assumptions, we establish a correspondence between the primal and dual equations that preserves the symmetries of their solutions. We derive an identity relating their negative spectral representations and show that the Galerkin-normalized local index of the primal linearization coincides with the equivariant gradient degree determined by the dual Hessian. This comparison extends to infinite-dimensional settings through stabilized Galerkin approximations. We also establish relations between primal and dual crossing forms, equivariant spectral flow, and local equivariant gradient degrees at nondegenerate fixed solutions. The results provide a framework for studying equivariant bifurcation in strongly indefinite systems. As an application, we consider periodic Hamiltonian ring equations with dihedral symmetry and illustrate the computation of equivariant gradient degrees and bifurcation invariants.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.40896 ↗

A nonlinear dynamical system can be characterized by the autonomous evolution of its state in time and its dependence on system parameters. As a parameter varies, the dynamics can change quantitatively and, at bifurcation points, qualitatively. When the governing equations are unknown, a data-driven model should therefore capture both elements from observational data, rather than reproduce the dynamics at a single operating condition only. Parameter-aware reservoir computing (PARC) is a model-free framework for this purpose, where the parameter is supplied to a reservoir computer through an additional input channel. PARC is trained on time series recorded at several parameter values, and the trained reservoir, run in closed loop as an autonomous dynamical system, generates trajectories at parameter values that were never observed. rc_prediction is an open-source Python package that implements this workflow. It provides multi-parameter training of a reservoir, autonomous closed-loop prediction at user-specified parameter values, parameter scans, and a standard echo state network for forecasting at a fixed parameter. Hyperparameter optimization and optional GPU acceleration through PyTorch are also provided. We use three benchmark dynamical systems as examples to demonstrate its usage. As one of the featured applications, here we illustrate the prediction of critical transitions and system collapse: trained only on pre-critical data, the model is queried beyond the training range to predict whether the system collapses, to estimate the critical parameter $p_c^{*}$, and to characterize transient lifetimes.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.40984 ↗

Let $\{T_t\}_{t\in\mathbb R}$ be a measure-preserving flow on a probability space $(Ω,Σ,μ)$ and let $T^*$ be the associated one-sided ergodic maximal operator. We show that the sharp Tauberian constant $C(α)=\sup_{μ(E)>0}\frac{μ(\{T^*χ_E>α\})}{μ(E)}$,$0<α<1$, equals $1/α$ for every $α\in(0,1)$, unless every measurable set is invariant modulo null sets, in which case $C(α)=1$ for every $α$. In particular the Solyanik estimate $\lim_{α\to1^-}C(α)=1$ holds for every measure-preserving flow, and for ergodic flows on nonatomic spaces the estimate $C(α)-1\le 1/α-1$ holds with equality. This answers the one-parameter case of a question of Hagelstein and Parissis. The main point is the lower bound, which is obtained from a lacunary cross-section built out of the upcrossings of a probe function along the orbits of the flow.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.41151 ↗

Let $U_1,\dots,U_n$ be commuting invertible measure-preserving transformations of a probability space, and let $C^{*}_{\mathrm{HL}}(α)$ denote the sharp Tauberian constant of the associated ergodic uncentered Hardy–Littlewood maximal operator. Hagelstein and Parissis showed that $C^{*}_{\mathrm{HL}}(α)-1\lesssim_n(1/α-1)^{1/(n(n+1))}$ as $α\to1^-$ and asked whether this exponent is sharp. We prove that $C^{*}_{\mathrm{HL}}(α)-1\lesssim_n(1/α-1)^{1/n}$ for all such transformations, and that this is sharp: equality of the ergodic and discrete Tauberian constants holds for every action admitting arbitrarily large Rokhlin towers (in particular for every free action, ergodic or not), and the discrete constant satisfies $\widetilde{C}_{\mathrm{HL}}(α)-1\asymp_n(1/α-1)^{1/n}$. For $n\ge2$ the sharp ergodic exponent $1/n$ is strictly smaller than the sharp Euclidean exponent $2/(n+1)$ recently obtained by Mukherjee; the discrepancy is caused by a lattice ``first layer'' effect. We also show that the discrete Tauberian constants of the Hardy–Littlewood and strong maximal operators dominate their Euclidean counterparts; for the strong maximal operator this yields $C^{*}_{\mathrm{S}}=C_{\mathrm{S}}$ for every free action.

fields
Dynamical Systems
pdf ↗ page arXiv:2610.41156 ↗

We develop a divergence-kernel method for jointly optimizing neural SDE drift and diffusion fields and the initial Gaussian distribution, using losses that depend directly on the terminal density. In two- and eight-dimensional Gaussian-mixture experiments, we learn time-independent fields and obtain far fewer off-mode samples than score-based generators with similar parameter counts. The drift attracts samples toward the modes, while the diffusion helps disperse excess probability between them.

comments
17 pages, 5 figures
fields
Dynamical Systems
pdf ↗ page arXiv:2610.41243 ↗

We study the worst-case relative growth rate of a fixed functionally generated portfolio under restrictions on market-weight dynamics over a compact convex market state space. For a nonempty compact forward shift-invariant path family, the pathwise wealth decomposition identifies the global rate with the minimum invariant average of the one-step $L$-divergence. For paths generated by a compact one-step transition relation, this rate is also the minimum over balanced transition measures, and a minimizing measure can be chosen on one relation basic set. For a fixed initial state, we study the lower limit of the minimum average $L$-divergence over the first $t$ periods as $t\to\infty$. We represent this value using empirical transition laws generated from that state. Their common state marginals are supported on chain-recurrent states that are chain-accessible from the initial state. The lower bound is obtained by minimizing the average $L$-divergence over balanced transition measures on chain-accessible basic sets. Finite exact reachability gives an upper bound. Examples show that both bounds can be strict. We also give a finite exact reachability condition under which the minimum average $L$-divergence over the first $t$ periods converges for paths confined to a basic set.

comments
24 pages,0 figures
fields
Dynamical Systems
pdf ↗ page arXiv:2610.41603 ↗

We give conditions that are all equivalent to surjectivity for endomorphisms of irreducible Smale spaces. This extends results previously obtained by Hedlund~\cite{hedlund-endomorphisms} for full shifts, by Coven and Paul and by Lind and Marcus for irreducible subshifts of finite type.

fields
Dynamical Systems, Mathematical Physics
pdf ↗ page arXiv:2610.41794 ↗

Analysis of mathematical models in ecology and epidemiology often emphasizes asymptotic dynamics, such as stable equilibria, periodic orbits, and attractors. However, many biologically relevant trajectories exhibit long transient dynamics in which a relevant observable evolves slowly for an extended period before a comparatively rapid transition occurs. On finite time scales relevant to data collection, management, and intervention, such transient behavior can be more informative than the eventual asymptotic outcome. In this work, we study long-lasting and slowly varying transient dynamics in continuous-time systems through the framework of nonequilibrium transient centers. These centers are points in the state space near which the observable can remain nearly constant for a long time, even when the full state is not near an equilibrium. We develop finite-time escape criteria for identifying such centers and their reachable counterparts. These criteria provide practical conditions under which nearby trajectories can remain in a slow regime for a long period of time before eventually leaving it. We apply the theory to an epidemic model with delay of stability loss. In this model, trajectories may remain close to the disease-free manifold for a long time before the infectious population resurges. The framework developed here describes this behavior as slow evolution near a nonequilibrium transient center followed by finite-time escape from the slow regime.

comments
27 pages
fields
Dynamical Systems
pdf ↗ page arXiv:2610.41930 ↗

Let $Γ\subset \operatorname{SL}(2, \mathbb{R})$ be discrete, co-finite with cusps, and let ${\frak b}\in \mathbb{R}\cup\{\infty\}$ be a cusp of $Γ$. Let $f: Γ{\frak b} \to \mathbb{R}$ be a quantum modular form in the sense that the maps $$ φ_γ(x) = f(x) - f(γx) \qquad (x\in Γ{\frak b}, γ\in Γ) $$ extend to Hölder-regular functions. Under bounded moments assumptions of the shape $\sum_n n^{-2} \int |φ_{γ^n}(x)|^αd x < \infty$ for some $α>0$ and every fixed parabolic element $γ\in Γ$, we prove asymptotic formulas for the characteristic functions of the values $f(x)$ for $x\in Γ{\frak b}$, ordered by growing denominators. These yield Gaussian limits when $α>2$ and other stable limit laws under suitable further assumptions. This generalizes to Fuchsian groups the work of Baladi-Vallée on rational orbits of the Gauss map. In companion papers, we use this result, and in particular its flexible assumptions on $φ_γ$, to deduce limit theorems for automorphic $L$-series at the central point. One ingredient in our proof is a particular cuspidal acceleration of the Bowen-Series map, whose inverse branches admit a convenient parametrization, and for which we describe explicitly a natural extension.

fields
Dynamical Systems, Number Theory
pdf ↗ page arXiv:2610.41973 ↗

Let $M$ be a closed Riemannian surface, and let $f_1,\ldots,f_s$ be area-preserving $C^2$ diffeomorphisms with fixed positive probabilities such that the resulting random dynamical system is uniformly expanding on average. For each $n$, choose $ω_1,\ldots,ω_n$ independently according to these probabilities and set $F=f_{ω_n}\circ\cdots\circ f_{ω_1}$. We prove that there exists $α>0$ such that, with probability at least $1-e^{-αn}$, the correlations of any two Hölder observables $φ,ψ$ under $F^m$ are bounded by $e^{-αn}\|φ\|_{C^β}\|ψ\|_{C^β}$, for all $m\leq e^{αn}$. These estimates imply a law of large numbers and a central limit theorem for the Birkhoff sums along $F$ of any zero-mean Hölder observable $φ$, for exponentially long times in $n$. The limiting Gaussian has variance $\int_Mφ^2\,d\mathrm{vol}$.

comments
19 pages
fields
Dynamical Systems
pdf ↗ page arXiv:2610.42083 ↗

Geometric Topology10

This thesis investigates two themes. The first studies systolic embeddings of graph on translation surfaces, proving existence results for essential and cellular systolic embeddings and determining extremal genera for certain graph families. The second examines hyperbolic surfaces in the Thurston set, constructing a one-parameter deformation that increases the sys- tole length and produces Bolza-like surfaces, i.e., surfaces whose systolic geodesics decompose the surfaces into triangles, for infinitely many genera. A symmetry property of geodesic inter- sections on these surfaces is also established, together with an application of these Bolza-like surfaces.

comments
Ph.D. thesis, Department of Mathematics and Statistics, Indian Institute of Technology Kanpur, 2026
fields
Geometric Topology
pdf ↗ page arXiv:2610.40750 ↗

Using Real Seiberg–Witten theory, Miyazawa constructed infinitely many exotic involutions on ${\overline{\mathbb{CP}}{}^2}$, and Baraglia extended this construction to $X\#{\overline{\mathbb{CP}}{}^2}$ for a broad class of $4$-manifolds $X$ equipped with an involution $c$. We show that, under suitable Real Seiberg–Witten hypotheses, the ${\overline{\mathbb{CP}}{}^2}$ summand can be omitted when $X$ is non-spin and the complement of the fixed locus is simply-connected. Our construction also yields infinite families of topologically isotopic but smoothly non-isotopic embeddings of the branch surface in the quotient $X/\langle c\rangle$. As applications, we construct anti-holomorphic involutions on irreducible, simply-connected projective surfaces of general type with infinitely many exotic copies, none of which is anti-symplectic with respect to any orientation-compatible symplectic form. These surfaces include cyclic branched covers of $\mathbb{CP}^2$ and complete intersections of multidegree $(2,4k)$ in $\mathbb{CP}^4$.

comments
24 pages, no figures, comments are welcome
fields
Geometric Topology
pdf ↗ page arXiv:2610.40802 ↗

Using an obstruction derived by Owens and relying on Donaldson's Diagonalisation Theorem, we obtain a complete classification of lens spaces of Lisca type $(2)$ or $(3)$ that are unobstructed from embedding in a homology complex projective plane. This classification is further refined for lens spaces $L(p,q)$ with $p$ even, using an obstruction due to Lidman–Moore–Vazquez based on work of Lin. This article is extracted from the author's PhD thesis.

comments
Comments are welcome!
fields
Geometric Topology
pdf ↗ page arXiv:2610.40957 ↗

In this note I prove that every non-trivial element of the fundamental group of a closed orientable surface X is non-trivial in the fundamental group of some handlebody with boundary X. This result appears to be folklore. I have uploaded a proof as the result is needed in a joint project with Clement Dupont on mapping class group actions on configuration spaces.

comments
4 pages
fields
Geometric Topology
pdf ↗ page arXiv:2610.40987 ↗

We study the asymptotic distribution of the number of primitive, closed geodesics on Weil-Petersson random hyperbolic surfaces with genus $g$ and $n(g)$ cusps, under various growth rates of $n(g)$ as $g\to\infty$. We prove that the limiting distribution is Poisson with parameter depending explicitly on the limit of $n(g)^2/g$. This work is inspired by that of Mirzakhani and Petri's on closed surfaces and Hide and Thomas when $g$ is fixed. Our proof relies on first order expansions for Weil-Petersson moduli space volumes in the $n(g)/g\to 0$ regime. We also obtain some new existence results for small Laplacian eigenvalues.

fields
Geometric Topology, Probability, Spectral Theory
pdf ↗ page arXiv:2610.41002 ↗

We prove a generalized Table Theorem for bounded convex domains in the plane. Then we prove a variation of the Table Problem for compactly supported functions on the plane. Our method is based on relating the table problem on a surface to the square peg problem for a configuration of star-shaped curves in the tangent bundle of the surface.

fields
Geometric Topology, Differential Geometry
pdf ↗ page arXiv:2610.41010 ↗

We show that rationally slice knots need not be biprojectively H-slice (BPH-slice), answering a question in the study of the rational knot concordance group. Our main tool is a lattice embedding obstruction to H-sliceness in $\#^n\mathbb{CP}^2$ for knots with vanishing signature, extending work of Kjuchukova–Miller–Ray–Sakallı. We apply the obstruction to an infinite family of rationally slice knots, beginning with $8_{18}$. As additional applications, we obstruct the BPH-sliceness of a family of 2-bridge knots which is not obstructed by the instanton knot concordance invariant $r_0$, we obstruct a non-alternating family of Montesinos knots, and we give prime alternating knots that are topologically but not smoothly biprojectively H-slice. We discuss why these knots remain plausible candidates for detecting exotic smooth structures on $\#^n\mathbb{CP}^2$ and suggest natural sources of further candidates.

comments
48 pages, 17 figures
fields
Geometric Topology
pdf ↗ page arXiv:2610.41996 ↗

We present a new diagrammatic construction of Lagrangian surfaces in all symplectic toric $4$-manifolds from objects called multi-grid diagrams. These are a generalisation of triple grid diagrams for Lagrangian surfaces in $\mathbb{CP}^2$ wherein the underlying trisection is replaced by the more general notion of a toric decomposition. We give examples of multi-grid diagrams for Lagrangian surfaces in a range of symplectic toric $4$-manifolds, including Evans' recent non-squeezable Klein bottle in $S^2 \times S^2$.

comments
50 pages, 40 figures
fields
Geometric Topology, Symplectic Geometry
pdf ↗ page arXiv:2610.42076 ↗

Every link in the \(3\)-sphere admits a generalised \(T\)-link presentation. In this paper, we give an explicit polynomial-time algorithm which, given a signed Artin braid word and the number of strands, computes such a presentation. Combining this construction with Yamada–Vogel braiding gives a polynomial-time conversion from arbitrary oriented combinatorial link diagrams to generalised \(T\)-link parameters. We prove polynomial bounds on the number of strands and crossings in the resulting braid. As an application, we prove that a link property or a link invariant is computable in polynomial time from diagrams if and only if the corresponding problem is solvable in polynomial time from generalised \(T\)-link parameters, with parameter input measured by the total number of strands and crossings in the corresponding braid diagram. Whenever a classification criterion or an explicit invariant formula is available in these parameters, it can be applied to the original link by substituting the parameters obtained from the diagram. A polynomial-time test or evaluation then gives a polynomial-time procedure on diagrams. We also establish the corresponding transfer for approximating real-valued invariants to prescribed accuracy.

comments
30 pages
fields
Geometric Topology
pdf ↗ page arXiv:2610.42077 ↗

Let $S$ be a closed oriented surface of genus $g\geq 2$. We prove that any representation of $π_1(S)$ into $\mathrm{PSL}(2,\mathbb{R})$ with Euler class $3-2g$ is the holonomy of a hyperbolic metric on $S$ with one simple branch point.

comments
52 pages, 14 figures
fields
Geometric Topology, Differential Geometry
pdf ↗ page arXiv:2610.42079 ↗

Group Theory9

In this paper, we investigate the low-depth circuit complexity of Group Isomorphism in the multiplication (Cayley) table model. We prove the first circuit lower bounds for Group Isomorphism: namely, we show that every family of depth-$2$ Boolean circuits deciding Group Isomorphism requires quasipolynomial-size. (We also extend this to a limited class of depth-3 circuits, namely $\textsf{CNF}$-of-$\textsf{XOR}$s.) We complement this with upper bounds of uniform depth-$2\frac{1}{2}$ circuits of quasipolynomial-size: depth 3 circuits with poly-logarithmic bottom fan-in. A sequence of previous results from 1970-2025 progressively reduced the circuit depth from polynomial to $3\frac{1}{2}$; all of these results relied on the generator-enumerator strategy and, in fact, applied more generally to quasigroups. In contrast, our depth-$2\frac{1}{2}$ construction follows a fundamentally different strategy that exploits structure more specific to groups. We guess a composition series for each group, together with generators for its terms and the isomorphism types of its composition factors. We then inductively verify that the corresponding extensions at each level of the two composition series are compatible. A central part in this approach brings to bear the extensive work on the Short Presentation Conjecture, in tandem with the algorithmic theory of group extensions and cohomology.

comments
New result: lower-bound against CNF-of-XORs. 69 pages
fields
Group Theory
pdf ↗ page arXiv:2608.26257 ↗

In this paper, we describe the centroid (according to S. Lioutikov and A. Myasnikov) of an arbitrary finitely generated graph two-step nilpotent Hall exponential group over a binomial domain whose commutativity graph has no central vertices. Based on this description, we prove that if the non-commutativity graph of such a group has exactly one connected component and the group is non-Abelian, then the group is rigid.

comments
The paper will be sent to the journal of Groups, Complexity, Cryptology
fields
Group Theory
pdf ↗ page arXiv:2610.40680 ↗

Given a finite group $G$, the factorization graph $Γ(G)$ has as vertices the nontrivial proper subgroups of $G$, with two distinct vertices $H$ and $K$ adjacent if and only if $HK = G$. In this paper, we study structural and combinatorial properties of $Γ(G)$ for three families of groups: cyclic groups of square-free order, generalized quaternion groups, and dicyclic groups. For cyclic groups $C_{p_1\cdots p_r}$, we describe the global structure of $Γ(G)$ and determine its order, degree sequence, center, girth, and diameter. We also characterize completely when $Γ(G)$ is claw-free, show that it is non-planar for $r \geq 4$, and prove that after removing all pendant vertices, the resulting induced subgraph is 2-connected; we further compute its exact circumference for small $r$, showing it is Hamiltonian precisely when $r=4, 5$. For generalized quaternion groups, we classify all isolated vertices and prove that the reduced factorization graph has diameter two and a unique central vertex. Moreover, we show that it is planar, has chromatic number~$3$, is Eulerian, and non-Hamiltonian, and that it contains maximal induced stars and triangular book subgraphs. Finally, we analyze the dicyclic case and show that the reduced factorization graph is a complete bipartite graph, from which all main invariants follow directly. An explicit SageMath algorithm for constructing $Γ(G)$ is included.

fields
Group Theory
pdf ↗ page arXiv:2610.40860 ↗

Let $p$ be an odd prime and $G$ a finite $p$-group whose Jennings series has length at most $p$, that is, $D_{p+1}(G)=1$; when $D_p(G)\ne1$ assume also that $G/D_p(G)$ has nilpotency class at most $p-2$. We prove that $G$ is determined up to isomorphism by its modular group algebra $\mathbb{F}_p G$, and indeed by the quotient $\mathbb{F}_p G/\mathrm{I}^{\,ν+1}$, where $\mathrm{I}$ is the augmentation ideal and $ν$ the length of the Jennings series; for an arbitrary finite $p$-group $G$, the quotient $\mathbb{F}_p G/\mathrm{I}^{\,p}$ determines $G/D_p(G)$; and that every automorphism of $\mathbb{F}_p G$ is the product of an automorphism of $G$ and an automorphism inducing the identity on $\mathrm{I}/\mathrm{I}^2$. In particular the modular isomorphism problem has a positive answer for all $p$-groups of exponent $p$ and class less than $p$, and for all $p$-groups of class at most $p-2$ and exponent dividing $p^2$ whose derived subgroup has exponent $p$. Consequently every counterexample in odd characteristic has class at least $p-1$, or a derived subgroup of exponent at least $p^2$, or an element of order $p^3$. The proof combines a descent theorem in the truncated group algebra of a free group, valid for every prime, with a relative rigidity theorem for the unipotent automorphisms of truncated group algebras of groups of exponent $p$, proved through the Lazard correspondence. The bound $p$ is the range in which the logarithm of the group algebra exists.

fields
Group Theory
pdf ↗ page arXiv:2610.40868 ↗

Let $G$ be a group and let $n\in\mathbb N$. We consider one-dimensional cellular automata of length $n$ over the alphabet $G$, with periodic boundary conditions and with local rules $φ$ given by words. We construct a functor $F$ from $\mathbf{Grp}$, the category of groups (or any similar algebraic structure), to the category of pointed digraphs, and show its faithfulness. We introduce the image category $\mathcal C_{n,φ}$ of state transition graphs of all such group-alphabet cellular automata, and show $F$ is an equivalence onto this category. Consequently we investigate the translation of categorical structures in $\mathbf{Grp}$ in terms of automata, and give examples of how standard methods of homological algebra may now be applied to automata. Additionally, a graph-theoretical construction of the product in $\mathcal C_{n,φ}$ is given, together with a formula for cycle lengths and multiplicities in this product. When $G$ is abelian, further structural properties of state transition graphs follow, and we contrast these results with non-abelian counterexamples.

fields
Group Theory, Combinatorics, Category Theory, Dynamical Systems
pdf ↗ page arXiv:2610.41013 ↗

A finitely generated group acting locally elliptically on a complete \CAT(0) space of finite geometric dimension fixes a point if distances to elliptic fixed sets are uniformly controlled by displacements. For arbitrary groups, the same hypothesis gives a fixed point in the space or its visual boundary. The proof combines a first-moment random-walk displacement estimate with Izeki's zero-drift theorem. For finite-rank $\R$-buildings whose Euclidean metric factor has finite linear image, we obtain an all-type local-to-global theorem, including the nondiscrete $\widetilde G_2$ case, with an explicit gap and without metric completeness; the linear-image restriction can be removed from the fixed-point conclusion when the Euclidean factor has dimension at most three. A tangent-cone criterion characterizes the gap and yields a cubical bound with optimal asymptotic growth $d/π$ in dimension $d$. For Hadamard manifolds, the gap is equivalent to uniformly bounded finite stabilizers in the isometry image, with stabilizer bounds of optimal polynomial growth in the gap constant.

fields
Group Theory, Geometric Topology, Metric Geometry
pdf ↗ page arXiv:2610.41468 ↗

We classify all non-abelian finite simple groups that possess conjugacy classes of size $3p^n$ for some prime $p$ and non-negative integer $n$. As an application, we study the structure of the normal closure $\langle g^G\rangle$ of an element $g$ in an arbitrary finite group $G$, provided that the conjugacy class $g^G$ has size $3p^n$. Specifically, we provide conditions for $\langle g^{G}\rangle$ to be solvable and completely determine its quotient modulo the solvable radical in the non-solvable case.

comments
12 pages
fields
Group Theory
pdf ↗ page arXiv:2610.41488 ↗

We prove that the braided Thompson groups $BV$ and $BF$ are Hopfian, answering the two parts of Zaremsky's question. We also prove that, for every discrete group $G$, the labeled Thompson group $V(G)$ is Hopfian if and only if $G$ is Hopfian, proving a conjecture of Wu–Wu–Zhao–Zhou. The second result allows us to prove a new embedding result for Hopfian groups: every Hopfian group of type $F_n$ embeds quasi-isometrically as a subgroup of a Hopfian, boundedly acyclic, acyclic group of type $F_n$ that has no proper finite-index subgroups.

comments
25 pages, any comments are welcome
fields
Group Theory
pdf ↗ page arXiv:2610.42050 ↗

We prove the existence of non-sofic Tarski monsters of every sufficiently large prime exponent and of torsion-free non-sofic Tarski monsters. The proof combines a theorem of Sandhu–Thom with quotient theorems of Semenov. These groups are minimal non-sofic, since all their proper subgroups are cyclic. A theorem of Minasyan–Osin–Witzel gives continuum many pairwise non-quasi-isometric torsion-free simple examples. Kulikova's quotient theorem also yields non-simple minimal non-sofic groups of exponent $p^m$ for every $m\geq2$ and every sufficiently large prime $p$. We then study descending chains of non-sofic subgroups in other constructions. We give proper self-embeddings of the binary Leavitt unit group, higher Brin–Thompson groups and the hyperbolic group studied by Douglas and Mghirbi. The explicit Kun–Thom wreath products and doubles are locally graded. These examples are $ω$-non-sofic, and the locally graded ones contain no minimal non-sofic subgroup.

fields
Group Theory
pdf ↗ page arXiv:2610.42140 ↗

Classical Analysis and ODEs8

We prove convex body sparse bounds for coordinatewise Hilbert transforms along monomial curves and local $C^d$ curves with uniformly nonvanishing torsion on the closed integration interval. Adapting the scalar estimates of Cladek–Ou \cite{CladekOu2018} and the lifting method of Laukkarinen–Lorist \cite{LaukkarinenLorist2026} to the anisotropic geometry, we obtain one sparse family of genuine curve-adapted boxes and quantitative $L^p(W)$ bounds for matrix weights. The convex body sparse exponent range is sharp up to its boundary: estimates hold in the interior of $Ω_d$, while any fixed finite number of sparse families fails outside $\overline{Ω_d}$ when $1/r+1/s>1$. We also establish an explicit endpoint-contact obstruction. For every interior planar exponent pair, a polynomial curve with positive torsion on the open integration interval violates the full-interval sparse estimate.

comments
Submitted; Comments are welcome!
fields
Classical Analysis and ODEs, Functional Analysis
pdf ↗ page arXiv:2610.40820 ↗

We revisit the one-dimensional Gagliardo-Nirenberg interpolation inequality $$\|u'\|_{L^p(\mathbb{R})}\leq C\|u\|_{L^q(\mathbb{R})}^{1-a}\|u''\|_{L^r(\mathbb{R})}^{a},\qquad u\in C^\infty_c(\mathbb{R}),\quad \tfrac12\leq a\leq 1,$$ and recall how its general form follows from the case $a=\frac12$ by interpolation with an endpoint estimate. For $a=\frac12$ and $p=q=r$, we give four alternative proofs, based on integration by parts (two proofs), on an ODE, and on the $L^p$ theory of singular integrals. The two proofs by integration by parts also cover most of the other admissible exponents. In $\mathbb{R}^N$, the integration by parts arguments give the inequality with explicit constants and, for $p\leq 2$, with the Laplacian in place of the Hessian and a constant independent of the dimension. For weighted measures $w(x)\,dx$, a pointwise version of the ODE argument gives the pointwise inequality $|\nabla u|^2\leq C_N\,Mu\,M(Δu)$, where $M$ is the Hardy-Littlewood maximal function, and hence the inequality for Muckenhoupt weights, with the Laplacian and an explicit constant. The integration by parts arguments give it for homogeneous weights of large degree and for superharmonic weights. These results apply, in particular, to power weights and to monomial weights.

fields
Classical Analysis and ODEs, Analysis of PDEs, Functional Analysis
pdf ↗ page arXiv:2610.40849 ↗

For two $1$-periodic functions $ψ_1,ψ_2\in L^2(0,1)$ with mean zero we characterize when $\{1\}\cup\{ψ_1(α\cdot x),ψ_2(α\cdot x)\}_{α\in\mathbb{Z}^d,\,α\succ0}$ is a Riesz basis of $L^2([0,1]^d)$, in terms of a $2\times2$ matrix of Dirichlet series, and we compute its optimal bounds. The tool is a $2\times2$ matrix version of the theorem of Hedenmalm, Lindqvist and Seip for systems of dilations with two generators, proved through the reproducing kernel of the Hardy space of the infinite polydisk. In particular, we get the optimal bounds $2/3$ and $3/2$ for $ψ_1,ψ_2$ the triangle waves considered by Daubechies, DeVore, Foucart, Hanin and Petrova in $L^2(0,1)$, and the extended version in $L^2([0,1]^d)$ defined by Schneider and Vybíral.

comments
16 pages, 1 figure
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.41154 ↗

We prove product-weight Pitt inequalities for real or complex functions on $\mathbb R^d$ whose full mixed Vitali variation satisfies an exterior-tail general monotonicity condition. A two-parameter condition gives spatial weights with different powers near the coordinate hyperplanes and at infinity. For the classical parameters, a reverse estimate yields a Boas norm equivalence without sign, parity or separability assumptions. The equivalence holds for finite spatial norm and, for $L^1$ data, characterizes finiteness of either norm; the unrestricted nonintegrable converse remains open. We also prove an equivalence under primitive control, give explicit examples, and establish sharpness of the Fourier-side exponent range.

comments
24 pages
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.41309 ↗

We consider probability measures whose distribution functions solve conjugate equations between two compatible systems of increasing maps on the unit interval. We give sufficient conditions for Ullman–Stahl–Totik regularity in terms of finite compositions of the source maps and a lower mass estimate which holds at Lebesgue almost every point. Our result substantially extends a regularity theorem of Mantica and Totik by allowing nonlinear target maps and replacing their assumptions on convexity, concavity, and displacement near fixed points. The proof uses the mass criterion of Stahl and Totik from logarithmic potential theory. We give further sufficient conditions involving uniform mesh bounds, sets where derivatives equal one, and endpoint drift. Examples distinguish these conditions and show that the uniform version of the lower mass estimate can fail for regular conjugacy measures.

comments
41 pages, 2 figures
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.41422 ↗

Elliptic Fekete configurations minimize logarithmic energy on the sphere. They also provide a natural source of complex polynomials with controlled Shub–Smale condition number. The link between these two problems is the product of the distances from each point to all the other points: logarithmic energy controls the geometric mean of these products, whereas local conditioning depends on each product separately. This paper studies the passage from global information to uniform estimates at individual nodes, exploring its connections with quantitative equidistribution, Riesz energies, and polynomial conditioning. We improve separation bounds for elliptic Fekete configurations and obtain quantitative Riesz-energy estimates for both these configurations and spherical roots of polynomials with optimal-order global conditioning. For $0<s<2$, these estimates also improve explicit upper bounds for the minimal Riesz $s$-energy.

fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.41660 ↗

Let $d\ge2$ and $t>0$. We study the bilinear triangle averaging operator $$\mathcal T_t(f,g)(x)=\int_{\mathcal I}f(x-ty)g(x-tz) dσ(y,z),$$ where $σ$ is the normalized surface measure on the manifold $\mathcal I=\{(y,z)\in\mathbb R^d\times\mathbb R^d:|y|=|z|=|y-z|=1\}$ of equilateral triangles of side length one with a vertex at the origin. We prove the strong-type asymmetric $L^p$-improving estimate $$\mathcal T_1:L^{\frac{d+1}{d-1}}(\mathbb R^d)\times L^{\frac{d^2-1}{d^2-3d+4}}(\mathbb R^d)\longrightarrow L^{\frac{d+1}{2}}(\mathbb R^d)$$ in every dimension $d\ge4$. This enlarges the range of exponents $(p,q,r)$ for which the triangle averaging operator is known to map $L^p(\mathbb R^d)\times L^q(\mathbb R^d)$ to $L^r(\mathbb R^d)$, and symmetrization produces new symmetric estimates in every such dimension as well. The proof combines Sobolev smoothing estimates, obtained through a microlocal decomposition of the triangle incidence relation, with complex interpolation in an analytic family of averages over triangles of varying shape that contains the triangle averaging operator. We also establish maximal estimates for triangle averages. Let $E\subset[1,2]$ be a nonempty set of upper Minkowski dimension $β$, and let $\mathcal T_E(f,g)=\sup_{t\in E}|\mathcal T_t(f,g)|$. We prove that $$\|\mathcal T_E(f,g)\|_2\lesssim\|f\|_{H^{-s_1}}\|g\|_{H^{-s_2}}$$ whenever $s_1,s_2\ge0$, $s_1+s_2<\frac{d-3-β}{2}$ and $d>3+β$. This improves the previously known Sobolev smoothing estimate by half a derivative and lowers the dimension requirement. We also obtain asymmetric localized maximal estimates of the form $\mathcal T_E:L^p(\mathbb R^d)\times L^q(\mathbb R^d)\to L^r(\mathbb R^d)$ in an enlarged range of exponents.

comments
43 pages, 7 figures
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.42061 ↗

For parabolic Ahlfors regular sets, we show that the bilateral weak geometric lemma (BWGL), bilateral approximation by unions of planes (BAUP), and corona decompositions by $Lip$(1,1/2) graphs (whose defining functions need not have half-order time derivatives in BMO) are equivalent. These geometric conditions do not imply the strong geometric lemma. A key tool is a parabolic version of Reifenberg's theorem for sets with holes, which we prove here. As an application of these geometric results, we prove that a parabolic Ahlfors $(n+2)$-regular measure $μ$ on $\mathbb{R}^{d+1}$, $1\leq n\leq d-1$, is parabolic uniformly rectifiable if its support satisfies any of these equivalent conditions and the parabolic Riesz transform is bounded on $L^2(μ)$.

comments
84 pages
fields
Classical Analysis and ODEs
pdf ↗ page arXiv:2610.42142 ↗

Functional Analysis8

This paper studies the strong law of large numbers for products of random uniformly continuous semigroups in Banach spaces. We propose a novel approach by treating the product of semigroups as a solution to an integral equation and establish a version of the SLLN in norm topology for arbitrary Banach spaces.

fields
Functional Analysis
pdf ↗ page arXiv:2610.41346 ↗

The Fock–Bargmann projection \(P_t\) is initially an orthogonal projection on a Gaussian Hilbert space. We characterize the boundedness and compactness of \[ P_t:L_α^{p_1,q_1}\longrightarrow \mathcal F_β^{p_2,q_2} \] and of its associated maximal projection \[ P_t^+:L_α^{p_1,q_1}\longrightarrow L_β^{p_2,q_2} \] for all \(t,α,β>0\) and \(0<p_1,q_1,p_2,q_2\le\infty\). Set \[ Δ(t;α,β):=t^2-2βt+αβ, \qquad d(p,q):=\frac1q-\frac1p. \] For both operators, boundedness holds exactly when \(p_1,q_1\ge1\) and either \(Δ(t;α,β)<0\), or \(Δ(t;α,β)=0\) with \[ q_1\le q_2, \qquad d(p_2,q_2)\le d(p_1,q_1). \] Both operators are compact exactly when \(p_1,q_1\ge1\) and either \(Δ(t;α,β)<0\), or \(Δ(t;α,β)=0\) with both of the above boundary inequalities strict. Thus boundedness forces the source exponents into the Banach range, whereas when \(Δ(t;α,β)<0\), every target exponent is admissible, including quasi-Banach ones. Annular discretization reduces the problem to drifted Gaussian sequence operators, whose comparison with diagonal sequence operators yields the endpoint criteria and quantitative tail estimates. A dilation–multiplier transfer gives corresponding results for Gaussian-twisted Fock operators, including Berezin transforms.

comments
30 pages
fields
Functional Analysis
pdf ↗ page arXiv:2610.41461 ↗

This paper continues the study of conditional symmetric systems initiated in arXiv:2601.15784 and solves several problems posed there. While the interaction between unconditionality and symmetry is well known in the Banach setting following the work in arXiv:2304.05888 and arXiv:2601.15784, we broaden the scope and address similar questions for spreading systems in the more general setting of quasi-Banach spaces. Our aim is twofold. On the one hand, we characterize which spreading systems of Banach spaces are unconditional. On the other hand, we provide examples of quasi-Banach spaces equipped with conditional systems with symmetry-related properties to illustrate the connections between symmetry, greedy-like properties, and the Schauder basis property in the absence of local convexity.

fields
Functional Analysis
pdf ↗ page arXiv:2610.41568 ↗

This paper addresses the expressivity and representation gap in Quantum Neural Networks by establishing a quantum analogue of the classical Kolmogorov–Arnold representation theorem. We propose a constructive approximation framework over fixed input coherent states, where the dependence of the function argument is embedded entirely in coordinate qubits, while the target function dictates a single outer unitary evolution. By applying a Sprecher-type representation to the rotation angle of the state, we demonstrate that the multidimensional operator factorises into \(2d+1\) commuting universal inner observables. We prove that the resulting measurement probability precisely reproduces the multidimensional Bernstein polynomial of the target function. For any continuous multivariate function, the joint quantum state converges uniformly to the target output register, linking classical superposition theory with probabilistic and coherent quantum circuits.

comments
8 pages, 1 figure
fields
Functional Analysis
pdf ↗ page arXiv:2610.41608 ↗

In this paper, we prove that if a Banach space $X$ has the hereditary approximation property, then $X\oplus_2 H$ has the hereditary approximation property for every Hilbert space $H$. This answers a question of Johnson and Szankowski. In addition, if $X$ is also reflexive, then every closed subspace of $X\oplus_2H$ has the metric approximation property. The proof uses Grothendieck's characterization of the approximation property, a factorization of nuclear operators through quotients of $X$, and a nuclear perturbation of the coordinate projection on a closed graph.

comments
17 pages
fields
Functional Analysis
pdf ↗ page arXiv:2610.41758 ↗

As the cornerstone of multivariable operator theory, (essential) Taylor's joint spectrum plays crucial roles in many areas. The present paper is devoted to developing the theory of (essential) Taylor-type joint spectrum of row-bounded sequences of commuting operators on Hilbert spaces and to establishing the associated Fredholm index theory. Furthermore, the holomorphic functional calculus for row-bounded sequences is established. At last, we obtain a Gleason-Richter-Sundberg type index theorem in the infinite-variable setting, which connects the Fredholm index, the asymptotic Arveson's curvature invariant and fibre dimensions.

fields
Functional Analysis
pdf ↗ page arXiv:2610.41882 ↗

In this paper we present results related to Donoho–Stark type uncertainty principles that were proved in the article "The uncertainty principle: variations on a theme" by Wigderson and Wigderson for $k$-Hadamard operators in the finite-dimensional setting. Namely, the authors showed that Donoho-Stark principles can be generalized to versions for approximate supports with respect to norms $\ell^{p}$ for $p=1$ and $p=2$. In the second case they assumed the so-called speciality condition, namely that a suitable matrix satisfies $A^{*}A = kI$. First we show that such theorems cannot be generalized to $p>1$ if matrices are not special $k$-Hadamard. Next we prove Donoho-Stark type principle for special $k$-Hadamard operators for $\ell^{p}$ norms within the range $1\leq p\leq 2$. Finally, we show that such theorems are false for $p>2$ even in the presence of the speciality condition.

fields
Functional Analysis, Information Theory
pdf ↗ page arXiv:2610.41902 ↗

In this article, we study Foguel-type operators $\EuScript{F}=\begin{bmatrix} V_1^{*} & T\\ 0 & V_2\end{bmatrix}$ on $H_1\oplus H_2$, where each $H_j$ is a separable Hilbert space, $V_j\in \mathcal{B}(H_j)$ is an isometry for $j=1,2$, and $T\in \mathcal{B}(H_2,H_1)$. We examine how the norm and minimum attainment properties of $\EuScript{F}$ depend on the corner entry $T$ and on the defect spaces of the isometries. The norm formula $\|\EuScript{F}\|=\frac{\|T\|+\sqrt{\|T\|^{2}+4}}{2}$ holds for every $T$ when $V_1,V_2$ are isometries, and more generally when $V_2$ is a contraction that is isometric on the range of $T^{*}$. Moreover, $\EuScript{F}$ attains its norm if and only if $T$ does. We characterize the positive $2\times 2$ block operator matrices that are absolutely norm attaining and absolutely minimum attaining, and use this to prove that $\EuScript{F}$ is absolutely norm attaining if and only if it is absolutely minimum attaining if and only if $\dim N(V_1^{*})<\infty$ and $T$ has finite rank. The closure of the absolutely norm attaining class is described by $\dim N(V_1^{*})<\infty$ and compactness of $T$. Further, we establish that the reduced minimum modulus is $γ(\EuScript{F})=\frac{\sqrt{\|T\|^{2}+4}-\|T\|}{2}=\frac{1}{\|\EuScript{F}\|}$, and $\EuScript{F}$ attains it if and only if $T$ is norm attaining. We also show that $\EuScript{F}$ is Fredholm if and only if $\dim N(V_j^{*})<\infty$ for $j=1,2$, with index $\dim N(V_1^{*})-\dim N(V_2^{*})$, and we give its Moore–Penrose inverse. Finally, for $V_1=V_2=T_z$ and more generally $V_j=T_{θ_j}$ with $θ_j$ inner, and Toeplitz or Hankel corner entries, the results become explicit conditions on the symbol.

comments
26 pages. Submitted to a journal. Comments and suggestions are welcome
fields
Functional Analysis
pdf ↗ page arXiv:2610.42102 ↗

General Mathematics8

The complete-lattice tensor product $\mathbb{E}=[0,1]\otimes[0,1]$ arises naturally in the algebra of truth values for type-2 fuzzy sets. For a monotone surjective map $S$ that is continuous on $[0,1[^2$ and a monotone map $T$ having $1$ as identity, we consider the generalized supremum convolution \[ (f\otimes_{_{S,T}}g)(x)=\sup_{S(u,v)=x}T(f(u),g(v)) \] on $\mathbb{E}$. Rather than assuming classical algebraic structures for $S$ and $T$ from the outset, we ask when they are forced by requiring the induced convolution itself to be a t-conorm. We prove that $\otimes_{_{S,T}}$ is a t-conorm on $\mathbb{E}$ if and only if $S$ is a continuous t-conorm and either $S=\max$ or $T$ is a right-continuous t-norm; in the exceptional case $S=\max$, the convolution is simply the pointwise maximum for every admissible $T$. The same convolution mechanism also yields an order-dual description of triangle functions on the space of distance distribution functions $Δ^+$. Using an order-reversing isomorphism between $Δ^+$ and a boundary variant of $\mathbb{E}$, we show that $τ_{_{T^*,L}}$ is a triangle function precisely when $L$ is a continuous t-conorm with no zero divisor and, unless $L=\max$, $T^*$ is a left-continuous t-conorm; moreover, when $L\neq\max$, continuity of $τ_{_{T^*,L}}$ is equivalent to continuity of $T^*$. This resolves the remaining $τ_{_{T^*,L}}$ part of the Schweizer–Sklar problem.

comments
19 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.40659 ↗

We define a binary operation $T_1\star T_2$ on the fuzzy unit interval $\I$ by extending the natural action of $T_1$ and $T_2$ on step functions. When both $T_1$ and $T_2$ are left-continuous, this operation coincides with the tensor product of the corresponding quantales and therefore yields a t-norm on $\I$. Without left-continuity, the situation becomes more subtle. We prove that if at least one of the two t-norms is continuous, then $T_1\star T_2$ is still a t-norm on $\I$. We also show by counterexample that continuity cannot simply be dropped. The operation $\DN$ fails to be associative. On the other hand, $T_D\star T_D$ is a t-norm despite $T_D$ being neither continuous nor left-continuous, indicating that continuity is not a necessary condition.

comments
16 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.40678 ↗

We prove that the largest absolute determinant of a $15\times15$ matrix with entries in $\{1,ω,ω^2\}$, where $ω^3=1$ and $ω\ne1$, is $120932352\sqrt{19}$. A matrix attaining this value was already known; the issue is maximality. The proof follows the Gram-matrix search method used in the real maximal-determinant problem. A sharp determinant bound in terms of the off-diagonal Gram energy, combined with Eisenstein congruences and a published classification of difference matrices, reduces a strict improvement to 23 possible energies. Componentwise Schur bounds and exact support catalogues then leave a small number of cases, excluded by orthogonality, rank, and spectral-intertwining conditions. We present the finite tests in a common framework and supply their replay programs. For the displayed extremizer we give a dephased representative, its Gram matrix and spectrum, and an exact computation of its monomial automorphism group, which has order 1008. The proof determines the value of the maximum, but does not classify all equality cases.

comments
19 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.40679 ↗

In this note, we address a traditional pedagogical asymmetry between algorithmic differentiation and heuristic integration by introducing an extra parametric-order differential framework. By treating the order of differentiation as a parametric axis (discrete or continuous), classical integration is reframed as a local differential evaluation at the non-positive boundary level. This analytical shift alleviates some traditional routines of integration by parts, by replacing the quadrature of a product of elementary functions with a direct differential series that either truncates into a finite algebraic sum or resolves through cyclic symmetries. This parametric-order approach also bridges the gap with fractional calculus, showing that non-integer derivatives emerge as a seamless extension of the same operational dimension. By reducing global quadrature to local differentiation, this note offers an efficient, conceptual tool for undergraduate analysis.

comments
5 pages. To appear as a Note in undergraduate/graduate analysis. Includes applications to fractional calculus
fields
General Mathematics
pdf ↗ page arXiv:2610.40681 ↗

Power series expansions in the polar distance about the center are derived for the normal direction and signed distance from an ellipse to points near its center. After extracting constant terms and common factors, the series coefficients are polynomials in the axis ratio and the absolute value of the sine of the polar angle, with rational coefficients. The region of convergence is shown to be the region bounded by the ellipse evolute. The expansions complement previously derived expansions valid for points far from the center.

comments
41 pages, 1 figure, 6 tables. Part II of arXiv:2507.08807. Code: https://github.com/jnil02/point2ellipse_series
fields
General Mathematics
pdf ↗ page arXiv:2610.40690 ↗

We show that $\frac1{g(x)}$, with $g(x)=\frac{x-1}{x\ln x}$, is a complete Bernstein function and give its Lévy measure explicitly. For $h(x)=\frac{(x^2-1)\ln(x+1)}{x^2\ln x}$, we prove that $h$ is not a Stieltjes function, so $\frac1h$ is not a complete Bernstein function, but $\frac1h$ is a Bernstein function. The latter rests on the positivity of a function built from the convolution of the exponential integrals $E_1$ and $E_2$. We also give a related completely monotonic function that is not a Stieltjes function.

comments
13 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.40718 ↗

Let $p(n)$ be the minimum cardinality of a finite planar point set in which each point has exactly $n$ unit-distance neighbours. Harborth proved $p(5)=18$ in work presented in 1985 and published in 1987, using a detailed case analysis. We revisit the proof using convex-hull arguments, incidence counts, and elementary properties of intersections of unit circles. These tools give an alternative treatment of configurations with $14$ and $16$ points and organize the final case distinction by a common vector-sum structure. We also treat a configuration not explicitly covered by the case analysis in the online PDF copy of Harborth's paper available to us, and give an explicit justification of the six-point classification used in the argument. No result subsequent to Harborth's proof is required.

comments
16 pages, 3 figures
fields
General Mathematics
pdf ↗ page arXiv:2610.40725 ↗

We consider special cases of the discontinuous Weber–Schafheitlin integral in which the orders $μ$ and $ν$ of the Bessel functions differ by an integer and the exponent $λ$ equals $0$ or $\pm1$, with $μ-ν+λ$ odd. Explicit closed-form results are obtained in terms of Jacobi polynomials.

comments
7 pages
fields
General Mathematics
pdf ↗ page arXiv:2610.40730 ↗

Commutative Algebra6

We prove that a semidomain has length-factorial additive and multiplicative monoids if and only if it is isomorphic to the semiring of nonnegative integers or is a unique factorization domain. Consequently, bi-length-factoriality and bi-factoriality coincide for all semidomains, settling the Bi-LF Positive Conjecture. The main step shows that additive length-factoriality and multiplicative factoriality force additive factoriality, without an additive reducedness assumption. Multiplicative factoriality forces every semidomain with a nonzero additive unit to be a unique factorization domain. In the additively reduced case, a uniform multiplier for lattice points in a finitely generated cone yields cancellation of divisibility across sums, which leads to an obstruction to additive atoms dividing positive integers and completes the structural argument. We also consider related questions on quasi-factoriality and classify the rational and transcendental monogenic positive semirings whose multiplicative monoids are quasi-$n$-factorial for a prescribed integer $n \ge 2$.

comments
16 pages
fields
Commutative Algebra
pdf ↗ page arXiv:2610.40688 ↗

A finitely generated module has unbalanced support if its support has positive dimension but meets a top-dimensional component of the spectrum only at the closed point. We ask what homological signals are left behind by this geometry. Separating the spectrum into two wings, we prove first Tor nonvanishing for positive-depth modules and give criteria for second Tor nonvanishing. In depth zero, we obtain a structural sequence describing the finite-length torsion under a first Tor vanishing. A recent counterexample to the Peskine–Szpiro dimension inequality conjecture establishes the existence of modules of finite projective dimension and unbalanced support \cite{Ma}. We apply our framework to these modules and establish a numerical torsion–spine balance for linked Cohen–Macaulay wings. We also construct such modules over a reduced Gorenstein ring with exactly two irreducible components.

fields
Commutative Algebra
pdf ↗ page arXiv:2610.40854 ↗

We compute top derived products and inverse limits on products of affine lines with doubled origin over $B[s]$, for any nonzero commutative ring $B$. For every $B$-module $M$, the two top obstruction modules fit into an explicit exact sequence and contain $M^{\mathbb N}/M^{(\mathbb N)}$as a natural $B$-module direct summand.This gives exact product dimensions, sharp vanishing bounds for inverse sequences, and faithfulness in the coefficient module. For a Grothendieck category of finite product dimension, we also obtain degree shifts under adjoining commuting locally nilpotent endomorphisms and prove that this construction preserves top-degree vanishing for epimorphic inverse sequences. A fixed Gabriel–Popescu presentation then describes the general top obstruction through actual Ext sequences and separates two module-theoretic terms.

fields
Commutative Algebra
pdf ↗ page arXiv:2610.41117 ↗

Let $M$ be a finitely generated module over a ring $R$ that is a finitely generated algebra over a Noetherian ring $A$ by $r$ elements. We prove that there are finitely many primes $P_1, \ldots, P_k$ of $A$ such that $M$ has a countable ascending filtration by finitely generated $A$-modules with the property that every factor $F_n$ has a filtration by cyclic modules of the form $A/P_i$, $1 \leq i \leq k$ such that the number of copies $H_i(n)$ of $A/P_i$ that occur in the filtration of $F_n$ is an eventually polynomial function of $n$. We prove a number of bounds on the degrees of these eventually polynomial functions: $r-1$ is always a bound, and the dimension of $\dim(M_P)-1$ is a bound under mild hypotheses. One application is an immediate proof Grothedieck's classic theorem on generic freeness. Several other applications are given.

fields
Commutative Algebra
pdf ↗ page arXiv:2610.41200 ↗

In this paper, we introduce and study the notion of 2-nil hyperideals of multiplicative hyperring.

fields
Commutative Algebra
pdf ↗ page arXiv:2610.41443 ↗

The reduction number of the maximal ideal of a localization (or even faithfully flat extension) need not increase under specialization. Nevertheless, for each fixed local ring the reduction numbers of the maximal ideals of all its localization have a finite common bound.

fields
Commutative Algebra
pdf ↗ page arXiv:2610.41659 ↗

Complex Variables6

For a complex polynomial $P$, its full shadow is the closure of the set of zeros of all derivatives of all powers $P^n$, and its limiting core is the Hausdorff limit, as $n\to\infty$, of the set of zeros of all derivatives of $P^n$. These sets retain geometric information that weak limits of root-counting measures discard. We determine complete zero-set limits for arrays of polynomials whose roots approach finitely many locations. A cluster containing $o(D)$ roots of a degree-$D$ polynomial can survive outside the limiting support of the derivative zeros, which depends only on the positive limiting masses: under repeated differentiation, the zeros it produces follow a straight path, selected by an explicit min–max condition, and preserve the cluster's exact multiplicity. When such a location instead carries a small positive mass $\varepsilon$, the support of the limiting zero distribution at a fixed ratio of derivative order to degree develops an arc on the $\sqrt{\varepsilon}$ scale with a universal local semicircle profile. We use these mechanisms to prove that polynomial shadows can have robust and arbitrarily complicated topology. For every $k\ge2$, there are polynomials with $k+2$ distinct root locations whose limiting cores and full shadows have at least $\binom{k}{2}$ holes; this persists under coefficient perturbations and among square-free polynomials without a reflection axis. In a four-location family, the core and full-shadow areas tend to zero while one hole has area tending to $8/1221$.

comments
38 pages, 7 figures
fields
Complex Variables, Classical Analysis and ODEs
pdf ↗ page arXiv:2610.40842 ↗

We study the class $CC_α$ of completely convex functions of order $α$, defined through a two-point analogous of the classical convexity condition. We give several equivalent characterizations of $CC_α$, together with representation formulas in terms of two auxiliary Schwarz functions linked by a compatibility equation. From these representations, we derive coefficient bounds, a subordination result $zf'/f$, and a two-point Taylor expansion in which the Schwarzian derivative $S_f$ arises naturally as the coefficient of the linear term. This yields a pointwise bound for $S_f$ and a bound for the pre-Schwarzian norm $\|P_f\|_{CC_α}$, improving the classical Yamashita bound when $α\leq 1-1/\sqrt2 $. We also show that, unlike the class $C_α$ of convex functions of order $α$, every $f\in CC_α$ is bounded when $α>0$.

fields
Complex Variables
pdf ↗ page arXiv:2610.40889 ↗

We study reciprocal moments of the critical points of a polynomial with a simple zero at the origin. A positive similarity transforms a reciprocal derivative companion into a complex symmetric matrix. This gives a short proof of the reciprocal Schoenberg inequality and a refinement for every elementary symmetric function. We determine the optimal constants in the corresponding inequalities for every positive exponent, including the sublinear range, and extend the results to arbitrary positive charges. We also characterize the normal powers of the companion and of its symmetric representative. The resulting criterion gives the complete equality cases in two quartic inequalities and explains the appearance of noncollinear equality configurations at higher orders. We also obtain a closed generating determinant and a quantitative collinearity estimate.

comments
12 pages
fields
Complex Variables, Combinatorics, Rings and Algebras
pdf ↗ page arXiv:2610.41325 ↗

We study the truncated Dirichlet functional \[ Φ_n (f) = \sum_{k = 1}^n k |a_k|^2, \quad f(z) = \sum_{k = 0}^\infty a_k z^k, \] when $f$ belongs to the Bloch space. If $n \geq 27$, we find a normalized Bloch function $f_n$ such that \[ Φ_n (f_n) > n B_n^2, \] where $B_n = \max \{ |a_n| : \| f \|_{\mathcal{B}} = 1 \}$, thus proving that an inequality by Kayumov and Wirths cannot be extended to large values of $n$. Nevertheless, we show that the inequality $Φ_n(f) \leq n B_n^2$ holds for all normalized functions when $n \leq 9$.

fields
Complex Variables, Classical Analysis and ODEs, Functional Analysis
pdf ↗ page arXiv:2610.41534 ↗

Let $G_0$ be a real form of a complex semisimple group acting on a flag manifold $Z$, with maximal compact subgroup $K_0$ and its complexification $K$. Matsuki duality pairs each $G_0$-orbit $γ$ with a $K$-orbit $κ$, and $M=κ\capγ$ is a compact $K_0$-orbit. For lower-dimensional $γ$, $M$ is a compact CR manifold, and we ask which functions are holomorphic near $M$ and what its holomorphic hull is. In general, we show that $K_0$-finite holomorphic functions near $M$ are regular functions on $K$-orbits, and that $M$ carries non-constant holomorphic germs if some $K$-stable prime divisor of $Z$ misses $κ$, and, under a connectedness hypothesis, only constants otherwise. For $G_0=SU(p,q)$ acting on any Grassmannian $\mathrm{Gr}_k(\mathbb{C}^{p+q})$ we give a complete answer. Holomorphic functions near $M$ are constant unless $M$ is of holomorphic type: $k\in\{p,q\}$ and $M$ lies in the Harish-Chandra chart of the bounded symmetric domain $\mathcal{D}$. Then $M$ is the manifold of rank-$c$ partial isometries, and a holomorphic hull of $M$ is $\overline{\mathcal{D}}\cap\{\mathrm{rank}\le c\}$. This agrees with the polynomial hull of Kaup and Zaitsev except in one totally real case, where it is strictly smaller because $\det^{-1}$ is holomorphic near $U(p)$. The same dichotomy decides when global CR functions on $M$ are constant and when holomorphic sections of line bundles near $M$ are global; its exceptions differ from the Gindikin-Matsuki exceptions for cycle spaces, except at the base point of $\mathcal{D}$. This settles the descent question left open in earlier work with K. Shabbir on rank-one orbits. We also determine which $K$-orbits carry non-constant regular functions, prove constancy via chains of rational curves for small $c$, and compute Levi forms in examples.

fields
Complex Variables, Representation Theory
pdf ↗ page arXiv:2610.41828 ↗

We develop a 4-dimensional residue calculus for quaternionic slice semiregular functions. Its local singularities are isolated real poles and 2-dimensional spheres. For the latter, we introduce a spherical residue and characterize it as a spherical Laurent coefficient, as the sum of two classical residues on any full complex slice, and by a finite differential procedure involving alternating spherical and slice derivatives. The residue theorem expresses a weighted boundary flux on a circular domain as the sum of the real and spherical residues it contains. Beyond this formula, polynomially weighted flux moments recover every coefficient of every finite principal part, including the companion coefficient at a spherical pole, invisible to ordinary residue summation. Once the polar components and order bounds are fixed, the quotient of the real polynomial ring by the global pole polynomial is the finite-dimensional algebra of weights. We prove that the residue pairing is well defined and nondegenerate, and identifies formal principal parts with the full quaternion-valued real dual of this algebra. Hermite interpolation and the Chinese remainder theorem provide global weights selecting each local coefficient. We separate the regularity domain of the function from the geometry of integration: for a smooth noncircular integration domain compactly contained in a circular ambient domain, away from the reals and with boundary disjoint from the polar set, we establish an exact residue formula with an explicit volume correction governed by the Cauchy-Fueter operator. The theory includes higher-order poles, exceptional points, product domains, and zero divisors. Applications give formulas for rational and oscillatory integrals on the imaginary hyperplane and characterize the rational data on the 3-dimensional imaginary subspace of the quaternions admitting a canonical slice regular extension.

comments
106 pages, 12 figures. The abstract in the PDF file is slightly different because it is too long for the arXiv web format
fields
Complex Variables
pdf ↗ page arXiv:2610.41869 ↗

Quantum Algebra6

We derive monopole-fermion scattering correlators from conformal boundary states in twisted sectors. The key ingredient is that symmetry-preserving boundary conditions contain sectors twisted by topological symmetry lines. Their overlaps determine correlators of operators attached to these lines, including the outgoing excitations required by charge conservation in the Callan-Rubakov effect. We construct the twisted boundary states of $N$ free Dirac fermions with $U(1)^N$-preserving boundary conditions, extending the familiar description in terms of charge lattices to shifted lattices describing twisted sectors. Using cluster decomposition and open-closed duality, we determine their coefficients and fix their normalization relative to the untwisted boundary state. We find that the Callan-Rubakov boundary condition yields a two-point function between an incoming fermion and an outgoing defect operator with a coefficient of unit modulus, even when the corresponding correlators of ordinary local fermions vanish. Finally, we relate these correlators to four-dimensional partial-wave scattering using gauge-covariant asymptotic states. We identify the missing $s$-wave contribution to the unitarity relation and argue that the twisted channel supplies it with precisely the required normalization.

comments
42 pages, 8 figures
fields
Quantum Algebra
pdf ↗ page arXiv:2610.40779 ↗

Let $V$ be a vertex operator algebra (VOA), not necessarily rational or $C_2$-cofinite, such that the category $\mathcal C_1(V)$ of $C_1$-cofinite grading-restricted generalized $V$-modules is semisimple and contains the generalized Verma modules associated to finite-dimensional modules over the Zhu algebra $A(V)$. Such VOAs include affine VOAs at admissible and generic levels. We prove that the bottom degree of Huang-Lepowsky's $P(w)$-tensor product module $M^1\boxtimes_{P(w)}M^2$ is isomorphic to the left $A(V)$-module $M^1\odot_{w,0}M^2$ arising from the restricted conformal blocks on $(¶^1,\infty,w,0)$. By shifting the marked points and using the parallel transport of the restricted conformal blocks, we construct an explicit braiding isomorphism on the category of finite-dimensional $A(V)$-modules, which is the restriction of the braiding isomorphism of Huang-Lepowsky-Zhang to the bottom degrees. For the affine VOA $L_k(\mathfrak g)$ at a positive integral level, it is given by the Casimir tensor of $\mathfrak g$ on certain quotients of tensor products of finite-dimensional $\mathfrak g$-modules, and at generic levels it recovers Drinfeld's commutativity isomorphism.

fields
Quantum Algebra
pdf ↗ page arXiv:2610.41605 ↗

We study an ``odd'' version of the Grothendieck–Teichmüller Lie algebra that is obtained by replacing free Lie algebras in even generators by free graded Lie algebras in odd generators in the classical definition. These odd variants are related to the ``odd'' Kontsevich graph complex in the same way as the classical Grothendieck–Teichmüller group is related to the even Kontsevich graph complex. We show an odd analog of the Deligne–Drinfeld conjecture, namely that the odd Grothendieck–Teichmüller Lie algebra is isomorphic to a free graded Lie algebra in one odd generator of weight 3. Thus we in particular compute the odd graph cohomology in the bottom cohomological degree, showing that it is mostly zero.

fields
Quantum Algebra, Algebraic Topology
pdf ↗ page arXiv:2610.41766 ↗

We consider noncommutative analogues of associated fibre bundles where the total space, base space and even the fiber may be nonaffine, that is, not represented by a single algebra, and where the structure group is replaced by a Hopf algebra $H$. We choose to describe gluing of affine pieces by corings. In the role of a principal bundle we take an internal $E$-coring $M$ in the monoidal category of right $H$-comodules where $E$ is a faithfully flat right Hopf-Galois extension. In a role of a fibre we either take a left $H$-comodule algebra $F$, or more generally an internal $F$-coring $N$ in the monoidal category of left $H$-comodules. Cotensor products combining these data is used to obtain an ordinary coring representing the total space of an associated bundle. If the fiber coring $N$ is equipped with a choice of an $H$-coinvariant grouplike element, then we exhibit a canonical morphisms of corings from the $H$-coinvariant coring of $M$ to the associated bundle coring which is viewed as a dual to the projection of an associated bundle.

comments
9 pages
fields
Quantum Algebra, Rings and Algebras
pdf ↗ page arXiv:2610.41774 ↗

We propose a program to classify tensor categories generated by prescribed seed tensor categories. We carry out this program for finitely many marked copies of the positive Fibonacci category with pairwise nonisomorphic generating simple objects, in the setting of split locally finite semisimple rigid $\CC$-linear tensor categories with simple unit. We show that the Fibonacci fusion relation forces the fusion rules to be those of a specialized Coxeter–Hecke ring. The systems admitting categorification have Coxeter exponents $2,3,\infty$. For every such system we construct a tensor category and prove uniqueness up to marked tensor equivalence. Its simple objects are indexed by the Coxeter group, which may be infinite. Finite rank occurs precisely for disjoint unions of finite ADE diagrams, giving the ADE classification of Fibonacci-generated fusion categories. These categories are unitarizable, with no unitary structure assumed on the candidate. In finite ADE type, allowing Yang–Lee seeds gives one independent tensor-branch choice on each connected component.

fields
Quantum Algebra
pdf ↗ page arXiv:2610.41811 ↗

We classify finite composites of the positive Fibonacci category with arbitrary pointed fusion categories, retaining the entire source functor and making no unitarity assumption on the target. Formal codegrees, dimension fields, and algebraic integrality exclude all dihedral Fibonacci rings $R_n$ with $n\ge4$ from categorification. Together with an independent graphical uniqueness argument, this gives the finite two-Fibonacci classification. For several marked Fibonacci generators, a geometric height-and-polygon argument identifies the based ring with a standard-basis Coxeter–Hecke ring. Transitive conjugation by the pointed image forces the noncommutation graph to be empty or a perfect matching. We then obtain a crossed-product classification in terms of a quotient group, a stabilizer, labelled Fibonacci or rank-six $\TT$ blocks, and compatible cocycle and categorical action data. This includes the Fibonacci--$\Z/3\Z$ composite problem without target unitarity. We also establish the corresponding forms and marked descent statements over characteristic-zero fields admitting the chosen Fibonacci-type input.

fields
Quantum Algebra
pdf ↗ page arXiv:2610.41891 ↗

Category Theory5

Let $(\V, \otimes, I,[-,-])$ be a closed symmetric monoidal category with finite (co)limits. Assume that $\V$ is weakly idempotent complete and additive. For a given small $\V$-category $\A$, we investigate sufficient conditions on $\A$ under which the ordinary category $[\A, \V]_0$ of $\V$-functors from $\A$ to $\V$, equipped with the exact structure $\mathcal{E}_{\textrm{lowt}}$, is a Frobenius exact category,and hence, induces a triangulated stable category. The exact structure $\mathcal{E}_{\textrm{lowt}}$ encompasses the degre-wise trivial exact structure on categories of $N$-chain complexes and $m$-periodic chain complexes, with terms in a variety of settings, including left $R$-modules, quasi-coherent sheaves on a scheme, and Banach spaces. We subsequently address the question of describing abstract left and right homotopy relations in $([\A, \V]_0;\mathcal{E}_{\textrm{lowt}})$ in terms of families of morphisms defined objectwise, as in the cases of chain-homotopy and relative homotopy relations. To this end, we introduce the notion of left/right-weak Serre $\V$-functors on $\A$, and investigate their relationships with the classical Serre functor relative to a commutative ring $k$.

fields
Category Theory, K-Theory and Homology
pdf ↗ page arXiv:2610.40746 ↗

Let $\mathcal{C}$ be a small category. We characterize the full initial subcategories of $\mathcal{C}$ as precisely those whose representables give a free presentation of the constant functor $Δ_{\mathcal{C}}\mathbb{K}:\mathcal{C}\to\mathrm{Vect}$. Such a presentation yields a formula for the limit of any functor $\mathcal{C} \to \mathrm{Vect}$ as the kernel of a morphism between products of evaluations. When the linearized endomorphism ring of every object is local, we show that a minimal free presentation, whenever one exists, determines the unique minimal full initial subcategory, up to equivalence. Bass's characterization of perfect rings gives a sufficient condition for the existence of such presentations. In particular, every finite category with no nonidentity endomorphisms, such as a finite poset, contains a unique minimal initial full subcategory. In the poset case, our results recover the minimal initial subposet constructed in [BDL27] and [DL26]. When $\mathcal{C}$ is the free category on a finite quiver, we give an explicit construction of its minimal initial full subcategories, show that every such subcategory arises from this construction, and provide algorithms with time-complexity bounds.

comments
27 pages, 0 figures
fields
Category Theory, Representation Theory
pdf ↗ page arXiv:2610.41068 ↗

Many kinds of generalised syntax, including internal categories, T-categories or polynomial monads, arise as monads in bicategories of a certain flavour: maps are some kind of span, and composition is some kind of pullback. We explain the pattern behind these similarities: each bicategory is stitched together from two simpler pieces by a distributive law, where the distributivity is governed by pullback. A key novelty is in the kind of distributive law we consider; where Rosebrugh and Wood considered distributive laws between categories, we instead consider pseudo-distributive laws between double categories. A key construction, drawing on work of Weber, obtains such a distributive law from a so-called co-crossed double category: one whose codomain functor is a fibration, and whose cartesian cells are stable under horizontal composition. We show the horizontal Kleisli double category of a monad T is co-crossed just when T is cartesian, and that the composite double category recovers T-spans; that the double category associated to an identity-on-objects functor is co-crossed just when that functor is a calibration in the sense of Bénabou; and that Street's calibrated bicategories yield co-crossed double categories in a similar manner, recovering the double category of polynomials of Gambino–Kock.

comments
29 pages
fields
Category Theory
pdf ↗ page arXiv:2610.41551 ↗

We extend the theory of lax-idempotent monads to the setting of $(\infty,2)$-categories. In particular, we prove an analogue of Kock and Zöberlein's characterization of the category of algebras over a lax-idempotent monad. We then apply this theory to many examples that arise naturally in homotopy theory. For example, we show that $(\infty,1)$-toposes are monadic over presentable $(\infty,1)$-categories, with left adjoint given by the symmetric $(\infty,1)$-topos. We also show that $\infty$-operads are comonadic over symmetric monoidal $(\infty,1)$-categories through the monoidal envelope construction, obtaining along the way a new proof of Barkan-Steinebrunner's result that the envelope functor is a faithful inclusion. We furthermore use this theory to simplify Ramzi's proof that dualizable stable $(\infty,1)$-categories are comonadic over $\aleph_1$-compactly generated stable $(\infty,1)$-categories.

comments
29 pages. Comments welcome!
fields
Category Theory
pdf ↗ page arXiv:2610.41904 ↗

We extend dependent type theory by type classifiers and construct an interpretation in the category of contextual categories with families. Type classifiers extend contexts by type variables, complementing the usual context extensions by term variables of specified types. This allows statements with arbitrary type parameters to be formulated without assuming a universe. We show that substituting isomorphic types induces isomorphisms between the resulting types even in extensional type theory, giving an analogue of the invariance supplied by the univalence axiom in homotopy type theory. Omitting function types from the theory, so that all type formers are covariant, extends this naturality to arbitrary, possibly noninvertible maps.

comments
7 pages
fields
Category Theory
pdf ↗ page arXiv:2610.41941 ↗

Metric Geometry5

We study a Poincaré-type inequality for Gaussian measures restricted to origin-symmetric convex sets. The inequality arises from the second variation at $t=0$ of the function $t\longmapsto γ_Σ\left(e^{tA}K\right)$, where $Σ$ and $A$ are symmetric matrices. The relevance of this function comes from Saroglou's reformulation of the log-Brunn–Minkowski conjecture. We show that $Σ$ and $A$ commute if and only if the inequality holds for every box of the form $UD B_\infty^n$, where $U$ is orthogonal and $D$ is diagonal. In the non-commuting case, this yields counterexamples for orthogonal boxes whose axes can be chosen arbitrarily close to the coordinate axes. We also show that the inequality holds when $K$ is unconditional with respect to an eigenbasis of $Σ$ and $A$ is an arbitrary symmetric matrix.

fields
Metric Geometry, Probability
pdf ↗ page arXiv:2610.07248 ↗

The paper investigates the Gromov–Hausdorff distance using a new tool, the so-called $\varkappa$-traces of metric spaces. A general lower bound for the Gromov–Hausdorff distance is obtained in terms of $\varkappa$-traces. Using this bound, exact estimates for the distance between an arbitrary metric space and its Hausdorff hyperspace are derived, and the behavior of the metric on transfinite towers (inductive limits) of hyperspaces is studied. In the final part of the paper, the trace machinery is applied to $l^p$-products of metric spaces. A connection is established between computing the Gromov–Hausdorff distance for infinite $l^\infty$-products and partition calculus. Using methods from Ramsey theory (in particular, the Erdos–Rado theorem), $\varkappa$-traces for $l^\infty$-products are computed, and sufficient conditions are obtained under which the Gromov–Hausdorff distance between infinite Cartesian products can be computed exactly. The final section also provides a general estimate for the Gromov–Hausdorff distance between $l^\infty$-products of metric spaces using the Erdos–Rado theorem.

comments
23 pages
fields
Metric Geometry
pdf ↗ page arXiv:2610.40873 ↗

We prove, in every dimension, the upper bound conjectured by Betke, Henk and Wills for the lattice-point enumerator in terms of volume and successive minima. For every convex body $K\subset R^n$ and full-rank lattice $Λ$, we establish $$ \#(K\capΛ)\le\frac{\mathrm{vol}_n(K)}{\detΛ}\prod_{i=1}^n\left(1+\frac{iλ_i}{2}\right), $$ where $λ_i$ are the successive minima of $(K-K)/2$ with respect to $Λ$. Equality holds precisely for lattice translates of positive integral dilates of unimodular simplices. We also prove $$ \#(\mathrm{int}K\capΛ)\ge\frac{\mathrm{vol}_n(K)}{\detΛ} \left(1-\frac{n}{4}\sum_{i=1}^{n-1}λ_i-\frac{n}{2}λ_n\right), $$ which implies the lower estimate with factor $1-\frac{1}{2}\sum_{i=1}^n iλ_i$. In dimension two we prove the lower bound with factor $(1-λ_1/2)(1-λ_2)$ when $λ_2\le1$. We also observe that a conjectured product lower bound for origin-symmetric bodies fails in dimension three. The proofs combine geometric compression, lattice-cell counts in coordinate sections, and an elementary coefficient comparison.

comments
18 pages, 1 figure
fields
Metric Geometry, Combinatorics
pdf ↗ page arXiv:2610.41125 ↗

Let $K \subset\mathbb{R}^d$, $d\ge2$, be a connected compact set and $φ(t)=tL(t)$ be a gauge function with $L$ non-increasing. We study intersections of $K$ with rectifiable curves and chord-arc curves. We show that if $\int_0^1\frac{dt}{φ(t)}<\infty$, then there exists a rectifiable curve $γ\subset\mathbb{R}^d$ such that $\mathcal H^φ(K \capγ)>0$. Moreover, for every $0<s<1$, there exists a chord-arc curve $γ$ such that $\textrm{H.dim }(K \capγ)\ge s$. Conversely, we show that if $K$ is uniformly porous on lines and $\int_0^1\frac{dt}{φ(t)}=\infty$, then $\mathcal H^φ(K \capγ)=0$ for every rectifiable curve $γ$ and $\textrm{H.dim }(K \capγ)<1$ for every chord-arc curve $γ$.

comments
15 pages
fields
Metric Geometry
pdf ↗ page arXiv:2610.41510 ↗

We show that any $n$-point Euclidean metric admits a non-expansive embedding into the line with average distortion at most $O\left(\sqrt{\frac{\log n}{\log\log n}}\right)$, matching a known lower bound. We also present two applications of this result. First, we give a cut-player strategy with an optimal spectral guarantee for the undirected cut-matching game. Second, we give an improved spectral approximation for vertex expansion in high-degree graphs.

fields
Metric Geometry
pdf ↗ page arXiv:2610.41945 ↗

Logic3

Many explicit constructions in algebra are not first-order definable in classical finitary logic, but are, however, definable in infinitary logic. The aim of this paper is to show how infinitary logic can be used to lift model-theoretic results in arithmetic to a quantifier-sensitive form. We apply this idea to a deep result of Jahnke and Kartas on the model theory of perfectoid fields. Our main result is that a first-order sentence $φ$ holds in the tilt $k^{\flat}$ of a perfectoid field $k$ if and only if a certain sentence $φ^{\natural}$ holds in a non-principal ultrapower of $k$, where $φ^{\natural}$ does not depend on $k$ and is only slightly more complicated than $φ$ itself. In particular, tilting preserves the Lévy hierarchy in a precise sense. To this end, we prove that the Standard Decomposition of a valued field at a constant $t$ is infinitarily interpretable by formulas of rank at most 1.

comments
13 pages
fields
Logic, Number Theory
pdf ↗ page arXiv:2610.41064 ↗

Let $K$ be a field which is strongly stable, that is, stable and strongly dependent. Then $K$ is finite or algebraically closed. This answers a question of Shelah.

comments
6 pages
fields
Logic
pdf ↗ page arXiv:2610.41430 ↗

We prove that KP, with Full Set Induction, is neither solid nor tight. More generally, every consistent theory obtained from KPC by adding a quantifier bounded set of sentences has a model with a parameter-free definable class surjection from its natural numbers onto its universe. Using such enumerations to define generic extensions by relative interpretations, we show that theories with the stable definable ground property are not tight. Ground definability over MOST and continuum coding over ZFC^{Π_3} yield applications to weak inductive set theories with large cardinals and forcing axioms. Finally, solidity for quantifier bounded extensions of MOST with Full Set Induction implies a Ground Axiom formulated using a fixed finite level of genericity; under the same solidity hypothesis, the corresponding extension of full ZFC proves the ordinary Ground Axiom.

fields
Logic
pdf ↗ page arXiv:2610.41862 ↗

Algebraic Topology2

Let $\mathcal{F}$ be a saturated fusion system over a finite $p$-group $S$, and $\mathcal{O}^c (\mathcal{F})$ denote the centric orbit category of $\mathcal{F}$. Suppose that $G$ is a discrete group realizing $\mathcal{F}$, and $\mathcal{C}$ is the collection of all $p$-centric subgroups in $G$. We prove that for every $\mathbb{Z}_{(p)} \mathcal{O} ^c (\mathcal{F})$-module $M$, the higher limits $\lim ^* _{\mathcal{O}^c (\mathcal{F})} M$ are isomorphic to the equivariant cohomology $H_ G ^* (|\mathcal{C}| ; \mathcal{M} )$, where $\mathcal{M}$ is the $G$-equivariant coefficient system on $|\mathcal{C}|$ induced by $M$. This is the fusion system version of a theorem by Grodal for higher limits over the $p$-orbit category $\mathcal{O}_p(G)$ of a finite group $G$. As a consequence, for every fusion system $\mathcal{F}$, we obtain a finite sequence of cohomology groups, involving the cohomology of $\mathcal{F}$, which is exact if and only if the mod-$p$ subgroup decomposition for $\mathcal{F}$ over $\mathcal{F}$-centric subgroups is sharp.

comments
27 pages
fields
Algebraic Topology
pdf ↗ page arXiv:2610.40915 ↗

Classical persistent homology provides a summary of the multiscale topology of a filtration of a space. These summaries are not naturally valued in a vector space, limiting their direct use in standard machine learning pipelines. To address this issue, Bubenik introduced persistence landscapes, vector-valued invariants of persistence modules, which operate by conveniently encoding the behavior of rank functions of modules along the parameter axis. When passing to multiparameter persistent homology, changes in the rank functions vary by different amounts in different directions. In this paper, we introduce the landscape interval, a subset of the parameter space of a multiparameter persistence module which records changes in the rank function in all directions simultaneously. This geometric perspective leads to two novel invariants which serve as notions of persistence landscapes for multiparameter modules. We show that our invariants generalize existing landscape-type invariants: in particular, they recover Bubenik's landscapes in the one-parameter setting and allow us to obtain the multiparameter landscapes introduced by Vipond. For our main theoretical contributions, we establish two exact identities connecting landscapes with erosion distances, prove that our invariants are stable under perturbations, and refine the interleaving-stability bounds for the existing landscape-type invariants. We also show that on a certain class of multiparameter persistence modules, the full invariants can be recovered from finite information about the module. On the practical side, we introduce a robust numerical framework for extracting our invariants from real data and for incorporating them into standard machine learning pipelines. The framework is illustrated on surface mesh and medical image data.

comments
67 pages, 6 figures, 4 tables
fields
Algebraic Topology
pdf ↗ page arXiv:2610.42090 ↗

History and Overview2

The Alabama paradox occurs when increasing the total number of seats in a legislature causes a state to lose a seat. We revisit a geometric analysis of this paradox published by Brent Bradberry in 1992, correct an erroneous probability formula, investigate the source of the error, and explore further consequences of the geometry.

comments
This article is written in an expository style expected for submission in a Mathematical Association of America journal
fields
History and Overview
pdf ↗ page arXiv:2610.40682 ↗

We revisit Fourier's method for proving the irrationality of $e$ and generalize it into a unified Diophantine framework, which applies to a broad class of series with bounded or mildly growing integer numerators, and suitably growing, divisible denominators. Using integer linear forms, we establish a general criterion that recovers classical irrationality results while also generating new ones within a single structural approach.

journal
The Mathematical Gazette (2026)
fields
History and Overview, Number Theory
pdf ↗ page arXiv:2610.40698 ↗

K-Theory and Homology2

By constructing a fully dualizable category $\mathrm{Bord}^{\mathcal{C}}_n$ as the left adjoint of the forgetful functor, we state a generalized cobordism hypothesis. As a corollary, we establish a generalized Pontryagin-Thom isomorphism that gives a geometric interpretation of higher algebraic $K$-theory using cobordism classes of framed manifolds and stratified spaces decorated by categorical string diagrams, which we call lasagnas. For finite-dimensional vector spaces over a field, these decorations are branched local systems. We also examine low-dimensional representatives, including $K_1$, Steinberg-group presentations of $K_2$, and fillings related to the Steinberg relation, providing a more explicit construction of Ghys' manifold.

fields
K-Theory and Homology, Algebraic Topology
pdf ↗ page arXiv:2610.41105 ↗

In this paper, we formulate Land-Tamme's $\boxtimes$-construction and $\odot$-construction for continuous exact monads on dualizable stable $\infty$-categories, and establish the behaviour of localizing invariants on pushouts of left modules over continuous exact monads on dualizable stable $\infty$-categories. Combining the Land-Tamme theory of continuous exact monads with the theory of analytic functors in Goodwillie calculus, we show that the reduced K-theory on modules over free tensor algebras of dualizable t-categories is modified 1-coanalytic or modified 2-coanalytic, depending on the coconnectivity of the base algebras. In particular, this answers Efimov's question concerning coanalyticity in the affirmative.

comments
Comments are welcome!
fields
K-Theory and Homology, Category Theory
pdf ↗ page arXiv:2610.41566 ↗

Operator Algebras2

We introduce the notion of Boomerang subalgebras of group von Neumann algebras and study their general properties. We show that every Boomerang subalgebra of $L(Γ)$ is the group von Neumann algebra of a Boomerang subgroup when $Γ$ is a countable $C^*$-simple acylindrically hyperbolic group or a finite product of such groups. Consequently, every such group is invariant random algebra (IRA) rigid: every IRA arises as the pushforward of an invariant random subgroup (IRS) under the group von Neumann algebra construction. We also establish IRA rigidity for adjoint $K$-split higher rank $S$-arithmetic groups, as well as for all countable mixed identity-free (MIF) groups and their finite products. Consequently, IRA rigidity holds for countable groups admitting a topologically free minimal extremely proximal action on a compact Hausdorff space with more than two points. The same conclusion holds for countable groups admitting a faithful minimal action of general type on a tree whose induced boundary action is topologically free, and for finite products of groups admitting either of these actions.

comments
36 pages, preliminary version. Comments are welcome
fields
Operator Algebras, Dynamical Systems, Functional Analysis, General Topology, Group Theory
pdf ↗ page arXiv:2610.41417 ↗

We prove that a von Neumann algebra $N$ is injective if and only if $\|T\|_{\mathrm{dec}}=\|T\|_{\mathrm{cb}}$ for every linear map $T:\ell^\infty_3to N$, answering a problem of Haagerup (1985).

fields
Operator Algebras
pdf ↗ page arXiv:2610.41746 ↗

Spectral Theory2

We consider the Laplacian with Robin boundary conditions on a smooth Riemannian manifold with boundary and investigate the behavior of various spectral quantities as the Robin parameter approaches infinity. Although the spectrum, heat trace, and Riesz means converge pointwise to their Dirichlet analogues, the asymptotics of these quantities as the spectral parameter approaches infinity, or as time approaches zero, do not. The remainders in the usual asymptotic expansions are not bounded as the Robin parameter approaches infinity. Our main result is a two-term asymptotic expansion for the Robin heat trace that depends jointly on the time variable and the Robin parameter, with a remainder that satisfies a uniform bound independent of the Robin parameter. We apply this result to obtain a similar two-term asymptotic expansion for Riesz means of order at least one, again with a remainder bound independent of the Robin parameter.

comments
32 pages
fields
Spectral Theory, Analysis of PDEs
pdf ↗ page arXiv:2610.40804 ↗

We prove that the spectrum of the Laplace–Beltrami operator on compact Riemannian manifolds without boundary is generically linearly independent over $\mathbb{Q}$. Furthermore, we prove that pointwise products of eigenfunctions generically are linearly independent. The proof is based on conformal perturbations of the metric. The result implies in particular that generically the truncation of the continuous functions on $M$ by spectral projections of the Laplace operator is given by symmetric matrices.

comments
22 pages
fields
Spectral Theory, Analysis of PDEs, Differential Geometry, Operator Algebras
pdf ↗ page arXiv:2610.40859 ↗

Symplectic Geometry1

Ekholm and Shende's skeins-on-branes formalism provides invariant counts of holomorphic curves with boundary on a Maslov zero Lagrangian within a Calabi-Yau 3-fold. This paper proves the orientation relations required for worldsheet skein relations to cancel codimension-one boundary strata. However, standard Cauchy-Riemann Fredholm orientations reveal an unexpected sign puzzle: the orientation relation at a curve with a hyperbolic boundary crossing depends on whether the crossing lies on distinct boundary components or the same component. This domain-dependent sign contradicts the uniform crossing relations of the HOMFLYPT skein algebra. We show that this discrepancy is resolved by passing to an anti-holomorphic orientation convention. We establish all required orientation comparisons in the general setting of an oriented Lagrangian submanifold of a 6-dimensional symplectic manifold.

comments
73 pages, 25 figures
fields
Symplectic Geometry, Geometric Topology
pdf ↗ page arXiv:2610.41233 ↗

Revised papers217

Sites in an infinite d-dimensional lattice, open with probability greater or equal to 1/d, form an infinite open path.

comments
There is an error in definition of percolation in the first sentence: "Vertices (sites), open with the smallest probability pH, percolate when they form an infinite open path from graph's origin v0." Vertices percolate when the probability of the open path from an origin to the infinity is positive
fields
Mathematical Physics, Combinatorics
pdf ↗ page arXiv:1307.2827 ↗

A kernel of a digraph is an independent set $K$ such that every vertex outside $K$ has an arc to a vertex in $K$. We present a polyhedral description of kernels in clique-acyclic orientations of line multigraphs. Given a digraph $D$, let $FK(D)$ denote its fractional kernel polytope, and let $σ(D)$ denote the linear system defining $FK(D)$. A digraph $D$ is called kernel perfect if every induced subdigraph has a kernel, called kernel ideal if the fractional kernel polytope of every induced subdigraph is integral, and called kernel Mengerian if the defining system of every induced subdigraph is totally dual integral. We show that a clique-acyclic orientation of a line multigraph is kernel ideal if and only if it is kernel Mengerian. These equivalent properties hold precisely when the orientation is kernel perfect and has no even-cycle obstruction.

comments
Corrects an error in the previous version; the main theorem and its proof have been revised
fields
Combinatorics
pdf ↗ page arXiv:1507.06053 ↗

In 1982, Uhlenbeck \cite {U2} established the well-known gauge fixing theorem, which has played a fundamental role for Yang-Mills theory. In this paper, we apply the idea of Uhlenbeck to establish a parabolic type of gauge fixing theorems for the Yang-Mills flow and prove existence of a weak solution of the Yang-Mills flow on a compact $n$-dimensional manifold with initial value $A_0$ in $W^{1,n/2}(M)$. When $n=4$, we improve a key lemma of Uhlenbeck (Lemma 2.7 of \cite {U2}) to prove uniqueness of weak solutions of the Yang-Mills flow on a four dimensional manifold.

comments
Replaced by a new paper: "The Yang-Mills flow in vector bundles over higher dimensional Riemannian manifolds" by Casey, Chan and Hong (arXiv:2610.38124)
fields
Differential Geometry
pdf ↗ page arXiv:1701.00601 ↗

Let $S$ be a closed surface of genus $g$. In this paper, we investigate the relationship between hyperbolic cone-structure on $S$ and representations of the fundamental group into $\text{PSL}_2\Bbb R$. We consider surfaces of genus greater than $g$ and we show that, under suitable conditions, every representation $ρ:π_1 S\longrightarrow \text{PSL}_2\Bbb R$ with Euler number $\mathcal{E}(ρ)=\pm\big(χ(S)+1\big)$ arises as holonomy of a hyperbolic cone-structure $σ$ on $S$ with a single cone point of angle $4π$. From this result, we derive that for surfaces of genus $2$ every representation with $\mathcal{E}(ρ)=\pm1$ arises as the holonomy of some hyperbolic cone-structure.

comments
The paper is withdrawn because a more recent work proves more general results that include our theorem as a special case
fields
Geometric Topology
pdf ↗ page arXiv:1802.00755 ↗

Denote by $u(n)$ the largest principal specialization of the Schubert polynomial: $ u(n) := \max_{w \in S_n} \mathfrak{S}_w(1,\ldots,1) $ Stanley conjectured in [arXiv:1704.00851] that there is a limit $\lim_{n\to \infty} \, \frac{1}{n^2} \log u(n), $ and asked for a limiting description of permutations achieving the maximum $u(n)$. Merzon and Smirnov conjectured in [arXiv:1410.6857] that this maximum is achieved on layered permutations. We resolve both Stanley's problems restricted to layered permutations.

comments
15 pages, 5 figures, fixed Figure 5, updated references, and added a footnote before Conjecture 1.4
journal
Proc. Amer. Math. Soc. 147 (2019), 1377-1389
fields
Combinatorics
pdf ↗ page arXiv:1805.04341 ↗

We give a new proof that free Burnside groups of sufficiently large even exponents are infinite. The method is very flexible and can also be used to study (partially) periodic quotients of any group which admits an action on a hyperbolic space satisfying a weak form of acylindricity.

fields
Group Theory, Metric Geometry
pdf ↗ page arXiv:1810.08372 ↗

This paper studies affine Weyl groups through the geometry of geodesic tubes and their hyperbolic stabilizers. We introduce a geometric framework relating affine Weyl group factorizations to Hecke-algebra-valued Poincaré series and zeta functions on quotient complexes. Using this framework, we prove a conjecture describing the connection between geodesic tubes and factorizations of Poincaré series for affine Weyl groups. We also establish partial results toward a Hecke-algebra-valued zeta identity for simply connected groups, including the cases of types $\widetilde{A}_{n-1}$ and $\widetilde{C}_n$. These results highlight the interaction between affine Weyl group geometry, Hecke algebras, and zeta functions arising from Bruhat–Tits buildings.

comments
35 pages. Accepted for publication in Journal of Group Theory. Corrected the definition and normalization of circular geodesic tube zeta functions, with corresponding proof amendments
fields
Group Theory
pdf ↗ page arXiv:1903.00703 ↗

In this paper,we consider the Floquet Hamiltonian $K$ associated with a three-body Schrödinger operator with time-periodic pair potentials $H(t)$. By introducing a conjugate operator $A$ for $K$ in the standard Mourre theory, we prove the Mourre estimate for $K$. As by-products of the Mourre estimate for $K$, the minimal velocity estimates for the physical propagator $U(t,0)$ generated by $H(t)$ as well as the propagator $e^{-iσK}$ generated by $K$ can be obtained.

fields
Mathematical Physics
pdf ↗ page arXiv:1904.10190 ↗

Assuming $G$ is unimodular and $K>1$, and that $μ(A^2) < K μ(A)$. We show that this is a continuous surjective group homomorphism $χ: G \to L$ with a compact kernel, where $L$ is a Lie group satisfying $$\dim(L) \leq \lfloor\log K\rfloor(\lfloor\log K\rfloor+1)/2.$$ This dimension bound is sharp, We also establish the relationship between $A$ and its image under the quotient map, and obtain a more general version of this result for the product set $AB$ without assuming unimodularity. We present two applications: (1) An inverse Kemperman theorem in noncompact groups. Let $G$ be a connected, noncompact, locally compact group with a left Haar measure $μ$, and let $A$ and $B$ be nonempty, compact subsets of $G$. We classify all cases where $G$, $A$, and $B$ exhibit nearly minimal expansion properties (in the unimodular case, this means $μ(AB)$ is close to $μ(A) + μ(B)$). This result completes the noncompact cases of the Kemperman inverse problem and resolves a question posed by Griesmer and Tao. (2) An effective generalized Margulis lemma for discrete subgroups of almost-connected locally compact groups acting properly by isometries. If $B(x,4r)$ can be covered by $K$ translates of $B(x,r)$ at one point and one scale, it gives a small radius $r/2M$, and the group generated at this radius has a subgroup of index at most $M \exp((\log K)^4)$ which is nilpotent modulo a finite normal subgroup. The nilpotency step is bounded by $(\log K)^2$, and \[ M\leq K^{\exp(C(\log K)^4\log\log K)} \] and $C$ is an absolute constant.

comments
33 pages, introduction and application updated
fields
Group Theory, Combinatorics
pdf ↗ page arXiv:2111.05236 ↗

In this article, we study Prasad's conjecture for regular supercuspidal representations based on the machinery developed by Hakim and Murnaghan to study distinguished representations, and the fundamental work of Kaletha on parameterization of regular supercuspidal representations. For regular supercuspidal representations, we give some new interpretations of the numerical quantities appearing in Prasad's formula, and reduce the proof to the case of tori. The proof of Prasad's conjecture then reduces to a comparison of various quadratic characters appearing naturally in the above process. We also have some new observations on these characters and study the relation between them in detail. For some particular examples, we show the coincidence of these characters, which gives a new purely local proof of Prasad's conjecture for regular supercuspidal representations of these groups. We also prove Prasad's conjecture for regular supercuspidal representations of G(E), when E/F is unramified and G is a general tamely ramified quasi-split reductive group.

comments
80 pages. Final version accepted by Adv.Math
journal
Adv. Math. 498 (2026), 111036, 126 pp
fields
Representation Theory, Number Theory
pdf ↗ page arXiv:2201.00447 ↗

We show that an $n\times n$ circulant Hadamard matrix must satisfy a family of congruence equations that have solutions only when $n \leq 4$, proving Ryser's 1963 conjecture that no such matrices exist for $n>4$.

fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2302.08346 ↗

We prove a fundamental theorem for tropical partial differential equations, analogous to the fundamental theorem of tropical geometry in this context. We extend results from Aroca et al., Falkensteiner et al. and from Fink and Toghani for the case of trivial valuation as introduced by Grigoriev to differential equations with power series coefficients over any valued field. Crucial ingredients are the framework for tropical partial differential equations introduced by Giansiracusa and Mereta and a result on infinite intersections of projections of fibers of tropicalizations, which we prove using Hrushovski and Loeser's model-theoretic interpretation of Berkovich analytification. As a corollary of the fundamental theorem, we show that the radius of convergence of solutions of an ordinary differential equation over a nontrivially valued field can be computed tropically.

comments
43 pages, extended the scope from univariate to multivariate case, fixed a gap in the proof of Proposition 3.5 of version 1 and fixed Lemma 2.13 of version 2, minor modifications at the end of section 2 and a small fix in the proof of Theorem 3.1
fields
Algebraic Geometry, Commutative Algebra, Rings and Algebras
pdf ↗ page arXiv:2303.12124 ↗

Latent variable modeling of discrete ordinal data typically relies on Euclidean priors or continuous divergence bounds, inducing a topological mismatch that leaves models vulnerable to factor score indeterminacy, rotational unidentifiability, and Neyman-Scott incidental parameter bias. We introduce an axiomatic, coordinate-free framework that constructs continuous latent coordinate spaces directly from empirical partial orders without extrinsic probability measures. By parameterizing ordinal choice topologies via a regularized log-odds field on a continuous projective product manifold $(\mathbb{P}^1)^Q$, pairwise field differences preserve the invariants of the projective linear group $\mathrm{PGL}(2, \mathbb{R})^Q$. Spectral decomposition of the integrated invariant Gramian matrix reduces the identification fiber strictly to the one-dimensional dilation group $\mathbb{R}_{>0}^Q$. This establishes ratio-scale measurement with generalized specific objectivity, transcending the restrictive uniform-discrimination assumptions of the Rasch model, identically resolving both classical rotational ambiguity and Guttman's factor score indeterminacy. We prove that an indicator array $P \ge 2Q+2$ is topologically necessary and sufficient to anchor a $Q$-dimensional metric space and structurally resolves incidental item parameters without joint, marginal, or conditional estimation. Finally, leveraging Hadamard differentiability and Hanson-Wright operator concentration, we establish that the projective spectral estimator is strongly consistent, asymptotically normal, and achieves optimal parametric concentration at rate $\mathcal{O}_P(P^{-1/2})$.

fields
Statistics Theory
pdf ↗ page arXiv:2307.10973 ↗

Integrated sensing and communications (ISAC) is one of the six usage scenarios of IMT-2030, with in-band full duplex (IBFD) being the most direct way to realize it. In this paradigm, a base station that transmits and receives on the same time-frequency resources can sense continuously with its own downlink waveform while serving uplink traffic. This article revisits full-duplex-enabled ISAC (FD-ISAC) in light of the technologies lately shaping sixth-generation (6G) radio design, such as extremely large multiple-input multiple-output (XL-MIMO) systems operating in the near field, holographic and dynamic metasurface apertures, fluid and movable antennas, beyond-diagonal reconfigucable intelligent surfaces (BS-RISs) and wave-domain processing with stacked intelligent metasurfaces. We present a holistic FD-ISAC architecture built around a multi-stage, echo-preserving self-interference cancellation chain, and map each 6G technological enabler to the gains it brings and the risks it introduces. A case study with a $28$~GHz full-duplex XL-MIMO base station showcases that near-field-aware beamforming combined with wave-domain, analog, and digital cancellation delivers more than twice the throughput of half-duplex and 3GPP sub-band full-duplex baselines at the same localization-accuracy bound. It is also demonstrated that incorporating a group-connected BD-RIS recovers most of the fully-connected gain at a small fraction of its circuit complexity. The article concludes discussing emerging use cases and open challenges.

fields
Information Theory
pdf ↗ page arXiv:2308.07266 ↗

We lay the geometric foundations for the study of the characteristic polynomial of tensors. For symmetric tensors of order $d \geq 3$ and dimension $2$, and symmetric tensors of order $3$ and dimension $3$, we prove that only finitely many tensors share any given characteristic polynomial, unlike the case of symmetric matrices and the case of non-symmetric tensors. We propose precise conjectures for the dimension of the variety of tensors sharing the same characteristic polynomial, in the symmetric and non-symmetric settings.

comments
23 pages, final version
fields
Algebraic Geometry
pdf ↗ page arXiv:2308.10957 ↗

Inspired by the study of transport properties in materials science, we investigate the Reeb graph of a map \(f:X\to\Ss\), interpreting \(X\) as a periodic tunnel system in which transport takes place. The map \(f\) encodes a chosen periodic direction and induces a directed structure on the graph. Under constructibility, we establish a lifting result for directed paths that makes precise how the graph represents tunnel segments in \(X\). To turn this representation into a numerical transport descriptor, we equip the graph with cross-sectional capacities defined, in a Riemannian setting, by weighted Hausdorff-measure integrals. Under boundary regularity and Morse assumptions in dimension \(d\ge2\), we establish continuity over regular Reeb values and continuity of the total capacity across fibers. Taking edgewise infima gives the bottleneck capacities for a circular max-flow formulation, which measures flow through fibers of the map to the circle without choosing an arbitrary inlet or outlet. The resulting descriptor is computable by linear optimization and is related to minimum-cost circulations. For nonnegative real capacities, we prove that the max-flow value is invariant under periodic unrolling: the same descriptor is obtained whether the medium is modeled by a toroidal cell or by its periodic replication along the chosen direction. Simulations on periodic point clouds illustrate the construction and the effects of imposing source–target cuts.

fields
Algebraic Topology
pdf ↗ page arXiv:2312.04357 ↗

This paper provides a bound for the supremum of sample averages over a class of functions for a general class of mixing stochastic processes with arbitrary mixing rates. Regardless of the speed of mixing, the bound is comprised of a concentration rate and a novel measure of complexity. The speed of mixing, however, affects the former quantity implying a phase transition. Fast mixing leads to the standard root-n concentration rate, while slow mixing leads to a slower concentration rate whose speed depends on the mixing structure. Our findings are applied to obtain new Glivenko-Cantelli type results.

comments
Corrects a few typos in the published version
fields
Probability, Statistics Theory
pdf ↗ page arXiv:2402.11394 ↗

This paper aims to study the birational geometry of log Calabi-Yau pairs $(\mathbb{P}^3, D)$ of coregularity 2, where in this case $D$ is an irreducible normal quartic surface with canonical singularities. We completely classify which toric weighted blowups of a point will initiate a volume preserving Sarkisov link starting with this pair. Depending on the type of singularity, our results point out that some of these weights do not work generically for a general member of the corresponding coarse moduli space of quartics.

comments
38 pages, 3 figures. Final version to appear in Publicacions Matemàtiques. I thank both referees for their valuable comments that contributed to the improvement of this paper
fields
Algebraic Geometry
pdf ↗ page arXiv:2402.13970 ↗

In a recent paper, the author and Stöhr established a bound on the number of iterated Frobenius pullbacks needed to transform a non-smooth purely inseparable point on a regular geometrically integral curve into a rational point. In this paper we improve this result, by establishing a new bound that is sharp in every characteristic $p>0$.

comments
13 pages. Final version. To appear in Indagationes Mathematicae
fields
Algebraic Geometry, Number Theory
pdf ↗ page arXiv:2402.14969 ↗

We study how the Bauer–Jost–Liu curvature bound applies to hypergraph Ollivier–Ricci curvature (ORC) and how its accuracy depends on local mass and geometry. An explicit weighted-graph representation shows how the graph theorem applies. A transport construction transfers the bounds to monotone pairwise hyperedge aggregations. We analyze endpoint cancellation under laziness, quantify the exact improvement over the earlier mixture bound, and characterize the extremal geometry at fixed endpoint and overlap statistics. Further results quantify worst-case slack, stability, and exactness on isolated hyperedges. A capacity-constrained short-detour refinement uses additional geometric information and improves monotonically as the query budget increases. As a computational application, a threshold cascade preserves exact transport decisions. Numerical verification and public-hypergraph experiments examine tightness, walk dependence, and aggregation, including evaluations on 42,940 finite pairwise comparisons and 64,620 public-data instances. Together, the results specify the scope, equality mechanisms, and computational uses of BJL-type hypergraph bounds.

fields
Optimization and Control
pdf ↗ page arXiv:2405.13302 ↗

This is a concise mathematical introduction to Monte Carlo methods, a rich family of algorithms with far-reaching applications in science and engineering. Monte Carlo methods are an exciting subject for mathematical statisticians and computational and applied mathematicians: the design and analysis of modern algorithms are rooted in a broad mathematical toolbox that includes ergodic theory of Markov chains, Hamiltonian dynamical systems, transport maps, stochastic differential equations, information theory, optimization, Riemannian geometry, and gradient flows, among many others. These lecture notes celebrate the breadth of mathematical ideas that have led to tangible advancements in Monte Carlo methods and their applications. To accommodate a diverse audience, the level of mathematical rigor varies from chapter to chapter, giving only an intuitive treatment to the most technically demanding subjects. The aim is not to be comprehensive or encyclopedic, but rather to illustrate some key principles in the design and analysis of Monte Carlo methods through a carefully-crafted choice of topics that emphasizes timeless over timely ideas. Algorithms are presented in a way that is conducive to conceptual understanding and mathematical analysis -- clarity and intuition are favored over state-of-the-art implementations that are harder to comprehend or rely on ad-hoc heuristics. To help readers navigate the expansive landscape of Monte Carlo methods, each algorithm is accompanied by a summary of its pros and cons, and by a discussion of the type of problems for which they are most useful. The presentation is self-contained, and therefore adequate for self-guided learning or as a teaching resource. Each chapter contains a section with bibliographic remarks that will be useful for those interested in conducting research on Monte Carlo methods and their applications.

comments
155 pages, 25 figures
fields
History and Overview, Numerical Analysis
pdf ↗ page arXiv:2405.16359 ↗

We prove the Geometric P=W conjecture in rank 3 on the three-punctured sphere. We describe the topology at infinity of the related character variety. We use asymptotic abelianization of harmonic bundles away from the ramification divisor and an equivariant approach near the branch points to find the WKB (also known as Liouville–Green or phase-integral) expansion of the involved maps. We analyze the Stokes phenomenon governing their behavior.

comments
66 pages, 7 figures; v2: expanded Section 6; v3: published version
journal
The Journal of Geometric Analysis (2026) 36:140
fields
Algebraic Geometry
pdf ↗ page arXiv:2407.20638 ↗

We consider the problem of learning maximal Lyapunov functions to compute the maximal region of attraction of nonlinear dynamical systems using neural networks. Two obstacles limit existing neural-network approaches: 1) finite-width approximations can fail to preserve the strict Lyapunov inequalities near the equilibrium, where they become arbitrarily weak, and 2) training is necessarily performed on a bounded computational domain for which the classical Lyapunov and Zubov theorems do not directly apply. To address these issues, first we derive Lyapunov and Zubov theorems directly on a prescribed compact computational domain, characterizing the maximal region of attraction within that domain as the solution of a boundary-value PDE. Second, we introduce the Taylor Lyapunov network, an architecture that prescribes the quadratic leading-order term via the system's linearization and represents the higher-order remainder through a structured cubic factorization, guaranteeing that any network in this class is a valid local Lyapunov function by construction. Finally, we prove that Taylor Lyapunov networks are universal approximators of maximal Lyapunov functions on compact domains, and combine these results into an unsupervised, physics-informed training procedure. Numerical experiments on systems of varying dimensions and stability types show that the proposed architecture is efficient at estimating the maximum region of attraction.

fields
Optimization and Control
pdf ↗ page arXiv:2408.17246 ↗

Let $B\subset \mathbb{P}^3$ be an slc quartic surface. The existence of an embedding $\mathbb{G}_m^3\hookrightarrow \mathbb{P}^3\setminus B$ implies that $B$ has coregularity zero. In this article, we initiate the classification of coregularity zero slc quartic surfaces $B\subset \mathbb{P}^3$ for which $\mathbb{P}^3\setminus B$ contains an algebraic torus $\mathbb{G}_m^3$, and equivalently, the classification of cluster type pairs $(\mathbb{P}^3,B)$. Along the way, we give criteria for a log Calabi–Yau pair $(X,B)$ over a toric variety $T$ to be of cluster type.

comments
24 pages. Minor corrections. We thank both referees for their comments that contribute to the sharpening of the paper
journal
Journal of the London Mathematical Society, 113(1):e70427, 2026
fields
Algebraic Geometry
pdf ↗ page arXiv:2411.03506 ↗

In this paper, we use the theory of nonlinear semigroups to establish the existence and uniqueness of both local and global solutions for a partial differential-algebraic equation (PDAE) of index one. This method is applied to a reaction-diffusion system coupled with an elliptic equation in one dimension by transforming the PDAE into a system of linear evolution equations with a Lipschitz-continuous perturbation

comments
The authors are withdrawing this manuscript to revise the theoretical framework and correct incomplete proofs
fields
Analysis of PDEs
pdf ↗ page arXiv:2411.15658 ↗

We generalize the higher Riemann-Hilbert correspondence in the presence of scalar curvature for a (possibly non-compact) smooth manifold $M$. We show that the dg-category of curved $\infty$-local systems, the dg-category of graded vector bundles with projectively flat $\mathbb Z$-graded connections and the dg-category of curved representations of the singular simplicial set of the based loop space of $M$ are all $A_\infty$-quasi equivalent. They provide dg-enhancements of the subcategory of the bounded derived category of twisted sheaves whose cohomology sheaves are locally constant and have finite-dimensional fibers. In the ungraded case, we reduce to an equivalence between projectively flat vector bundles and a subcategory of projective representations of $π_1(M; x_0)$. As an application of our general framework, we also prove that the category of cohesive modules over the curved Dolbeault algebra of a complex manifold $X$ is equivalent to a subcategory of the bounded derived category of twisted sheaves of $\mathcal O_X$-modules which generalizes a theorem due to Block to possibly non-compact complex manifolds.

fields
Algebraic Topology, Differential Geometry
pdf ↗ page arXiv:2411.19595 ↗

The \emph{$r$-neighbourhood complexity} of a graph $G$ is the function counting, for a given integer $k$, the largest possible number, over all vertex-subsets $A$ of size $k$, of subsets of $A$ realized as the intersection between the $r$-neighbourhood of some vertex and $A$. A~refinement of this notion is the \emph{$r$-profile complexity}, that counts the maximum number of distinct distance-vectors from any vertex to the vertices of $A$, ignoring distances larger than~$r$. Typically, in structured graph classes such as graphs of bounded VC-dimension or chordal graphs, these functions are bounded, leading to insights into their structural properties and efficient algorithms. We improve existing bounds on the $r$-profile complexity (and thus on the $r$-neighbourhood complexity) for graphs in several structured graph classes. We show that the $r$-profile complexity of graphs excluding $K_h$ as a minor is in $O_h(r^{3h-3}k)$. For graphs of treewidth at most~$t$, we give a bound in $O_t(r^{t+1}k)$, which is tight up to a function of~$t$ as a factor. These bounds improve results of Joret and Rambaud and answer a question of their paper [Combinatorica, 2024]. We also apply our methods to other classes of bounded expansion such as graphs excluding a fixed complete graph as a subdivision. For outerplanar graphs, we can improve our treewidth bound by a factor of $r$ and conjecture that a similar improvement holds for graphs with bounded simple treewidth. For graphs of treelength at most~$\ell$, we give the upper bound of $O(k(r^2(\ell+1)^k))$, which we improve to $O\left (k\cdot (r 2^k + r^2k^2) \right)$ in the case of chordal graphs and $O(k^2r)$ for interval graphs. Our bounds also imply relations between the order, diameter and metric dimension of graphs in these classes, improving results from [Beaudou et al., SIDMA 2017].

comments
various improvements, accepted to SIDMA
fields
Combinatorics
pdf ↗ page arXiv:2501.08895 ↗

Linguistic steganography aims to conceal information within natural language text without being detected. An effective steganography approach should encode the secret message bits into a minimal number of language tokens while preserving the natural appearance and fluency of the stego-texts. We present a new framework to enhance the embedding efficiency by framing the steganographic embedding procedure as a Constrained Markov Decision Process (CMDP), which takes into consideration the long-term dependencies instead of merely the immediate effects. We constrain the solution space such that the discounted expected cumulative total variation between the selected conditional probability distribution and the original one given by the LLM is below a threshold. To find the optimal policy, we first show that the functional optimization problem can be simplified to a convex problem with a finite number of variables. For the binary-state model, we give a full characterization of the optimal policy, including closed-form expressions in several regimes, and show that it displays water-filling-like behavior in certain cases. The solution indicates that states with highly deterministic transition distributions are preferable for adjustment in the binary-state case. For larger state spaces, the optimal solution is more complex and explicit solutions are difficult to obtain, but the optimal policies show similar structural behaviors.

comments
Long journal version, 33 pages
fields
Information Theory
pdf ↗ page arXiv:2502.01827 ↗

We investigate the small-time local controllability (STLC) near the ground state of a bilinear Schrödinger equation when the linearized system is not controllable. It is well known that, for single-input systems, quadratic terms in the state expansion can then lead to obstructions to the STLC of the nonlinear system. In this work, we extend this phenomenon to the multi-input setting, presenting the first example of multi-input quadratic obstructions for PDEs. Our results build upon our previous study of such obstructions for ODEs and provide a functional framework for analyzing them in the bilinear Schrödinger equation.

comments
arXiv admin note: text overlap with arXiv:2407.07446
fields
Optimization and Control
pdf ↗ page arXiv:2502.14999 ↗

Let $\mathbb{F}_q$ be a finite field with $q$ elements, where $q$ is a prime power and let $A:= \mathbb{F}_{q}[T]$ and $F:= \mathbb{F}_{q}(T)$. By~\cite{PR09}, the adelic image of the Galois representation attached to a generic Drinfeld $A$-module $\varphi$ of rank $2$ over $F$ satisfying $\mathrm{End}_{\bar{F }}(\varphi)=\varphi(A)$ is open, and determining when it is surjective remains a subtle problem. To resolve this question, in this article, we study the $\mathfrak{p}$-adic surjectivity of the Galois representations attached to $\varphi$, where $\mathfrak{p} \in Ω_A:= \mathrm{Spec}(A) \setminus \{ (0) \}$. There are two directions to investigate this problem: one by fixing the prime $\mathfrak{p}$, and the other by fixing $\varphi$. In the horizontal direction, for a fixed $\mathfrak{p} \in Ω_A$, we give explicit and easily verifiable conditions on Drinfeld $A$-modules $\varphi$ of rank $2$ over $F$ which ensure the surjectivity of the $\mathfrak{p}$-adic Galois representation $ρ_{\varphi,\mathfrak{p}}$. This work extends~\cite{Ray24} to arbitrary $\mathfrak{p}$, and, for $\mathfrak{p}=(T)$, recovers the results of~\textit{loc. cit.} under weaker hypotheses. In the vertical direction, we show that for a fixed Drinfeld $A$-module $\varphi$ of rank $2$ over $F$, whose coefficients satisfy certain congruence and valuation conditions, the $\mathfrak{p}$-adic Galois representation $ρ_{\varphi,\mathfrak{p}}$ is surjective for all $\mathfrak{p} \in Ω_A$. This recovers the example of \cite{Zyw11} and yields new examples beyond those considered in \cite{Zyw25}. As a consequence, we obtain the surjectivity of the associated adelic Galois representation.

comments
The title of the article has been slightly revised, while the results have been significantly generalized to a broader framework. Additionally, the overall presentation has been substantially enhanced for greater clarity and impact
fields
Number Theory
pdf ↗ page arXiv:2502.19084 ↗

We present a program for enumerating all pseudoline arrangements with a small number of pseudolines and abstract order types of small point sets. This program supports computer experiments with these structures, and it complements the order-type database of Aichholzer, Aurenhammer, and Krasser. This system makes it practical to explore the abstract order types for 12 points, and the pseudoline arrangements of 11 pseudolines.

comments
27 pages, 16 figures. Full version of a paper that has appeared (without appendices) in the 41st European Workshop on Computational Geometry, Liblice, Czech republic, April 9-11, 2025. Version 2: The entries in the lowest three lines of Table 6 have been corrected
fields
Combinatorics
pdf ↗ page arXiv:2503.02336 ↗

We study the asymptotic joint degree distribution of uniformly chosen vertices in uniform recursive directed acyclic graphs, a natural generalization of random recursive trees. Here, a graph is built by adding vertices one-by-one and connecting each new vertex to $m\in\mathbb{N}$ uniformly selected vertices already present. Using a representation based on Kingman's coalescent, we establish a transition in the joint degree distribution at the threshold $(m+1)\log n$, arising from the interplay between geometric degree asymptotics and fluctuations in the coalescent construction. Our results extend previously known asymptotics toward this threshold and provide a unified description of the transition, including mixed regimes where prescribed degrees lie on different scales.

comments
11 pages. The present version and arXiv:2606.09386 supersede the first submission
fields
Probability
pdf ↗ page arXiv:2503.05829 ↗

This paper discusses the possibility of applying the velocity averaging theorems in [F. Golse, P.-L. Lions, B. Perthame, R. Sentis: J. Funct. Anal. 76(1):110–125, 1988] to the Wigner equation governing the quantum evolution of the Wigner transform of quantum density operators. Our first main results address the case of the Wigner function of a special class of density operators associated to mixed states, whose Hilbert-Schmidt norm is of order $\hbar^{d/2}$, where $d$ is the space dimension and $\hbar$ the reduced Planck constant. In space dimension $d=1$, we prove that the density function belongs to the Sobolev space $H^s(\mathbb R)$ for some $s>0$. In the case of pure states, we first obtain a characterization of the Wigner transform of rank-one quantum density operators, and apply this characterization (1) to analyze a rather general setting in which velocity averaging cannot apply to the Wigner functions of a family of rank-one density operators whose evolution is governed by the von Neumann equation, and (2) to obtain a quick derivation of Madelung's system of quantum hydrodynamic equations. This derivation provides a physical explanation of one key assumption used in the proof of the negative result (1) described above.

comments
36 pages
fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2503.09924 ↗

We show that Propositional Dynamic Logic (PDL) has the Craig Interpolation Property. This question has been open for many years. Three proof attempts were published, but later criticized in the literature or retracted. Our proof is based on the main ideas from Borzechowski (1988, master thesis). We define a cyclic tableau system for PDL with a loading mechanism to recognize successful repeats. For this system, we show soundness and completeness via a game. To show interpolation, we modify Maehara's method to work for tableaux with repeats: we first define pre-interpolants at each node, and then use a quasi-tableau to define interpolants for clusters (strongly connected components). In different terms, our method solves the fixpoint equations that characterize the desired interpolants, and the method ensures that the solutions to these equations can be expressed within PDL. The proof is constructive and we show how to compute interpolants. We make available a Haskell implementation of the proof system that provides interpolants, and a formal verification of our proof in the interactive theorem prover Lean. Lastly, we discuss several questions for future work.

fields
Logic
pdf ↗ page arXiv:2503.13276 ↗

We investigate a rainbow analogue of the Ruzsa–Szemerédi problem. For $t \in \mathbb{N}$, we say that a colouring of $E(K_n)$ is almost $t$-Gallai if no two rainbow $t$-cliques share an edge. Motivated by a lemma of Berkowitz on bounding the modulus of the characteristic function of clique counts in random graphs, we study the maximum number $τ_t(n)$ of rainbow $t$-cliques in an almost $t$-Gallai colouring of $E(K_n)$. For every $t \ge 4$, we show that $n^{2-o(1)} \leq τ_t(n) = o(n^2)$. For $t=3$, surprisingly, the behaviour is substantially different. Our main result establishes that $$\left ( \frac{1}{2}-o(1) \right ) n\log n \le τ_3(n) \le O\big (n^{\sqrt{2}}\log n \big ),$$ which gives the first non-trivial improvements over the simple lower and upper bounds. Our proof combines various applications of the probabilistic method and a generalisation of the edge-isoperimetric inequality for the hypercube.

comments
23 pages
fields
Combinatorics
pdf ↗ page arXiv:2503.17334 ↗

In this paper, we study the use of outer metrics, in particular Sobolev-type metrics on the diffeomorphism group in the context of PDE-constrained shape optimization. Leveraging the structure of the diffeomorphism group we analyze the connection between the push-forward of a smooth function defined on the diffeomorphism group and the classical shape derivative as an Eulerian semi-derivative. We consider in particular, two predominant examples on PDE-constrained shape optimization. An electric impedance tomography inspired problem, and the optimization of a two-dimensional bridge. These problems are numerically solved using the Riemannian steepest descent method where the descent directions are taken to be the Riemannian gradients associated to various outer metrics. For comparison reasons, we also solve the problem using other previously proposed Riemannian metrics in particular the Steklov-Poincaré metric.

fields
Optimization and Control, Differential Geometry
pdf ↗ page arXiv:2503.22872 ↗

We provide $m^{1+o(1)}kε^{-1}$-time algorithms for computing multiplicative $(1 - ε)$-approximate solutions to multi-commodity flow problems with $k$-commodities on $m$-edge directed graphs, including concurrent multi-commodity flow and maximum multi-commodity flow. To obtain our results, we provide new optimization tools of potential independent interest. First, we provide an improved optimization method for solving $\ell_{q, p}$-regression problems to high accuracy. This method makes $\tilde{O}_{q, p}(k)$ queries to a high accuracy convex minimization oracle for an individual block, where $\tilde{O}_{q, p}(\cdot)$ hides factors depending only on $q$, $p$, or $\mathrm{poly}(\log m)$, improving upon the $\tilde{O}_{q, p}(k^2)$ bound of [Chen-Ye, ICALP 2024]. As a result, we obtain the first almost-linear time algorithm that solves $\ell_{q, p}$ flows on directed graphs to high accuracy. Second, we present optimization tools to reduce approximately solving composite $\ell_{1, \infty}$-regression problems to solving $m^{o(1)}ε^{-1}$ instances of composite $\ell_{q, p}$-regression problem. The method builds upon recent advances in solving box-simplex games [Jambulapati-Tian, NeurIPS 2023] and the area convex regularizer introduced in [Sherman, STOC 2017] to obtain faster rates for constrained versions of the problem. Carefully combining these techniques yields our directed multi-commodity flow algorithm.

comments
Accepted at STOC 2025
fields
Optimization and Control
pdf ↗ page arXiv:2503.24373 ↗

We consider a random graph in which vertices can have one of two possible colours. Each vertex switches its colour at a rate that is proportional to the number of vertices of the other colour to which it is connected by an edge. Each edge turns on or off according to a rate that depends on whether the vertices at its two endpoints have the same colour or not. We prove that, in the limit as the graph size tends to infinity and the graph becomes dense, the graph process converges, in a suitable path topology, to a limiting Markov process that lives on a certain subset of the space of coloured graphons. In the limit, the density of each vertex colour evolves according to a Fisher-Wright diffusion driven by the density of the edges, while the underlying edge connectivity structure evolves according to a stochastic flow whose drift depends on the densities of the two vertex colours.

comments
60 pages, 3 figures, 5 tables
fields
Probability
pdf ↗ page arXiv:2504.06493 ↗

We consider mixed-integer Nash equilibrium problems in which each player solves a mixed-integer optimization problem parameterized in the rivals' strategies. We distinguish standard Nash equilibrium problems (NEPs), where the parameterization acts only on the cost functions, and generalized NEPs (GNEPs), where the strategy spaces may also depend on the rivals' strategies. We introduce a branch-and-cut (B&C) algorithm for such games that, upon termination, either computes a pure Nash equilibrium or decides that none exists. The main idea is a bilevel reformulation based on the Nikaido–Isoda function. Using bilevel-optimization techniques, we obtain a tractable relaxation of this reformulation and embed it into a B&C framework. Depending on the setting, we derive sufficient conditions for the existence of suitable cuts and finite termination. For GNEPs, we adapt intersection cuts (ICs) from bilevel and mixed-integer linear optimization. Such cuts exist under assumptions that hold, e.g., for pure-integer GNEPs with decoupled concave objectives and linear coupling constraints. Since ICs may become arbitrarily shallow, we complement them with a fallback strategy that fixes recurring integer profiles and ensures finite termination in a relevant setting. For NEPs, suitable cuts always exist via best-response inequalities, and our method terminates finitely whenever the set of best-response sets is finite. This holds, e.g., if (i) the cost functions are concave in the players' own continuous strategies and the strategy sets are polyhedral, or (ii) the cost functions only depend on the player's own strategy and the rivals' integer strategy components. Numerical results on various NEPs and GNEPs show that our tailored B&C method outperforms directly applying the state-of-the-art bilevel solver MibS on the tested integer and generalized knapsack instances.

fields
Optimization and Control
pdf ↗ page arXiv:2506.02520 ↗

We present a unified theoretical framework for analyzing the stability and consistency of Physics-Informed Neural Networks (PINNs), grounded in operator coercivity, variational formulations, and non-asymptotic perturbation theory. PINNs approximate solutions to partial differential equations (PDEs) by minimizing residual losses over sampled collocation and boundary points. We formalize both operator-level and variational notions of consistency, proving that residual minimization in Sobolev norms leads to convergence in energy and uniform norms under mild regularity. Deterministic stability bounds quantify how bounded perturbations to the network outputs propagate through the full composite loss, while probabilistic concentration results via McDiarmid's inequality yield sample complexity guarantees for residual-based generalization. A unified generalization bound links residual consistency, projection error, and perturbation sensitivity. Empirical results on elliptic, parabolic, and nonlinear PDEs confirm the predictive accuracy of our theoretical bounds across regimes. The framework identifies key structural principles, such as operator coercivity, activation smoothness, and sampling admissibility, that underlie robust and generalizable PINN training, offering principled guidance for the design and analysis of PDE-informed learning systems.

comments
arXiv admin note: This submission has been withdrawn because it does not meet arXiv's research content quality standards
journal
109909 (2026)
fields
Functional Analysis, Numerical Analysis
pdf ↗ page arXiv:2506.13554 ↗

Using the modular symbol approach, we prove the rationality and period relations for certain critical values of Rankin-Selberg L-functions for $\mathrm{GL}(n)\times\mathrm{GL}(n)$ over any number field that contains a CM field.

comments
We made some corrections to the previous version
fields
Number Theory, Representation Theory
pdf ↗ page arXiv:2506.20942 ↗

We recently introduced the Fast Newton Transform (FNT), an hierarchical algorithm for performing multivariate Newton interpolation in arbitrary downward closed polynomial spaces of spatial dimension $m$. Here, we analyze the FNT in the context of a specific family of downward closed sets $A_{m,n,p}$, defined as all multi-indices with $\ell^p$ norm less than $n$ with $p \in [0,\infty]$. The FNT performs with time complexity $\mathcal{O}(|A_{m,n,p}|mn)$ on the induced downward closed polynomial spaces $Π_{m,n,p}$. We show that the $Π_{m,n,p}$ choice compared to the tensor product spaces $Π_{m,n,\infty}$, reduces time complexity by a factor of $ρ_{m,n,p}$, decaying super exponentially with spatial dimension when $m \lesssim n^p$. We showcase the efficiency of the FNT by computing activity scores in sensitivity analysis.

comments
The work on which this manuscript is based has undergone substantial revisions that significantly change the underlying formulation. Consequently, the analysis presented here is no longer applicable to the revised work and has become obsolete
fields
Numerical Analysis
pdf ↗ page arXiv:2507.13640 ↗

We prove that coherent configurations can be represented as modules over Frobenius structures in the category of real nonnegative matrices. We generalize the notion of admissible morphism from association schemes to coherent configurations. We show that the Frobenius structure associated to a coherent configuration can be modified to become a dagger Frobenius structure, and use this to connect the coherent configurations to groupoids and $H^*$-algebras. We examine the properties of the dagger Frobenius structure with respect to admissible morphisms. We introduce the matrix $O$ obtained as the composition of comultiplication and multiplication of the dagger Frobenius structure and prove that we may obtain the valencies of colors, and thus recover the original coherent configuration, as an eigenvector of $O$. In the last part of the paper, we examine the spectrum of $O$ and apply it to generalize the Lagrange theorem from groups to association schemes.

comments
61 pages, 77 figures
fields
Combinatorics, Rings and Algebras
pdf ↗ page arXiv:2507.21774 ↗

Dirac's classical theorem asserts that, for $n\ge 3$, any $n$-vertex graph with minimum degree at least $\frac{n}{2}$ is Hamiltonian. Furthermore, if we additionally assume that such graphs are regular and sufficiently large, then, by the breakthrough work of Csaba, Kühn, Lo, Osthus, and Treglown, they admit a decomposition into Hamilton cycles and at most one perfect matching, solving the well-known Nash-Williams conjecture. In the pseudorandom setting, it has long been conjectured that similar results hold in much sparser graphs. We prove two overarching theorems for graphs that exclude excessively dense subgraphs, which yield asymptotically optimal resilience and Hamilton-decomposition results in sparse pseudorandom graphs. Firstly, our results imply that for every fixed $γ>0$, there exists a constant $C>0$ such that if $G$ is a spanning subgraph of an $(n,d,λ)$-graph satisfying $δ(G)\ge\bigl(\tfrac12+γ\bigr)d$ and $d/λ\ge C$, then $G$ must contain a Hamilton cycle. Secondly, we show that for every $\varepsilon>0$, there is $C>0$ so that any $(n,d,λ)$-graph $G$ with $d/λ\ge C$ contains at least $\bigl(1-\varepsilon\bigr)\tfrac d2$ edge-disjoint Hamilton cycles, and, finally, we prove that the entire edge set of $G$ can be covered by no more than $\bigl(1+\varepsilon\bigr)\tfrac d2$ such cycles. All bounds are asymptotically optimal and significantly improve earlier results on Hamiltonian resilience, packing, and covering in sparse pseudorandom graphs.

comments
34 pages
fields
Combinatorics
pdf ↗ page arXiv:2507.22807 ↗

We show that for diffeomorphisms of closed simply-connected $4$-manifolds dissolving under a single stabilization, there is a natural induced diffeomorphism on an exotic copy of $\RR^4$. We thereby construct uncountably many exotic copies of $\RR^4$ with nontrivial compactly supported mapping class groups of arbitrarily large rank. This follows from a modification of the construction of the diffeomorphism corks of arXiv:2407.04696 that makes their exteriors simply-connected.

comments
Main conclusion strengthened, incorporated feedback from reviewer
fields
Geometric Topology
pdf ↗ page arXiv:2508.02017 ↗

Let G = Aut(T) be the automorphism group of a regular tree T. We study continuous irreducible representations of G that preserve a continuous strongly nondegenerate sesquilinear form of finite index on a Hilbert space. These are already classified for index 0 (unitary case) and for index 1. We show that there are no more representations for index > 1, which completes the classification.

comments
v2: we have proved the result for continuous representations into Pontryagin spaces, adding an appendix where we show admissibility of such representations. v3: Final version, published in Geometriae Dedicata. Corrected a few minor errors and improved the exposition. Added a statement declaring minor use of AI, limited to the later stages of preparation and revision of the paper. 43 pages
journal
Geom Dedicata 220, 68 (2026)
fields
Group Theory, Metric Geometry, Representation Theory
pdf ↗ page arXiv:2508.02367 ↗

It is well known that the global well-posedness of the Navier-Stokes equations with temperature-dependent coefficients is a challenging problem, especially in multi-dimensional space. In this paper, we study the 3D Navier-Stokes equations with temperature-dependent coefficients in the whole space. When the initial density and the initial temperature are linearly equivalent to some large constant states, we establish the first result on the global existence of large strong solution. Moreover, the optimal decay rates of the solution to its associated equilibrium are established when the initial data belong to $L^{p_0}(\mathbb{R}^3)$ for some $p_0\in[1,2]$.

comments
81 pages, all comments are welcome
fields
Analysis of PDEs
pdf ↗ page arXiv:2508.06933 ↗

We construct standard resolutions for analytic local modules on complex hypersurfaces using standard basis methods, with extensions to complete intersections. The algebraic version over arbitrary infinite fields is also suggested. Applications include derived category formula of local hypersurface blow-ups, revealing new connections between standard bases and categorical geometry.

fields
Algebraic Geometry
pdf ↗ page arXiv:2508.11306 ↗

In modern retail and e-commerce, customers often purchase multiple items across distinct product categories, exhibiting both substitution and complementarity. We consider the cross-category assortment optimization problem where retailers jointly determine assortments across categories to maximize expected revenue. Most prior work on the topic either overlooks complementarity or proposes models that lead to intractable optimization problems, despite being based on the multinomial logit (MNL) choice model. We propose a sequential multi-purchase choice model for cross-category choice that incorporates complementarity through a Markovian transition structure across categories, while allowing general Random Utility Maximization (RUM)-based choice models to capture the within-category substitution. We develop an Expectation-Maximization algorithm for estimation, and a polynomial-time algorithm for unconstrained assortment optimization that yields the optimal solution when the within-category substitution follows a Markov chain choice model. Furthermore, we introduce an empirical metric to quantify the strength of complementarity across product categories and conduct extensive numerical experiments using both synthetic data and a large-scale transaction-level dataset from a major U.S. grocery store. Our model improves predictive accuracy and model fit in both synthetic and real-data experiments and improves expected revenue in synthetic settings with complementarity. It also reveals intuitive market structures such as brand-loyal cross-category purchasing. Overall, we believe that our model provides a theoretically grounded and practical framework for modeling complementarity and making better cross-category assortment decisions.

fields
Optimization and Control
pdf ↗ page arXiv:2508.18566 ↗

We bound the supremum of stochastic processes using a rate-distortion functional analogous to classical entropy integrals. A lifting argument embeds the process into a higher-dimensional space with permutation symmetry, reducing the analysis to a stationary process. This gives a two-sided characterization of the supremum of Gaussian processes, including cases where Dudley's entropy integral is not sharp. Combining the sharpness of Dudley's integral for stationary processes with information-theoretic arguments gives a proof of the majorizing measure theorem. The method also yields localized generalization bounds without a slicing argument. As an application, we derive risk bounds for empirical risk minimization over Lipschitz images of Sobolev ellipsoids. Our proof does not use the majorizing measure theorem.

comments
1 figure; Annals of Applied Probability, to appear
fields
Statistics Theory
pdf ↗ page arXiv:2508.18682 ↗

For each endotrivial complex for a $p$-group arising from Bredon homology of a representation sphere, we construct $p$-local quasi-isomorphisms, called forerunners. These enable us to extend Balmer–Gallauer's results concerning the tensor-triangular geometry of permutation modules for elementary abelian $p$-groups to all $p$-groups. We construct an open cover of the Balmer spectrum under which all endotrivials are tt-line bundles, that is, every endotrivial is locally isomorphic to a shift of the tensor unit. We define a remixed permutation twisted cohomology ring for which the canonical comparison map from the Balmer spectrum to the homogeneous spectrum of the twisted cohomology ring is injective. If the twisted cohomology ring is Noetherian, the comparison map is an open immersion, and the open cover endows the Balmer spectrum with Dirac scheme structure. We prove Noetherianity holds for Dedekind groups and the dihedral group of order 8, and conjecture it holds for all $p$-groups.

comments
Significant rewrite of Sections 6 and 7 due to an erroneous lemma discovered by an AI tool. Paper and proofs remain human-generated, main results unchanged. 49 pages
fields
Representation Theory, Algebraic Geometry, Category Theory, Group Theory
pdf ↗ page arXiv:2509.00954 ↗

Important nonlinear dynamics, such as those found in plasma and fluid systems, are typically hard to simulate on classical computers. Thus, if fault-tolerant quantum computers could efficiently solve such nonlinear problems, it would be a transformative change for many industries. In a recent breakthrough [Liu et al., PNAS 2021], the first efficient quantum algorithm for solving nonlinear differential equations was constructed, based on a single condition $R<1$, where $R$ characterizes the ratio of nonlinearity to dissipation. This result, however, is limited to the class of purely dissipative systems with negative log-norm, which excludes application to many important problems. In this work, we correct technical issues with this and other prior analysis, and substantially extend the scope of nonlinear dynamical systems that can be efficiently simulated on a quantum computer in a number of ways. Firstly, we extend the existing results from purely dissipative systems to a much broader class of stable systems, and show that every quadratic Lyapunov function for the linearized system corresponds to an independent $R$-number criterion for the convergence of the Carleman scheme. Secondly, we extend our stable system results to physically relevant settings where conserved polynomial quantities exist. Finally, we provide extensive results for the class of non-resonant systems. With this, we are able to show that efficient quantum algorithms exist for a much wider class of nonlinear systems than previously known, and prove the BQP-completeness of nonlinear oscillator problems of exponential size. In our analysis, we also obtain several results related to the Poincaré-Dulac theorem and diagonalization of the Carleman matrix, which could be of independent interest.

comments
73+78 pages, 5 figures. Fixed broken internal hyperlinks in the PDF; scientific content unchanged
journal
Quantum 10, 2233 (2026)
fields
Numerical Analysis
pdf ↗ page arXiv:2509.07155 ↗

While modal extensions of decidable fragments of first-order logic are usually undecidable, their monodic counterparts, in which formulas in the scope of modal operators have at most one free variable, are typically decidable. This only holds, however, under the provision that non-rigid constants, definite descriptions and non-trivial counting are not admitted. Indeed, several monodic fragments having at least one of these features are known to be undecidable. We investigate these features systematically and show that fundamental monodic fragments such as the two-variable fragment with counting and the guarded fragment of standard first-order modal logics $\mathbf{K}_{n}$ and $\mathbf{S5}_{n}$ are decidable. Tight complexity bounds are established as well. Under the expanding-domain semantics, we show decidability of the basic modal logic extended with the transitive closure operator on frames without infinite ascending chains; this logic, however, is Ackermann-hard.

fields
Logic
pdf ↗ page arXiv:2509.08165 ↗

The main aim of this work is to present an explicit construction of a 2-design of ${\rm U}(2)$, relying only on a tool that belongs to every physicists toolbox: the theory of angular momentum. Unitary designs are a rich and fundamental mathematical topic, with numerous fruitful applications in quantum information science and technology. In this work we take a peek under the hood. We begin with a minimal set of definitions and characterizations. Then we derive all 1-designs of ${\rm U}(2)$ of minimum size. Finally, we set out, step by step, a completion procedure extending such 1-designs to 2-designs. In particular, starting from the Pauli basis $\unicode{x2014}$ the prototypical unitary 1-design $\unicode{x2014}$ one $\unicode{x201C}$naturally$\unicode{x201D}$ obtains the 2-design originally employed by Bennett and coauthors in $\textit{Mixed State Entanglement and Quantum Error Correction}$. The present work also serves as a gentle and largely self-contained introduction to the subject.

comments
pdfTeX-1.40.27, 42 pages, 5 figures, 1 table
journal
Phys. Scr. 101, 305201 (2026)
fields
Mathematical Physics
pdf ↗ page arXiv:2509.17127 ↗

We prove that it is decidable whether a one-sided shift of finite type is conjugate to a one-sided Hom-shift, and whether a tree-shift of finite type is conjugate to a Hom tree-shift. The proof uses Williams's theory for one-sided shifts

fields
Dynamical Systems
pdf ↗ page arXiv:2509.24754 ↗

Semantic communication shifts the focus from bit-level accuracy to task-relevant semantic delivery, enabling efficient and intelligent communication for next-generation networks. However, existing multi-modal solutions often process all available data modalities indiscriminately, ignoring that their contributions to downstream tasks are often unequal. This not only leads to severe resource inefficiency but also degrades task inference performance due to irrelevant or redundant information. To tackle this issue, we propose a novel task-aware distributed information bottleneck (TADIB) framework, which quantifies the contribution of any set of modalities to given tasks. Based on this theoretical framework, we design a practical coding scheme that intelligently selects and compresses only the most task-relevant modalities at the transmitter. To find the optimal selection and the codecs in the network, we adopt the probabilistic relaxation of discrete selection, enabling distributed encoders to make coordinated decisions with score function estimation and common randomness. Extensive experiments on public datasets demonstrate that our solution matches or surpasses the inference quality of full-modal baselines while significantly reducing communication and computational costs.

fields
Information Theory
pdf ↗ page arXiv:2510.04000 ↗

A quaternionic matrix-valued regular function is a map $F: Ω\rightarrow M_n(\mathbb{H})$ whose entries are (left) regular functions of a quaternionic variable, where $Ω$ is a domain in $\mathbb{H}$. Given any $n \times n$ matrix-valued regular function $F(q)$ over $\mathbb{H}$ having norm at most $1$, we derive a decomposition of $F(q)$ in the form $U \star \begin{bmatrix} ||F(q_0)|| & 0\\0 & G(q)\end{bmatrix} V$ for some constant unitary matrices $U, V$, some vector $q_0$ and an $(n-1) \times (n-1)$ matrix-valued regular function $G(q)$, where $\star$ denotes the star product over $\mathbb{H}$. We use this to prove a Fisher-type approximation result for matrix-valued regular functions $F(q)$ on the quaternionic unit ball $\mathbb{B}$: if $F(q)$ has norm at most $1$ and is continuous on the boundary of the ball, then $F(q)$ can be uniformly approximated by a convex combination of matrix-valued rational unitary functions on $\overline{\mathbb{B}}$.

fields
Functional Analysis, Complex Variables
pdf ↗ page arXiv:2510.07866 ↗

Symplectic Lefschetz fibrations can be described via classifying maps with values in the Deligne-Mumford compactification of the moduli space of curves, by means of constructions relying on symplectic geometry. In this note we prove the existence of classifying maps for achiral Lefschetz fibrations using Riemannian geometry. We further extend Smith's signature formula to the achiral case, providing a more elementary proof of this statement.

comments
22 pages, multiple minor fixes and a fixed gap in the proof of thm 4.1
fields
Geometric Topology
pdf ↗ page arXiv:2510.18593 ↗

We develop a derived-geometric framework for Donaldson–Thomas invariants in semistable Calabi–Yau threefold degenerations. Derived restriction from the two smooth components of the central fiber to their common anticanonical divisor is Lagrangian. We give explicit universal-family diagrams for this structure, compare the deformation complexes with perfect complexes on the singular union, and construct the fourfold duality using a canonical bundle pulled back from the base. On specified Quot presentations we describe the real Lagrangian foliation and its shifted potential. We formulate proper-completion and cohomological-comparison conjectures for sheaves arising from sections of an extending line bundle. Their specialization is described by oriented two-sided bar complexes and a finite Gauss–Manin system.

comments
Revised the article based on the comments of the referee. 89 pages. Comments are very welcome!
fields
Algebraic Geometry, Mathematical Physics
pdf ↗ page arXiv:2510.20325 ↗

We consider the empirical eigenvalue distribution for a class of non-Hermitian random block tridiagonal matrices $T$ with independent entries. The matrix has $n$ blocks on the diagonal and each block has size $\ell_n$, so the whole matrix has size $n\ell_n$. We assume that the nonzero entries are i.i.d. with mean 0, variance $(3\ell_n)^{-1}$ and having finite moments of all orders. We prove that when the entries have a bounded density, then whenever $\lim_{n\to\infty}\ell_n=\infty$ and $\ell_n=O(\operatorname{Poly}(n))$, the normalized empirical spectral distribution of $T$ converges almost surely to the circular law. The growing bandwidth condition $\lim _{n\to\infty}\ell_n=\infty$ is the optimal condition for the circular law with small bandwidth. This confirms the folklore conjecture that the circular law holds whenever the bandwidth increases with the dimension; earlier circular-law proofs required the delocalized regime $\ell_n\gg n$.

comments
50 pages. Major revision. To appear in PTRF
fields
Probability
pdf ↗ page arXiv:2511.01744 ↗

We investigate the tensor-triangular geometry of the categories of isotropic Tate motives, isotropic Artin motives and isotropic Artin–Tate motives. In particular, we study the categories $DTM_{gm}(k/k;\mathbb{F}_2)$, $DAM_{gm}(k/k;\mathbb{F}_2)$ and $DATM_{gm}(k/k;\mathbb{F}_2)$ where $k$ is a flexible field and we fix $\mathbb{F}_2$-coefficients. In this case, we prove their Balmer spectra are singletons. The proof of this relies on the fact that the categories in question are generated by objects whose morphisms are `nilpotent enough.' Furthermore, we investigate their homological spectra, and verify Balmer's Nerves of Steel conjecture in these cases by showing these spaces are also singletons, and we give an explicit description of the homological primes. We also investigate stratification, and we conclude by studying generation properties of these categories -- the latter results we see also apply to global (i.e. non-isotropic) motives.

comments
27 pages, 1 figure; accepted version
journal
Sparks, F. (2026), On the tensor-triangular geometry of isotropic motives. J. London Math. Soc., 114: e70752
fields
Algebraic Geometry, Algebraic Topology, Category Theory
pdf ↗ page arXiv:2511.05922 ↗

In this paper, we study the dynamics of set-valued maps whose graphs are open and such that the image of each point is an open and connected set. Building upon the work of P. Duarte and M. Torres, who introduced and analyzed the combinatorial structure of the final recurrent set, we incorporate the notions of transitivity and mixing, thereby bringing their framework into the spirit of Smale's spectral decomposition theorem. We also demonstrate that such maps exhibit infinite topological entropy, and we establish a connection between transitive open set-valued maps and transitive Anosov diffeomorphisms.

comments
19 pages and 2 figures. Published in the journal Nonlinearity
fields
Dynamical Systems
pdf ↗ page arXiv:2511.07642 ↗

We derive an expression for the joint distribution function of the first jump times of a continuous state and continuous time branching process with immigration (CBI process) with jump sizes in given Borel sets having finite total Lévy measures, where by total Lévy measure, we mean the sum of the measures appearing in the branching and immigration mechanisms of the CBI process in question. Our result generalizes a corresponding result of He and Li (2016), who considered this problem in case of a single Borel set having finite total Lévy measure.

comments
47 pages
fields
Probability
pdf ↗ page arXiv:2512.04935 ↗

We construct and study solvable lattice models whose partition functions are Demazure characters. Two types of colored generalizations of the five-vertex model can be defined that are called the open and closed models. States of both models are associated with Gelfand-Tsetlin patterns, or equivalently, semistandard Young tableaux. We characterize the tableaux arising from states of the closed model and prove that they form a Demazure crystal. This is stronger than identifying the partition function as a Demazure character: it requires constructing Kashiwara crystal operators on the states through lattice moves and recoloring. For the open model, Brubaker, Bump, Buciumas, and Gustafsson showed that the states realize (crystal) Demazure atoms. We also give a new, independent proof of their result. Bump and Naprienko considered bosonic models involving a parameter t. In these models, the edges and vertices are constructed by fusion of simpler edge and vertex types. Taking t=0 produces bosonic analogs of the open model. By reversing the fusion order, we construct closed bosonic models and establish their Yang-Baxter equations, yielding a Demazure-Lusztig recursion for the partition functions. At t=0, these models specialize to bosonic analogues of the closed five-vertex model, and we prove that their states form Demazure crystals.

comments
revised on Oct29, 34 pages
fields
Combinatorics, Representation Theory
pdf ↗ page arXiv:2512.05479 ↗

Within the framework of Finite Element Systems (FES) we define a numerical method for computing differential forms, inspired by the Discrete de Rham (DDR) method. The discrete $L^2$ product of DDR is given a new interpretation, as a computable approximation of the true $L^2$ product on local spaces of forms defined by partial differential equations. Without significantly modifying the numerical method itself, this point of view provides an alternative approach to the analysis. The conformity and consistency properties we obtain are stronger than those previously shown, even in low dimensions. We can also recover some of the other results that have been proved about DDR, from those that have already been proved, in principle, in the general context of FES. Parts of the analysis involve a conjectural hypothesis of stability of constants, with respect to small mesh perturbations, that we detail. We also bring VEM, the Virtual Element Method, into the discussion. For convenience, background material on the FES framework is reviewed.

comments
v1: 20 pages, v2: 27 pages, v3: 37 pages
fields
Numerical Analysis
pdf ↗ page arXiv:2512.05912 ↗

We investigate the dimension dependence of Bramble–Pasciak–Xu~(BPX) preconditioners for simplicial $\mathbb{P}_1$ finite element discretizations of high-dimensional partial differential equations and establish that the condition numbers of BPX-preconditioned systems grow only polynomially with the spatial dimension. Our analysis requires a careful derivation of the dimension dependence of several fundamental tools in the theory of finite element methods, including elliptic regularity, the Bramble–Hilbert lemma, trace inequalities, and inverse inequalities. We further analyze an averaged Scott–Zhang-type quasi-interpolation operator, and show that its associated constants scale polynomially with the dimension. Building on these ingredients, we prove a multilevel norm equivalence theorem and derive a BPX preconditioner with explicit polynomial bounds on its dimensional dependence. The analysis is motivated in part by recent tensor and quantum finite element methods, where dimension-explicit conditioning estimates for BPX preconditioners play an important role.

comments
34 pages, 0 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2512.06166 ↗

This paper addresses two fundamental problems posed by Qi regarding the sufficiency of eigenvalues for the classification of symmetric tensors in the two-dimensional setting. For $2\times2\times2$ and $2\times2\times2\times2$ complex symmetric tensors, we classify their equivalence classes using canonical forms of their associated binary cubics and quartics. We then determine the number of eigenpair classes and the multiplicities of zero eigenvalues for the canonical forms over the complex domain. We demonstrate that these spectral counts do not distinguish all real equivalence classes. Finally, we extend this approach to derive canonical forms and classification for real third- and fourth-order linear partial differential equations in two variables using their bijective relationship to binary forms.

comments
An earlier version was published in Communications on Applied Mathematics and Computation (doi: 10.1007/s42967-026-00591-w). This revision corrects errors in the published version
fields
Rings and Algebras
pdf ↗ page arXiv:2512.17431 ↗

We study complete four-dimensional Riemannian metrics on connected manifolds that are critical for the curvature functional $\A(g)=\int_M|\Rm|^2\,\vol$ and have finite curvature energy. If $\langle\Ract\E,\E\rangle\ge0$, where $\E=\Ric-\frac14\Scal\,g$ and $\Ract$ is the curvature action on symmetric two-tensors, then the metric is either Einstein or locally isometric to a product of surfaces of constant Gaussian curvatures $c$ and $-c$, with $c\ne0$. This extends the corresponding compact classification to the complete setting with finite curvature energy.

comments
12 pages. Revised version with updated notation, expanded discussion of the curvature assumption, and a counterexample showing that this assumption cannot be omitted. Title slightly shortened
fields
Differential Geometry
pdf ↗ page arXiv:2512.18758 ↗

We develop a finite-dimensional sensitivity framework for studying stability in learning systems whose states include representations, parameters, and update variables. The central object is the \emph{Learning Stability Profile}, a collection of directional sensitivity operators that records how perturbations in inputs, parameter initialization, and update mechanisms propagate along a specified learning trajectory. The main result is a Lyapunov criterion for controlling this profile. Under explicit regularity, coercivity, and dissipation assumptions, an incremental Lyapunov energy yields uniform or exponentially decaying bounds on the associated linearized transition operators. The result is stated as a sufficient stability criterion, not as an unconditional converse theorem. The framework also distinguishes terminal decay, profile-wise boundedness, and subexponential growth, avoiding the identification of nonpositive growth exponents with uniform boundedness. The profile is then specialized to several standard learning mechanisms. Spectral bounds give forward sensitivity estimates for feedforward networks. Dissipativity and step-size restrictions give stability bounds for residual architectures. Mean-square contraction assumptions yield parameter and update sensitivity bounds for stochastic gradient methods. Locally Lipschitz systems, including piecewise-linear networks, proximal maps, projected updates, and recurrent or state-space recursions, are handled through Clarke generalized Jacobians and variational Lyapunov inequalities. The resulting framework provides a common stability language for architecture, optimization, stochasticity, and nonsmoothness. Its role is structural: it organizes known stability mechanisms within one perturbation calculus while keeping the hypotheses needed for each guarantee explicit.

comments
The stochastic sensitivity analysis needs stronger justification when moving from fixed-perturbation expectations to expected operator-norm bounds. The Lyapunov framework remains relevant and potentially repairable
fields
Dynamical Systems, Optimization and Control
pdf ↗ page arXiv:2512.21208 ↗

An orthogonal n-frame is an ordered set of n pairwise orthogonal vectors. The set of all orthogonal n-frames in a d-dimensional quadratic vector space is an algebraic variety V(d,n). In this paper, we investigate the singularities and components of the variety V(d,n), as well as the algebraic properties of the quadratic ideal I(d,n) generated by the orthogonality relations, which cuts out V(d,n) set-theoretically. By a general result of Ananyan and Hochster, these objects satisfy increasingly strong regularity conditions when d becomes large compared to n; we establish optimal effective bounds to quantify this behavior. We give criteria for the ideal I(d,n) to be prime or a complete intersection, thereby resolving open problems proposed by Conca and Welker. We give criteria for the variety V(d,n) to be normal or factorial. We provide strong evidence in support of the conjecture that I(d,n) defines V(d,n) scheme-theoretically. As applications, we improve the existing bounds for when the Lovász-Saks-Schrijver ideals of simple graphs satisfy these four properties.

comments
Final version. To appear on Transactions of the American Mathematical Society
fields
Commutative Algebra, Algebraic Geometry
pdf ↗ page arXiv:2512.25058 ↗

It is well-known that every tournament has a spanning path. We consider hypergraph analogues. In an \emph{$r$-uniform fully directed hypergraph}, or \emph{$r$-digraph}, every edge is a list or $r$ distinct vertices. An $(r,k)$-tournament is an $r$-digraph $G$ such that for every $r$-set $S$ of vertices in $G$, exactly $k$ of the orderings of $S$ are edges in $G$. A \emph{directed tight path} is an $r$-digraph $G$ whose vertices can be ordered so that the intervals of size $r$ are the edges in $G$. Let $f(n,r,k)$ be the maximum $s$ such that every $n$-vertex $(r,k)$-tournament contains a tight path on $s$ vertices. Since every tournament has a spanning path, we have $f(n,2,1)=n$. In this paper, we show that the minimum $k$ such that $f(n,r,k)$ tends to infinity with $n$ is in the interval $\left[\left(1-\frac{1}{r}-O(\frac{\log r}{r^2\log\log r})\right)r!, ~\left(1-\frac{1}{r} - \frac{\varphi(r)-1}{r!}\right)r!\right]$ where $\varphi(r)$ is the Euler Totient Function, and we find the exact value when $r\le 5$. We also show that $Ω(\sqrt{\log n/\log \log n}) \le f(n,3,3) \le O(\log n)$ and $f(n,3,4)\ge Ω(n^{1/5})$.

comments
18 pages, 5 figures; (v2) minor revisions; (v3) added an appendix; (v4) minor revisions
fields
Combinatorics
pdf ↗ page arXiv:2601.00144 ↗

In this paper, we introduce Mackey functors for a t.d.l.c. group and define the cohomological dimension of this group over the Mackey category. We then compare this dimension to the rational discrete cohomological dimension defined by Castellano and Weigel, as well as to the Bredon cohomological dimension of that t.d.l.c. group with respect to the family of compact open subgroups. We also extend results about the geometric dimension of a t.d.l.c. group.

comments
Relevant changes in Definition 3.3, Proposition 4.2 and Corollary 5.5. Some details have been added
fields
Group Theory
pdf ↗ page arXiv:2601.12958 ↗

In many learning tasks, certain requirements on the processing of individual data samples can arguably be formalized, in an intelligible way, as strict constraints of the underlying optimization problem, rather than by means of arbitrary penalties that often require heavy trial-and-error tuning. In this paper we show that, in these scenarios, learning can be carried out exploiting a sequential penalty method that allows to explicitly deal with constraints. For the proposed algorithm we show that, under classical assumptions and in the considered learning scenario, penalty subproblems can provably be solved in finite time with suitable guarantees of approximate stationarity in expectation; we also show that, when the latter condition is guaranteed for the subproblems, the overall sequential scheme possesses almost sure convergence properties to KKT-stationary points of (general) constrained problems. The results of experiments on sandbox and real-world image processing tasks show that the method is indeed viable to be used in practical deep learning scenarios.

journal
Transactions on Machine Learning Research, 2026
fields
Optimization and Control
pdf ↗ page arXiv:2601.16812 ↗

Plavnik, Schopieray, Yu, and Zhang have drawn attention to those (automatically premodular) fusion subcategories of modular fusion categories which are submodules for the Galois action on the ambient category. In particular, they showed that a subcategory is a Galois submodule if and only if its centralizer is integral. In the other direction, Schopieray has conjectured that every premodular fusion category can be embedded as a Galois-closed subcategory of a modular category; Schopieray calls such an embedding a \emph{Galois-modular extension}. We prove Schopieray's conjecture for pseudounitary categories; in the nonpseudounitary case, we prove that every braided fusion category admits a "Galois nondegenerate extension" but we do not know it to be modular. Along the way we record some general comments about the minimal nondegenerate extension problem for braided fusion categories.

comments
9 pages. v2 removes a pseudounitarity hypothesis
journal
Annals of Representation Theory, Volume 3 (2026) no. 3, pp. 579-588
fields
Quantum Algebra
pdf ↗ page arXiv:2601.23192 ↗

We develop a combinatorial and order-theoretic framework for shuffles, understood as ordered partitions of the natural numbers. We introduce decomposable shuffles as generalised sums of finite ordinals, copies of $ω$, and copies of its dual $ω^*$, indexed by countable scattered linear orders. These building blocks are called benches, ladders and snakes, respectively. We introduce a subclass, layered shuffles, and for this subclass we construct an explicit finite-address representation which reduces comparison in the induced order on $\mathbb N$ to lexicographic comparison of integer tuples. We determine the order types of represented layered shuffles, establish that all represented orders are scattered, and define an involution that realises order duality. Finally, we introduce a hierarchical composition on bench-free injective uniform shuffles of fixed sign. For each sign, this operation yields a monoid graded by reduced degree. Its degree-one component is isomorphic to $\operatorname{Sym}(\mathbb N)$ and consists precisely of the invertible elements.

fields
Combinatorics, Logic
pdf ↗ page arXiv:2602.00461 ↗

We investigate the existence and nonexistence of traveling wave solutions near monotonic shear flows with non-constant background density for the two-dimensional inhomogeneous Euler equations in a finite channel. For any small $τ>0$, first, we construct nontrivial traveling waves with velocity and density in $H^{5/2-τ}$ and $H^{3/2-τ}$, respectively, showing that inviscid damping fails at these regularities. Second, when the distorted Rayleigh operator has no eigenvalues, we prove that such traveling wave solutions cannot exist in higher regularity spaces ($H^{5/2+τ}$ for velocity and $H^{3/2+τ}$ for density).

fields
Analysis of PDEs
pdf ↗ page arXiv:2602.00824 ↗

We develop a unified matrix-spectral framework for analyzing stability and interpretability in deep neural networks. Representing networks as data-dependent products of linear operators reveals spectral quantities governing sensitivity to input perturbations, label noise, and training dynamics. We introduce a Global Matrix Stability Index that aggregates spectral information from Jacobians, parameter gradients, Neural Tangent Kernel operators, and loss Hessians into a single stability scale controlling forward sensitivity, attribution robustness, and optimization conditioning. We further show that spectral entropy refines classical operator-norm bounds by capturing typical, rather than purely worst-case, sensitivity. These quantities yield computable diagnostics and stability-oriented regularization principles. Synthetic experiments and controlled studies on MNIST, CIFAR-10, and CIFAR-100 confirm that modest spectral regularization substantially improves attribution stability even when global spectral summaries change little. The results establish a precise connection between spectral concentration and analytic stability, providing practical guidance for robustness-aware model design and training.

comments
Theorem 4.2 incorrectly claims that an upper spectral scale controls NTK conditioning without a lower eigenvalue bound. Theorem 5.1 also omits the Jacobian's magnitude from its spectral-entropy sensitivity bound. These problems undermine the central theoretical contribution
fields
Dynamical Systems, Optimization and Control
pdf ↗ page arXiv:2602.01136 ↗

Modern learning systems often interpolate training data while still generalizing well, yet it remains unclear when algorithmic stability explains this behavior. We model training as a function-space trajectory and measure sensitivity to single-sample perturbations along this trajectory. We propose a contractive propagation condition and a stability certificate obtained by unrolling the resulting recursion. A small certificate implies stability-based generalization, while we also prove that there exist interpolating regimes with small risk where such contractive sensitivity cannot hold, showing that stability is not a universal explanation. Experiments confirm that certificate growth predicts generalization differences across optimizers, step sizes, and dataset perturbations. The framework therefore identifies regimes where stability explains generalization and where alternative mechanisms must account for success.

comments
Theorem 5.2 does not establish the required quantitative neighbouring-dataset instability from its cited benign-overfitting setting. The claimed necessity direction and sharp stability boundary therefore lack a complete proof
fields
Optimization and Control
pdf ↗ page arXiv:2602.03514 ↗

Given a smooth, complete Riemannian manifold $M$ with bounded Ricci curvature and positive injectivity radius, we derive a sharp Sobolev inequality for the embedding of $W^{1,p}(M)$ into $L^{\frac{np}{n-p}}(M)$, when $1\le p< n$. We will first reduce the inequality to functions having support with small enough volume. In turn, we will show that the inequality for small volumes is implied by a first order uniform asymptotic expansion of the isoperimetric profile for $M$, for small volumes. We will then show that such an expansion follows from a local, uniform Sobolev inequality for functions in $W^{1,1}$, having support with small enough diameter.

comments
29 pages. In v2 several minor errors and typos have been corrected, and some small aesthetical changes and clarifications have been made. The generalized Polya-Szego inequality is now stated more precisely in theorem form (see Proposition 1), and the original outline of its proof has been expanded and fine-tuned in the Appendix
fields
Analysis of PDEs, Differential Geometry
pdf ↗ page arXiv:2602.06648 ↗

Given a Borel class of trees, we show that there is a tree in that class whose Scott sentence is not too much more complicated than the definition of the class. In particular, if the class is definable by a $Π_α$ sentence, then there is a model of Scott rank at most $α+ 2$. This gives another proof-and one that does not require first proving Vaught's conjecture for trees-of the fact that trees are not faithfully Borel complete.

fields
Logic
pdf ↗ page arXiv:2602.07166 ↗

We establish Sobolev and Moser-Trudinger inequalities with best constants on noncompact Riemannan manifolds with Ricci curvature bounded below, and positive injectivity radius.

comments
27 pages. In v2 a few parts of the introduction have been slightly revised or improved. The statement of Talenti's comparison theorem is now stated more precisely in an a priori form in Proposition 1, and the original outline of its proof has been expanded and fine-tuned in the Appendix. Minor errors and typos have been corrected
fields
Analysis of PDEs, Differential Geometry
pdf ↗ page arXiv:2602.09677 ↗

We investigate homological and depth-theoretic properties of finitely generated modules of finite projective dimension over Noetherian local rings. A central theme is the study of criteria for freeness and reflexivity derived from the torsion-freeness or reflexivity of tensor products of the form \( M \otimes_R M \) and \( M \otimes_R M^* \). Under mild homological assumptions, we prove that such properties of these tensor products impose strong structural constraints on \( M \), often forcing it to be free. These results generalize classical theorems of Auslander beyond the regular case. The second part of the paper is devoted to the dimension and support of Ext-modules, particularly \( \operatorname{Ext}^i_R(M, R) \) for critical values of \( i \), when \( M \) has finite projective dimension. We establish sharp bounds on their Krull dimensions, analyze their behavior for prime and equidimensional modules, and relate these findings to the grade conjecture and other homological conjectures, i.e., whenever $\operatorname{grade}(M) = \operatorname{ht}(\operatorname{Ann}(M))$ where Gdim$(M)<\infty$. We consider the problem that asks whenever is \( \pd_R(M \otimes_R N) = 1 \)? Applications include new cases of a question of Jorgensen, which asks whether \( \operatorname{pd}(M) < i \) whenever \( \operatorname{Ext}^i_R(M, M) = 0 \) and \( M \) has finite projective dimension over a complete intersection ring. Finally, we examine the projective dimension of prime ideals in rings that fail chain conditions.

fields
Commutative Algebra
pdf ↗ page arXiv:2602.09899 ↗

Electric vehicle (EV) adoption in cold regions is hindered by degraded EV charging performance at low temperatures, which necessitates effective battery thermal management during charging. Given the coupling of battery charging and heating dynamics, this paper explores the benefits of online coordinated charging and heating control, rather than performing them separately. Specifically, we first build queue models for both battery charging and thermal dynamics. Then, we formulate an optimization problem to minimize the system cost of a charging station, which allows us to coordinate charging and heating through maintaining queue stability. To solve the problem, we develop an online coordinated charging and heating control algorithm using Lyapunov optimization techniques. The proposed algorithm is prediction-free and has provable feasibility and optimality guarantees. Numerical experiments based on real-world data demonstrate the effectiveness and robustness of our proposed method compared to existing benchmarks.

comments
11 pages, 7 figures
fields
Optimization and Control
pdf ↗ page arXiv:2602.11561 ↗

When sampling multi-modal probability distributions, correctly estimating the relative probability of each mode, even when the modes have been discovered and locally sampled, remains challenging. We test a simple reweighting scheme designed for this situation, which consists in minimizing (in terms of weights) the Kullback-Leibler divergence of a weighted (regularized) empirical distribution of the samples with respect to the target measure. We show in a theoretical toy model that the convergence in terms of the number of samples follows the Monte Carlo rate. Numerical experiments on a variety of simple benchmarks are provided, exploring the effect of the ambient dimension, number of modes, number of samples and definition of the modes. We find that, particularly in the situation of interest with well-identified separated modes, the method is able to correct the error in the estimation from the initial sampling of the relative probabilities of the modes.

fields
Statistics Theory, Probability
pdf ↗ page arXiv:2602.12027 ↗

We introduce a general equivalence relation on the global class of morphisms of a category that subsumes several classical notions of equivalence in mathematics. We show that group-action equivalence relations, pullback equivalence relations, and coarse analysis-operator-based equivalence relations on continuous Bessel families are special cases of this general definition.

fields
Category Theory, Functional Analysis
pdf ↗ page arXiv:2602.12363 ↗

We study the approximation of $L_p$-functions, $p\in (0,\infty]$, on cylindrical space-time domains $Ω_T:=[0,T]\times Ω$, $0<T<\infty$, $Ω\subset \mathbb{R}^d$ Lipschitz, $d\in \mathbb{N}$, with respect to continuous anisotropic space-time finite elements on prismatic meshes. In particular, we propose a suitable refinement technique which creates (locally refined) prismatic meshes with sufficient smoothness and the desired anisotropy, and prove complexity estimates. Furthermore, we define a (quasi-)interpolation operator on this type of meshes and use it to characterize the corresponding approximation classes by showing direct and inverse estimates in terms of anisotropic Besov norms.

fields
Numerical Analysis
pdf ↗ page arXiv:2602.14921 ↗

Predicting the response of an observed system to a known input is a fruitful first step to accurately control the system's dynamics. Despite the recent advances in fully data-driven algorithms, the most interpretable way to reach this goal is through mechanistic mathematical modeling. Here, we leverage optimal control and propose a closed-loop iterative method to choose among a set of candidate models the one that most accurately predicts an observed system. We assume that one has control over an input of the observed system and access to measurements of its response. Our approach is to identify the input control that maximally discriminates the response of the candidate models, allowing us to determine which model is best by comparing such responses with the observed data. We demonstrate our proposed framework in numerical simulations before applying it during an electrophysiology experiment, successfully discriminating between different models for photocurrents produced via opsin dynamics.

comments
24 pages, 15 figures
fields
Dynamical Systems
pdf ↗ page arXiv:2603.00709 ↗

Quasi-Monte Carlo (QMC) is widely used in computational finance, and its efficiency strongly depends on the smoothness and effective dimension of the integrand. In this work, we propose a three-step approach, referred to as the IS\_AS\_PI method, which sequentially applies importance sampling (IS), the active subspace (AS) method and preintegration (PI). The proposed method overcomes the problem of empirical failure in constructing active subspaces when nonzero payoff gradients are rarely sampled. Here, empirical failure means that the estimated gradient information matrix is zero and thus provides no useful directions. We establish the orthogonal invariance of active subspaces and IS-active subspaces under suitable conditions. The method is applied to pricing Asian and basket options and estimating the Delta of Asian options under the Black–Scholes model. Numerical experiments show that the proposed method avoids this empirical failure in the deep out-of-the-money cases. Moreover, in all experiments, the proposed method with Laplace IS achieves the largest variance reduction factors, indicating the best performance, among the methods compared.

fields
Numerical Analysis
pdf ↗ page arXiv:2603.01763 ↗

We present two results regarding the relation of the index of equilibria and the index of fixed points, and the index of equilibria and payoff-robustness: any integer can be the index of an isolated completely mixed equilibrium of a finite game. In a particular class of isolated completely mixed equilibria, called monogenic, the index can be 0, +1 or -1 only. In this class non-zero index is equivalent to payoff-robustness. We also discuss extensions of the method of computation to extensive-form games, and cases where the equilibria might be located on the boundary of the strategy set. These results follow from a new method of computing indices of equilibria, which we also develop in this paper, that is based on algebraic properties of the system of polynomials describing completely mixed equilibria.

fields
Algebraic Geometry
pdf ↗ page arXiv:2603.04298 ↗

In many practical applications, prior knowledge about unknown parameters is available in the form of ordering constraints, and incorporating such information into the inferential procedure can improve the efficiency of the resulting estimators. In this paper, we consider the problem of estimating Shannon's entropy of two normal populations with ordered location parameters under a general location-invariant loss function. We derive the maximum likelihood estimator, the restricted maximum likelihood estimator, and the uniformly minimum variance unbiased estimator. Furthermore, we obtain an estimator that dominates the best affine equivariant estimator. In addition, we propose a smooth estimator that improves upon the best affine equivariant estimator. Moreover, we obtain improved estimators with respect to the Pitman closeness criterion. We consider interval estimation and construct several confidence intervals, such as asymptotic confidence intervals, bootstrap confidence intervals, highest posterior density credible intervals, and the intervals based on generalized pivotal quantities. A Monte Carlo simulation study is conducted to assess the performance of the proposed point and interval estimators. Finally, the proposed methods are illustrated by using two real data sets.

comments
38 pages, 18 figures, 6 tables, 41 references
fields
Statistics Theory
pdf ↗ page arXiv:2603.07971 ↗

Optimizing an implicational base of a closure system consists in turning this implicational base into an equivalent one with premises and conclusions as small as possible. This task is known to be hard in general but tractable for a number of classes of closure systems. In particular, several classes of convex geometries are known to have tractable optimization, while the problem was recently claimed to remain hard in general convex geometries. Continuing this line of research, we give a characterization of the optimum bases of a convex geometry in terms of what we call quasi-closed hypergraphs. We then use this characterization to show that when each quasi-closed hypergraph has disjoint edges, any implicational base of the convex geometry can be optimized in polynomial time with existing minimization and reduction algorithms. Finally, we prove that this property applies to double-shelling convex geometries, $D$-geometries, convex geometries with the $n$-Carousel property and $q$-acceptant convex geometries, thus unifying and extending the existing results regarding the tractability of optimization.

comments
Long version, 28 pages with appendices, 14 figures
fields
Combinatorics
pdf ↗ page arXiv:2603.14615 ↗

In this paper, we extend hitting times for imprecise Markov chains to the framework of weighted imprecise Markov chains (WIMCs), in which each transition is associated with a strictly positive weight encoded by a matrix $W$. Given a convex set $\mathcal{T}$ of admissible transition matrices, we define lower and upper expected hitting times for WIMCs as the infimum and supremum of the (weighted) expected hitting times over $\mathcal{T}$, and we characterise these quantities as the unique solutions of nonlinear fixed-point equations. We show that any weighted hitting time problem can be transformed into an unweighted hitting time problem on an augmented state space, enabling the reuse of existing IMC theory and algorithms. In particular, we are able to adapt known iterative methods for the numerical computation of lower and upper hitting times for WIMCs.

comments
Published for International Conference on Soft Methods in Probability and Statistics (SMPS)
fields
Probability
pdf ↗ page arXiv:2603.16665 ↗

In this paper, we investigate the relation between hitting times and hitting probabilities in discrete-time imprecise Markov chains (IMCs). We define lower and upper hitting times and probabilities for IMCs whose set of transition matrices $\T$ is compact, convex, and has separately specified rows. Building on reachability-based partitions of the state space, we prove two key implications: (i) finiteness of the upper expected hitting time entails the lower hitting probability equals one, and (ii) finiteness of the lower expected hitting time entails the upper hitting probability equals one. We further show an equivalence: the upper expected hitting time is finite if and only if the lower hitting probability is one. Finally, by presenting a counterexample, we show that the converse of the second implication can fail.

comments
Published for International Conference on Soft Methods in Probability and Statistics (SMPS)
fields
Probability
pdf ↗ page arXiv:2603.16675 ↗

We show that the classifying space of the flow category of a \emph{tame} Morse function on a smooth, closed manifold $M$ recovers the homotopy type of $M$, thereby addressing a claim in a preprint of Cohen–Jones–Segal. The tameness assumption is that the compactified moduli spaces of broken gradient trajectories are locally contractible, ensuring the flow category is topologically well-behaved. We construct a Morse function and Riemannian metric on $S^2\times S^1$ for which the associated flow category fails to recover the correct homotopy type, showing that the tameness hypothesis is crucial. Together, these results clarify the extent to which transversality assumptions can be relaxed so that the flow category models the homotopy type of the underlying manifold.

comments
v2: 37 pages, with minor revisions in response to referee comments. Final version to appear in Math. Z
fields
Algebraic Topology
pdf ↗ page arXiv:2603.23695 ↗

At the end, the house always wins! This simple truth holds for all public games of chance. Nevertheless, since lotteries have existed, people have tried everything to give luck a helping hand. This article compares objective scientific approaches to tackle the 6/49 lottery: probabilistic methods and combinatorial designs. The mathematical models developed herein can be modified and applied to other lotteries. The newly constructed (49, 6, 5) covering design is introduced, which meets the Schönheim bound. For lottery designs and for covering designs, a benchmark based on probabilistic methods is presented. It is demonstrated that common attempts to outwit the odds correspond to limitations of numbers to subsets, which disproportionately reduce the chances of winning.

comments
16 pages
fields
Combinatorics, Probability
pdf ↗ page arXiv:2603.24170 ↗

We derive a nonperturbative bound on the distance between evolutions of open quantum systems described by time-dependent generators. We show how this result can be employed to provide an explicit upper bound on the error of the rotating-wave approximation in the presence of dissipation and decoherence. We apply the derived bound to the strong-coupling limit in open quantum systems and to the secular approximation used to obtain a master equation of the Gorini-Kossakowski-Lindblad-Sudarshan form from the Redfield master equation.

comments
34 pages, 2 figures; improved error bounds for the secular approximation and clarified their time dependence and range of validity;
fields
Mathematical Physics
pdf ↗ page arXiv:2603.26606 ↗

While Physics-Informed Neural Networks offer a promising framework for solving partial differential equations, the standard $L^2$ loss formulation is fundamentally insufficient when applied to the Bhatnagar-Gross-Krook (BGK) model. Specifically, simply minimizing the standard loss does not guarantee accurate predictions of the macroscopic moments, causing the approximate solutions to fail in capturing the true physical solution. To overcome this limitation, we introduce a velocity-weighted $L^2$ loss function designed to effectively penalize errors in the high-velocity regions. By establishing a stability estimate for the proposed approach, we shows that minimizing the proposed weighted loss guarantees the convergence of the approximate solution. Also, numerical experiments demonstrate that employing this weighted PINN loss leads to superior accuracy and robustness across various benchmarks compared to the standard approach.

comments
29 pages, 12 figures
fields
Numerical Analysis
pdf ↗ page arXiv:2604.04971 ↗

This paper considers the problem of multi-bit generative watermarking for large language models under a worst-case false-alarm constraint. Prior work established a lower bound on the achievable miss-detection probability in the finite-token regime and proposed a scheme claimed to achieve this bound. We show, however, that the proposed scheme is in fact suboptimal. We then develop two new encoding-decoding constructions that attain the previously established lower bound, thereby completely characterizing the optimal multi-bit watermarking performance. Our approach formulates the watermark design problem as a linear program and derives the structural conditions under which optimality can be achieved. In addition, we identify the failure mechanism of the previous construction and compare the tradeoffs between the two proposed schemes.

comments
42 pages, 8 tables
fields
Information Theory
pdf ↗ page arXiv:2604.08759 ↗

Accurate and efficient representation of non-Gaussian state distributions is critical for estimation, control, and decision-making, especially when the system is highly nonlinear. This paper develops a mathematical framework for estimating non-Gaussian uncertainty via semi-nonparametric (SNP), or Gallant-Nychka, densities. SNP densities use squared probabilists' Hermite polynomials to guarantee nonnegative probability densities while representing non-Gaussian behavior. For uncertainty propagation, once fitted to propagated state samples, SNP densities enable direct analytical density and probability evaluations without needing to tune additional hyperparameters besides the polynomial order. To enable their application to nonlinear dynamical systems, we derive analytic marginal and cumulative distribution functions and develop a lossless convex formulation of the coefficient estimation problem along a given sign branch. The approach is demonstrated on the chaotic Lorenz system and compared against Gaussian mixture models and Edgeworth approximations. Results demonstrate that SNP densities can accurately represent non-Gaussian state distributions while enabling direct probability and quantile evaluation, making them a promising tool for density estimation and uncertainty quantification in nonlinear estimation and control.

fields
Optimization and Control, Statistics Theory
pdf ↗ page arXiv:2604.09375 ↗

We establish finite-step probabilistic upper bounds on the contraction ratios $ρ_k = Δ_{k+1}/Δ_k$ for iterated Pearson correlation dynamics. Let $(P_k)_{k\ge 0}$ be the sequence generated by the Pearson update. Define $Δ_k := \|P_{k+1}-P_k\|_F$, $ρ_k := Δ_{k+1}/Δ_k$ for $Δ_k > 0$, and $δ_k := Δ_k/n$. Although $Δ_k \to 0$ along convergent trajectories, the ratios $ρ_k$ may exceed unity in finitely many steps. This behavior is invisible to local linearization. Our main contribution is a probabilistic bounding framework that captures these finite-step expansions. We initialize $P_0$ with i.i.d. $\mathcal{U}[-1,1]$ entries and let $\mathbb{P}$ be the induced measure. For $k \ge 2$, we construct state-dependent bounds $B_p : \mathbb{R}_+ \to \mathbb{R}_+$ satisfying $\mathbb{P}(ρ_k \le B_p(δ_k)) \ge p$. The functions $B^{\mathrm{q}}_p(δ)$ are empirical conditional $p$-quantiles of $\log ρ_k$ given $δ_k$ under logarithmic binning. Larger families $B^{\mathrm{TC}}_{p,τ}(δ)$ and $B^{\mathrm{tol}}_{p,τ}(δ)$ are obtained via multiplicative adjustments, yielding pointwise larger bounds that preserve the $δ$-dependence. Validation on held-out trajectories confirms the bounds hold with empirical coverage matching nominal levels for all $n \in [3,2000]$. The baseline $0.95$-quantile bound $B^{\mathrm{q}}_{0.95}(δ)$ yields two concrete results: $\mathbb{P}(ρ\le 1 \mid δ\le 0.03) \ge 0.95$ uniformly in $n$, and $\mathbb{P}(ρ\le 1.7) \ge 0.95$ for 21 of 22 dimensions. The exception $n = 69$ attains $2.35$, revealing a rare extreme upper tail discontinuity not captured by asymptotic analysis. These are the first finite-step probabilistic bounds for Pearson correlation dynamics. The framework is fully reproducible with provided code and data.

comments
arXiv admin note: This submission has been withdrawn because it does not meet arXiv's research content quality standards
fields
Statistics Theory, Dynamical Systems
pdf ↗ page arXiv:2604.14071 ↗

The idea of a $\bigvee$-system was introduced by Veselov in the study of rational solutions of the WDVV equations of associativity. These are algebraic/geometric conditions on the set of covectors that appear in rational solutions to the WDVV equations. Here, this idea is generalized to open WDVV equations, which are an additional set of PDEs originating from open Gromow-Witten Theory. We develop -- for rank-one extensions -- algebraic/geometric conditions on the covectors that supplement the $\bigvee$-system to give rational solutions to the open WDVV equations. Examples, and the relation to superpotentials and to Dubrovin almost-duality, are given.

comments
22 pages; comments welcome! Version 2 - minor changes
fields
Mathematical Physics, Representation Theory
pdf ↗ page arXiv:2604.14077 ↗

We introduce \emph{joint exclusivity} (JE), a relaxation of mutual exclusivity (ME) for non-negative random vectors that forbids only the simultaneous positivity of all $n$ components. Whereas ME requires the probabilities of positive outcomes to sum to at most one, a JE random vector with prescribed marginals $F_1,\ldots,F_n$ exists if and only if $$ \sum_{i=1}^n \overline{F}_i(0)\le n-1. $$ An inclusion–exclusion identity shows that JE laws are determined by their $(n-1)$-dimensional margins and decomposes aggregate stop-loss premiums into those of sub-portfolios. A single wrap-around construction attains the existence bound, minimises the number of claims in convex order and, in frequency–severity models with exchangeable severities, the aggregate loss, extending the safest-dependence property of ME beyond its admissible range. We develop canonical and distortion-based constructions, an $m$-exclusivity hierarchy between ME and JE, and admissibility criteria for general families of claim patterns, explicit for hierarchical caps. Finally, we identify the most dangerous exclusive structure in convex order for lines sharing a severity law, which yields sharp stop-loss and Tail Value-at-Risk bounds.

fields
Statistics Theory
pdf ↗ page arXiv:2604.17490 ↗

This article investigates channel capacity under mismatched stochastic decoding. Feinstein- and Verdú-Han-type bounds on the error probability of coded communication are derived for sequences of $n$-letter channels and decoding metrics. These are used to obtain an information-spectrum formula for the corresponding channel capacity: the supremum, over all input distribution sequences, of the limit inferior in probability of the sequence of normalized mismatched information densities. This formula yields the mismatched analog of the Verdú-Han channel capacity formula. For discrete-memoryless channels and product metrics, uniform integrability allows to show that the limit product-space improvements of the generalized mutual information and the LM rate give the mismatch capacity under stochastic decoding. This establishes the stochastic decoding counterpart of the Csiszár-Narayan conjecture. Relationships among the capacity formulas of related decoders are investigated. A number of examples are discussed including the case of binary-input binary-output channels and decoding metrics, where capacity is characterized in single-letter form.

comments
Submitted to IEEE Transactions on Information Theory
fields
Information Theory
pdf ↗ page arXiv:2604.17964 ↗

We consider a Ziegler pair of plane arrangements, that is two plane arrangements $\mathcal{A}:f=0$ and $\mathcal{A}':f'=0$ in the projective space $\mathbb{P}^3$, such that the intersection lattices $L(\mathcal{A})$ and $L(\mathcal{A}')$ are isomorphic, but the Betti numbers of the minimal resolutions of their Jacobian algebras are not the same. We introduce several properties for such pairs and relate them to cones over Ziegler pairs of line arrangements in $\mathbb{P}^2$.

comments
16 pages, version 2.0
fields
Algebraic Geometry
pdf ↗ page arXiv:2604.25637 ↗

We propose a unifying framework for characterizing pure and mixed state phases of matter across equilibrium, non equilibrium, and metastable regimes. We introduce the concept of locally stable states, defined by the operational property that any local operation (including post selection) can be reversed by a local channel. We prove that local stability is equivalent to a state being short range correlated, defined by the decay of both correlations and conditional mutual information. We demonstrate that these properties are invariant under locally reversible channels, thus defining locally stable phases. Furthermore, we prove that local stability implies both the decay of a family of nonlinear correlators, including the fidelity correlator, and the decay of correlations in the canonical purification, thus bridging the gap between mixed and pure states. Along the way, we establish two results which may be of independent interest: we show that post-selection on locally stable (short range correlated) states can be implemented via local channels and that quantum Markov chains can be characterized by the local computability of nonlinear observables.

comments
17 + 16 pages, 5 + 1 figures; v2: strengthened Theorems 9, 11, Corollary 12 and added Corollary 13
fields
Mathematical Physics
pdf ↗ page arXiv:2605.00088 ↗

We develop a reduced-order framework for optimizing mixing in two-dimensional incompressible flows. Instead of optimizing the full transport PDE, the method maximizes the length of advected material interfaces, leading to a Hamiltonian control problem with a finite spatial basis based on parametrized stream functions. We derive the continuous adjoint equations and reduced gradients, and discretize the forward and adjoint dynamics with the implicit midpoint rule. The resulting discrete adjoint is algebraically consistent with the derivative of the fully discrete objective when the implicit equations are solved exactly. The approach applies to bounded two-dimensional domains with sufficiently regular finite-dimensional stream-function parametrizations. Numerical experiments on cellular-flow and Doswell frontogenesis benchmarks show interface stretching and mix-norm decay over the simulated horizons. In four comparisons, the interface runs have lower recorded optimization times, whereas the Sobolev controls attain smaller terminal mix-norms. The different constraints, stopping rules and timing scopes preclude an equal-accuracy speedup claim.

comments
28 pages
fields
Numerical Analysis
pdf ↗ page arXiv:2605.04688 ↗

Let $\mathfrak{J}$ be a JB$^*$-algebra with no quotients isomorphic to any spin factor. Let $μ$ be a local quasi-linear Jordan functional on $\mathfrak{J}_{sa}$. We show that $μ$ is a linear functional on $\mathfrak{J}_{sa}$ if and only if the restriction of $μ$ to the closed unit ball of $\mathfrak{J}_{sa}$ is uniformly weakly continuous.

fields
Operator Algebras
pdf ↗ page arXiv:2605.06120 ↗

Marginally trapped surfaces are spacelike surfaces in the Minkowski space whose mean curvature vector is lightlike at each point. In general, the marginally trapped surfaces are determined by seven functions satisfying several conditions (differential equations). In the present paper, we introduce special principal parameters, called canonical, and prove that every marginally trapped surface of general type admits (at least locally) canonical principal parameters which allow us to reduce the number of functions. We prove a Fundamental existence and uniqueness theorem formulated in terms of canonical parameters, which states that every marginally trapped surface of general type is determined up to a motion by three smooth functions satisfying a system of partial differential equations. The special subclass of marginally trapped surfaces with parallel mean curvature vector field is described in terms of only one differential equation. Finally, the theory is enriched with examples of marginally trapped surfaces for which canonical parameters are determined.

fields
Differential Geometry
pdf ↗ page arXiv:2605.08367 ↗

Transformers with self-attention modules as their core components have become an integral architecture in modern large language and foundation models. In this paper, we study the evolution of tokens in deep encoder-only transformers at inference time which is described in the large-token limit by a mean-field continuity equation. Leveraging ideas from the convergence analysis of interacting multi-particle systems, with particles corresponding to tokens, we prove that the token distribution rapidly concentrates onto the push-forward of the initial distribution under a projection map induced by the key, query, and value matrices, and remains metastable for moderate times. Specifically, we show that the Wasserstein distance of the two distributions scales like $\sqrt{{\log(β+1)}/β}\exp(Ct)+\exp(-ct)$ in terms of the temperature parameter $β^{-1}\to 0$ and inference time $t\geq 0$. For the proof, we establish Lyapunov-type estimates for the zero-temperature equation, identify its limit as $t\to\infty$, and employ a stability estimate in Wasserstein space together with a quantitative Laplace principle to couple the two equations. Our result implies that for time scales of order $\logβ$, which correspond to the early-phase behavior most relevant for practical finite-depth models, the token distribution concentrates at the identified limiting distribution. Numerical experiments confirm this and, beyond that, complement our theory by showing that for finite $β$ and large $t$ the dynamics enter a different terminal phase, dominated by the spectrum of the value matrix.

comments
34 pages, 11 figures
fields
Analysis of PDEs, Dynamical Systems
pdf ↗ page arXiv:2605.10931 ↗

We study the terminal behavior of singular deterministic and stochastic dynamics driven by convex or extrapolative combinations of logarithmic heat gradients. These equations arise in exact-score guided diffusion sampling, including classifier-free guidance. The rescaled potentials converge to weighted sums of squared support distances, but extrapolative mixing can produce limiting directions that depend on probability masses and lie outside the Clarke subdifferential of the limiting potential. A componentwise chain rule gives dissipation for the limiting differential inclusion. Posterior relative entropy yields finite deterministic kinetic action in logarithmic time, while a conditional free-energy estimate bounds expected energy increments uniformly in the restart time. For compactly supported laws, these estimates give energy convergence and equilibrium localization, almost surely for each fixed stochastic initialization; extrapolation requires a lower small-ball mass bound on the reference law. For finite atomic measures, we prove that the equilibrium set is finite for arbitrary site configurations. Every deterministic trajectory therefore converges to an equilibrium, and stochastic convergence holds almost surely for each fixed initialization, including endpoints at projection interfaces. At empirical endpoints with a unique nearest atom in each active support, deterministic trajectories admit a square-root heat-time expansion with an exponentially small remainder. The exponent 1/2 is sharp when the leading coefficient is nonzero. These results provide a geometric and dynamical framework for understanding terminal convergence in guided diffusion models.

comments
34 pages, 3 figures
fields
Optimization and Control
pdf ↗ page arXiv:2605.12231 ↗

The increasing availability of experimental data has intensified interest in calibrating stochastic models, raising fundamental questions about parameter identifiability. Structural identifiability determines whether parameters can be uniquely recovered from idealised data, a prerequisite to allow for parameter estimation in real-world scenarios. However, there is a lack of methodologies generally applicable to analyse the structural identifiability of such stochastic processes. Therefore, we develop a methodology for a class of spatio-temporal stochastic processes. We investigate how structural identifiability depends on the type of available data, distinguishing between single-particle trajectories and total particle density measurements. For trajectory data, we use a particle-based description that explicitly represents single-particle dynamics. For population-level data, we derive a partial differential equation representation, that describes the evolution of total particle density, and apply a differential algebra approach, common to ordinary differential equation analysis. We further introduce Taylor-series-expansion and Fourier-transform approaches to study the initial condition. We apply our methodology to an example model with a specified class of initial conditions. We show that the model is globally structurally identifiable given infinitely many, infinitely long single-particle trajectories, but only locally structurally identifiable from total particle density data, demonstrating that parameter identifiability depends on the type of data available.

comments
Main: 34 pages, 3 figures. Supplementary Information: 17 pages, 5 figures
fields
Analysis of PDEs, Dynamical Systems, Probability
pdf ↗ page arXiv:2605.13504 ↗

We introduce the median uniformity $\mathcal U_{\mathrm m}$, an intrinsic precompact convex uniform structure on a median algebra. It is Hausdorff under natural assumptions, for instance for finite-rank median algebras. In the Hausdorff case, its uniform completion yields the Minimal Median Compactification (MMC). The induced topology $τ_{\mathrm m}$ provides a natural higher-rank analogue of the interval topology on linearly ordered sets and of the shadow topology on rank-one median algebras. When the intrinsic topology $τ_{\mathrm m}$ is Hausdorff and all intervals are compact, the MMC is the unique proper median compactification and is canonically isomorphic to the Roller–Fioravanti compactification. We apply this uniform framework to continuous actions of a topological group $G$ by median automorphisms. We show that the MMC is a median $G$-compactification. In the finite-rank case, the resulting compact $G$-system is Rosenthal representable and dynamically tame.

comments
30 pages, 1 figure
fields
General Topology, Dynamical Systems, Functional Analysis
pdf ↗ page arXiv:2605.16096 ↗

Unbalanced optimal transport (UOT) provides a principled framework for modeling single-cell transitions and birth-death dynamics, but its high computational cost limits scalability to large-scale datasets. Although single-cell data often contain hierarchical annotations and known transition priors, existing UOT approximations rarely exploit this multiscale structure or prior knowledge. We introduce Multiscale Supervised Unbalanced Optimal Transport Flow Matching (MUST-FM), a simulation-free framework that scales UOT by leveraging hierarchical data structure. MUST-FM further supports an optional supervised formulation that incorporates transition priors, such as cell lineages, to guide the learning of displacement fields and mass variations. Experiments show that MUST-FM reduces computational overhead while achieving robust and biologically meaningful trajectory inference, enabling dynamic modeling of atlas-scale single-cell datasets. The code is available at https://github.com/QiangweiPeng/MUST-FM.

fields
Optimization and Control
pdf ↗ page arXiv:2605.16529 ↗

A pseudo-Euclidean Novikov superalgebra $A$ is a Novikov superalgebra endowed with a non-degenerate symmetric bilinear form $\langle, \rangle$ such that all left multiplication operators are $\langle, \rangle$-antisymmetric. In this case, the associated Lie superalgebra $(A^{-},\langle, \rangle)$ is a flat pseudo-Euclidean Lie superalgebra. In this paper, we investigate the structure of pseudo-Euclidean Novikov superalgebras. In particular, we introduce a distinguished subclass, called \emph{Milnor superalgebras}, and prove that any pseudo-Euclidean Novikov superalgebra whose two-sided ideal is non-degenerate belongs to this class. We provide a method for constructing pseudo-Euclidean Novikov superalgebras. We also introduce a double extension procedure for pseudo-Euclidean Novikov superalgebras and show that every such superalgebra with a degenerate two-sided ideal can be obtained via this method. Furthermore, we establish that any pseudo-Euclidean Novikov superalgebra is either a Milnor superalgebra or can be obtained by a sequence of double extensions starting from a Milnor superalgebra. As an application, we provide a complete classification of pseudo-Euclidean Novikov superalgebras of total dimension at most four over the field of real numbers.

comments
54 pages; major corrections; to appear in Bulletin des Sciences Mathematiques
fields
Rings and Algebras
pdf ↗ page arXiv:2605.20775 ↗

Some time ago, we published the full count of degrees of freedom in the linearised weak gravity limit of arbitrary New GR models. We did it by considering the linear weak gravity equations and presented a thorough analysis with no ambiguity left. A bit later, we generalised it to linear cosmological perturbations and discussed the strong coupling issues that appear already at this level. Recently, there were claims that some dynamical modes had been missed in our work. However, the authors of the new claims did not look at the equations of motion and analysed the quadratic Lagrangian densities instead. In this paper, I take one of the most elementary cases, namely the vector modes in New GR of Type 3, and show what was their mistake that had led them to claiming that those were dynamical. The main message: Do not substitute constraint equations into a Lagrangian.

comments
minor additions, the final version; 7 pages
fields
Mathematical Physics
pdf ↗ page arXiv:2605.22329 ↗

In this article we provide an integration formula enabling us to integrate random variables defined on the moduli space of hyperbolic surfaces which involve the lengths of closed geodesics of any prescribed topological type, at fixed genus and number of boundaries. We then compute the general form of length functions of closed loops in Fenchel-Nielsen coordinates. Using this expression together with our integration formula, we express the expectations of geometric random variables as integrals over R with almost polynomial density. As a consequence, given an arbitrary closed loop $γ$, we obtain an improvement of Mirzakhani's asymptotic equivalent of the Weil-Petersson expectation E[N$γ$(a)], when a $\rightarrow$ $\infty$, of the number of geodesics with same topological type as $γ$ of length at most a. We finally provide, for a loop $γ$ filling a one-holed torus or a four-holed sphere, an interpretation of the density function associated with $γ$ as the period of the Hamiltonian flow of the length function of $γ$.

fields
Geometric Topology
pdf ↗ page arXiv:2605.23626 ↗

The Floquet-Magnus expansion is a widely used tool to derive effective descriptions of time-periodic quantum systems by approximating their dynamics with a time-independent Hamiltonian. However, its standard formulation is, strictly speaking, restricted to bounded Hamiltonians. In this work, we extend its definition and analysis to a broad class of time-periodic unbounded Hamiltonians. Our approach is based on an a priori distinct nonperturbative framework for the construction of effective Hamiltonians, which we show to reproduce the Floquet-Magnus expansion. A particular strength of our framework is that it allows us to prove that the resulting effective dynamics approximates the original time evolution propagators to arbitrary order in the high-frequency limit without requiring convergence of the Floquet-Magnus expansion, a condition that is already highly restrictive even in the bounded setting. We illustrate the scope of the method on representative models: the quantum Rabi Hamiltonian in the interaction picture, and the periodically driven quantum harmonic oscillator.

comments
61 pages, revision, main results unchanged
fields
Mathematical Physics, Functional Analysis
pdf ↗ page arXiv:2605.23734 ↗

We exploit the connection between quantum dot Dirac operators and $\overline\partial$-Robin Laplacians. First, we find a graphical relation between their smallest positive eigenvalues, which allows us to deduce a recipe for translating bounds (from above and below) from one to the other. As an application, we provide new upper and lower bounds for the eigenvalues of the quantum dot Dirac operators, which depend only on geometric quantities of the underlying domain. In particular, we obtain some Faber-Krahn type inequalities for convex thin domains.

comments
20 pages, 3 figures. v2: added a Note before Section 1.1 discussing a proof of Dirac-Faber-Krahn that appeared during the peer review process of this work
journal
Complex Anal. Oper. Theory 20, 210 (2026)
fields
Analysis of PDEs
pdf ↗ page arXiv:2605.30228 ↗

We show that strong log-concavity emerges in probit regression likelihoods without ridge penalization (i.e. Gaussian priors), unlike for the logistic case. Specifically, we provide: (a) a non-asymptotic characterization of strong log-concavity for fixed designs, similar to that for the existence of the maximum likelihood estimator (MLE) and (b) an asymptotic analysis for random designs, in the proportional regime when the sample size $n$ and the number of covariates $d$ grow proportionally $d/n\rightarrow r\in[0,1)$. In the latter case we show that, provided $r$ is small enough, the resulting condition number is finite and bounded independently of $n,d,r$ with high-probability. Numerically tractable estimates are given in case $r=0$. Thus, probit regression provides a non-trivial example of a high-dimensional, well-conditioned log-concave objective.

fields
Statistics Theory
pdf ↗ page arXiv:2605.31218 ↗

We investigate several $σ$-ideals on the Baire space $ω^ω$ $(\mathbb{Z}^ω)$, introducing and studying the ideals $\mathcal{G}$ and $\mathcal{SMZ}^+$, alongside the classical ideals of meager sets, strong measure zero sets, the eventually different ideal and infinitely equal ideal We establish structural relationships and proper inclusions among these ideals. Also we compute the cardinal invariants of $\mathcal{M}_-$, proving that they are same as invariants of $σ$-ideal of meager sets. We further analyze the operation $^*$ on families of sets, establishing dual relationships such as $\mathcal{ED}^* = \mathcal{IE}$, $\mathcal{IE}^*=\mathcal{ED}$, $\mathcal{M}_-^*=\mathcal{SMZ}^+$ and $\mathcal{H}^* = \mathcal{G}$, and derive separations between ideals under additional set-theoretic assumptions. Finally, we prove a tree dichotomy theorem for the ideal $\mathcal{M}_-$ and we study the associated forcing notion.

fields
Logic
pdf ↗ page arXiv:2606.01180 ↗

We revisit the convergence guarantees of the Extragradient (EG) method for unconstrained biaffine min-max optimization. It is known that EG with a fixed stepsize achieves a $Θ(T^{-1/2})$ last-iterate convergence rate, which is slower than the optimal $\mathcal{O}(T^{-1})$ rate attainable by incorporating additional mechanisms such as anchoring. Motivated by recent advances showing that dynamic stepsizes alone can significantly accelerate gradient descent, we ask whether dynamic stepsizes can similarly accelerate the last-iterate convergence of EG. We present the first positive result in this direction. Specifically, we provide a deterministic dynamic stepsize schedule that accelerates the convergence rate of EG to $\mathcal{O}(T^{-2/3+\varepsilon})$ for any $\varepsilon > 0$. We also show that this rate is nearly tight when the extrapolation and update steps of EG use the same stepsize. We then show that allowing different stepsizes for the extrapolation and update steps further improves the convergence rate to the near-optimal $\mathcal{O}(T^{-1+\varepsilon})$. Our analysis reduces stepsize scheduling to an optimization problem, whose solution leads to a stepsize schedule that follows (a discretization of) a power-law distribution. Our proposed stepsize schedules and analysis extend to other methods, such as Optimistic Gradient (OG), and suggest broader applicability to general min-max optimization problems.

comments
58 pages. Accepted at NeurIPS 2026
fields
Optimization and Control
pdf ↗ page arXiv:2606.01764 ↗

We study local finiteness for modal $K4$ algebras via the tunability of their dual general frames. In particular, we provide a sufficient condition for modal $K4$ algebras to be locally finite by identifying a structure which must be present in non-locally finite modal $K4$ algebras. We then show that this condition becomes both necessary and sufficient for $σ$-complete modal $K4$ algebras. Next, we translate this condition into a pair of order-theoretic conditions on transitive Kripke frames, providing a classification of local finiteness on their dual modal algebras. We further show that the logic of any class of well-founded transitive relations with no infinite antichains has the finite model property, and conclude that the logic of the class of well-quasi orderings has the finite model property.

fields
Logic
pdf ↗ page arXiv:2606.02941 ↗

We construct a stacky resolution of singularities for certain noncommutative spaces which can be viewed as 'finite noncommutative extensions' of schemes. More precisely, we show that any order over a reduced separated finite type scheme over a field of characteristic zero can be resolved by an Azumaya algebra over a smooth Deligne–Mumford stack by a sequence of stacky blow-ups.

comments
17 pages. v2: Final version, to appear in J. Algebra
fields
Algebraic Geometry, Rings and Algebras, Representation Theory
pdf ↗ page arXiv:2606.05137 ↗

Physics-informed learning is increasingly used for partial differential equation (PDE)-governed inverse problems, but its reliability remains difficult to certify. This paper develops a no-harm certification-and-selection framework for physics-informed inverse learning. A learned reconstruction is accepted only when its residual-calibrated radius is no worse than the baseline radius, namely when $$R_{\mathrm{learn}}\le R_{\mathrm{base}}+\varepsilon_{\mathrm{safe}};$$otherwise, the method returns the baseline. The certificate combines data, physics, boundary or initial-condition, and optimization residuals. Under a conditional stability estimate, these residuals yield an a posteriori reconstruction-error bound and a deterministic uncertainty radius. A high-probability certificate is also derived for physics residuals estimated from independent random collocation points. Numerical tests on Poisson source recovery, inverse heat reconstruction, limited-angle tomography, elliptic coefficient identification, and stochastic residual validation show that the selector accepts certified improvements, rejects shifted, hallucinated, or unfinished candidates, and becomes conservative in strongly ill-posed regimes. The framework is therefore a certification-and-selection layer, not another reconstruction architecture.

comments
I need to clarify that a smaller certified upper bound does not necessarily imply a smaller actual reconstruction error. Strengthen the stability assumptions, calibration, and independent residual validation
fields
Numerical Analysis, Optimization and Control
pdf ↗ page arXiv:2606.07153 ↗

Let $G$ be the group of power series $x+a_2x^2+a_3x^3+\cdots\in R[[x]]$ under substitution, where $R$ is a commutative ring with $1\neq 0$ of prime characteristic $p$. Given any $n\geq 1$, the subgroup $K_n=\{x+a_{n+1}x^{n+1}+a_{n+2}x^{n+2}+\cdots\,|\, a_i\in R\}$ is normal in $G$, and the quotient $G_n=G/K_n$ is the group of truncated polynomials over $R$ of degree $\leq n$ under substitution. In this paper, we compute the exponent of the image of $K_r$ in $G_n$, for all $r,n\geq 1$, indicating in every case a family of elements realizing this exponent.

fields
Group Theory
pdf ↗ page arXiv:2606.11461 ↗

We study the topology of isometric classes of tensorial $H$-structures for closed and connected subgroups $H\leqslant\mathrm{SO}(n)$. The map assigning an $H$-structure its induced metric admits a global trivialization; consequently, the space of $H$-structures is homotopy equivalent to any fixed isometric class. On flat tori, these classes are mapping spaces into $\mathrm{SO}(n)/H$. We give examples with infinitely many connected components for $\mathrm{G}_2$, $\mathrm{Spin}(7)$, and almost Hermitian structures, determine the number of components for $\mathrm{SU}(m)$-structures on flat tori, and study the corresponding moduli spaces. For closed manifolds of dimension $n>2$, the intrinsic torsion energy on the full space of $H$-structures has infimum zero on every nonempty path component, and its critical points are precisely the torsion-free structures. In fixed isometric classes, we contrast the concentration phenomena underlying finite-time singularities with cohomological classes admitting positive energy lower bounds and harmonic representatives. Finally, we establish a lifting principle for metric-dependent flows, derive an evolution identity for isometric flows, and extend harmonic-flow estimates to general $H$-structures.

comments
28 pages. v2: Substantially revised. Corrected the SU(m) examples and the moduli-space discussion; strengthened the metric-map theorem; expanded and clarified several proofs and analytical arguments
fields
Differential Geometry
pdf ↗ page arXiv:2606.12031 ↗

Dyer (1991) proved that, for each fixed length $k$, only finitely many isomorphism types of Bruhat intervals occur in finite Coxeter groups. In this short paper, we prove that this result holds also for affine Coxeter groups, and moreover that this characterizes the affine groups among all infinite Coxeter groups. We also show that the coefficients of $q$ in the Kazhdan-Lusztig polynomials of a Coxeter group are bounded if and only if the group is finite or affine.

comments
7 pages; v2: added appendix. Final version to appear in Math. Research Letters
fields
Combinatorics, Representation Theory
pdf ↗ page arXiv:2606.16894 ↗

We propose, analyze and test computational methods for solving a continuous-time heterogenous agent model with Epstein-Zin utility. Such recursive utilities allow the model to disentangle between risk aversion and intertemporal substitution. Having discretized the Hamilton-Jacobi-Bellman (HJB) equation arising in the model, we propose a Howard-Newton algorithm for the late resolution preference case, and a Howard-Tarski-Kantorovich algorithm for the early resolution preference case. We prove the convergence of the iterative algorithms. We obtain as a consequence the existence of solutions to the discretized HJB equations. In the late resolution case, we supply a priori estimates between the unique solutions of the continuous and discretized HJB equations.

fields
Optimization and Control, Numerical Analysis
pdf ↗ page arXiv:2606.23408 ↗

A graph with maximum degree $Δ$ is \emph{$Δ$-critical} if it is connected, has chromatic index $Δ+1$, and the deletion of any edge reduces its chromatic index to $Δ$; it is \emph{nontrivial} if it has no subgraph $H$ with $|E(H)|>Δ\lfloor |V(H)|/2\rfloor$. We prove a characterization of all nontrivial $3$-critical graphs: every such graph is obtained either from smaller $3$-critical graphs by replacing a vertex with a triangle, by a Hajós-join, or by replacing a vertex with a suitable class~$1$ graph, or from a snark by deleting at most one vertex and some independent edges. The third operation, which we call a Meredith-type extension, generalizes the Meredith extension by allowing other class~$1$ graphs in place of $K_{2,3}$, and we give a sufficient condition under which it produces new nontrivial $3$-critical graphs from smaller ones. To the best of our knowledge, this is the first characterization of this kind for edge-chromatic $3$-critical graphs, and the Meredith-type extension has not previously been considered in this generality. The characterization was found by analyzing the output of an exhaustive computer search for nontrivial $3$-critical graphs of orders $13$ to $24$, and we make the resulting data publicly available. Based on these data, we also propose two conjectures that are closely related to the case $Δ=3$ of Seymour's exact conjecture and to Grötzsch's conjecture on the $3$-edge-colorability of planar graphs.

fields
Combinatorics
pdf ↗ page arXiv:2606.27131 ↗

We study standard spherical growth rates of right-angled Coxeter groups through the clique polynomial of the defining graph. We prove that every even degree at least four occurs as the degree of a strongly primitive Salem growth rate: for each $d\geq2$, there are infinitely many connected $K_{2d+1}$-free defining graphs whose full reciprocal-radius polynomial is an irreducible Salem polynomial of degree $2d$. We also establish independence-polynomial obstructions for prescribed Salem polynomials, including a sharp first-coefficient bound $a_1\leq-5$. Finally, an elementary golden-ratio growth gap excludes Lehmer's number from the standard spherical growth rates of right-angled Coxeter groups.

comments
12 pages; corrected errors and revised Sections 6 and 7
fields
Group Theory, Combinatorics
pdf ↗ page arXiv:2606.29397 ↗

Mutual information estimation is a central problem in machine learning and statistics; however, existing benchmarks typically evaluate estimators on simplified, low-dimensional distributions, leaving their performance on complex, realistic data largely unexplored. We address this gap with a comprehensive benchmarking framework grounded in a unified copula-theoretic perspective that subsumes existing benchmarks as special cases. Within this framework, we propose two complementary families of tests: a copula-first family that systematically varies ground-truth mutual information, dimensionality, and marginal complexity using synthetic and flow-based transformations; and a marginals-first family that couples real-world image data with controlled dependency structures, extending the classic same-class-pairing paradigm. We use this suite to extensively evaluate three classes of estimators: non-parametric, variational, and generative. Contrary to prevailing assumptions, our results indicate that there is no universal winner: each category can systematically outperform all other estimators under specific setups. By analyzing these cases, we identify fundamental estimation barriers and propose new tests that more effectively stress these specific limitations. We share the open source code at https://github.com/VanessB/mutinfo.

fields
Information Theory
pdf ↗ page arXiv:2607.03487 ↗

We show that the data of a Riemannian metric on a differentiable stack presented by an orbifold groupoid is equivalent to the data of a Riemannian metric on its diffeological orbit space. As a consequence, we conclude that the classical notion of Riemannian orbifold is equivalent to that of a Riemannian diffeological orbifold. We use the framework for Riemannian diffeology introduced by Kuribayashi, Sakai, and Shiobara, and our result answers a problem they posed in the affirmative. More generally, we show that a Riemannian metric on a Lie groupoid, namely a 2-metric in the sense of del Hoyo and Fernandes, induces a Riemannian metric on its diffeological orbit space only if the Lie groupoid is regular, and that properness is a sufficient but not necessary condition.

comments
27 pages. Added a comparison between two notions of diffeological Riemannian metric. Added a reference to "A baby Baire property" by Watts and Ziegler that simplifies the proof of Proposition 4.18
fields
Differential Geometry
pdf ↗ page arXiv:2607.08939 ↗

In the present paper, with the help of the Faà di Bruno formula and several identities of partial Bell polynomials, the author establishes explicit formulas of the Gauss hypergeometric functions \begin{gather*} {\,}_2F_1\biggl(\frac{1-n}{2},\frac{2-n}{2};\frac{3}{2}-m;z^2\biggr), \quad {\,}_2F_1\biggl(-\frac{n}{2},\frac{1-n}{2};\frac{1}{2}-m;z^2\biggr),\\ {\,}_2F_1\biggl(a,a+\frac{1}{2};\frac{3}{2}-m;z^2\biggr), \quad {\,}_2F_1\biggl(a,a+\frac{1}{2};\frac{1}{2}-m;z^2\biggr) \end{gather*} for $m,n\in\mathbb{N}$ and $a\in\mathbb{C}$, and then derives two combinatorial identities \begin{equation*} \sum_{k=0}^{m}\frac{2^k}{k!} \binom{2m-2k}{m-k} \sum_{\ell=0}^{k} \frac{(-1)^\ell}{2^\ell} \frac{(2k-2\ell-1)!!}{(n-\ell)!} \binom{2k-\ell-1}{\ell-1} =\frac{1}{n!}\binom{2m-n}{m} \end{equation*} and \begin{equation*} \sum_{k=1}^{m}\frac{1}{(k!)^2}\binom{2m-2k}{m-k} \sum_{\ell=1}^{k} \binom{k}{\ell}\ell(2k-\ell-1)! (2a)_\ell =\binom{2m+2a}{m}-\binom{2m}{m}, \end{equation*} where $m\in\mathbb{N}_0$, $n\in\mathbb{Z}$, and $a\in\mathbb{C}$. These newly-established identities extend and generalize the beautiful combinatorial identity \begin{equation*} \sum_{k=0}^{n} \frac{2^{k}}{k!}\binom{2n-2k}{n-k} \sum_{j=0}^{k}\frac{(-1)^{j}}{2^j} \frac{(2k-2j-1)!!}{(n-j)!} \binom{2k-j-1}{j-1} =\frac{1}{n!}, \quad n\in\mathbb{N}_0, \end{equation*} which was obtained in Theorem 4 of the paper "F. Qi, C.-Y. He, and D. Lim, Explicit formulas of two Gauss hypergeometric functions and several combinatorial identities, Discrete Appl. Math. Vol. 393 (2026), 215–229. DOI: https://doi.org/10.1016/j.dam.2026.06.023."

comments
16 pages. Corrected version
fields
Combinatorics, Number Theory
pdf ↗ page arXiv:2607.10643 ↗

We propose a new strategy for gradient-based unconstrained optimization, involving two parallel sequences of iterates that cooperate to determine their stepsizes via a \textit{Twin-Step} principle. Rather than minimizing the objective function individually, the algorithm selects steplengths that minimize the Euclidean distance between the two gradient based processes occurring simultaneously at each iteration. The theoretical analysis shows that the convergence of the mutual distance is governed by the angle between the search directions. In particular the effectiveness of the overall process degrades as the directions approach parallelism. To ensure robustness against collinearity, we introduce a hybrid framework, Twin-ABB$_{\min}$, which switches to the Adaptive Barzilai–Borwein method when the geometric cooperation becomes ineffective. Extensive and very promising numerical results evidence that the Twin phase creates favorable initial conditions for subsequent BB-type iterations.

fields
Numerical Analysis, Optimization and Control
pdf ↗ page arXiv:2607.11575 ↗

We study Heilbronn's triangle problem in the unit right triangle, where $n$ points are placed to maximize the smallest of the $\binom{n}{3}$ triangle areas they span. We prove a boundary-structure result: unless all three vertices are occupied, some optimal configuration with $n \ge 5$ has at least four points on the boundary, one edge containing two of them. With the affine $S_3$ symmetry this fixes four boundary points and $n$ orientation variables in a mixed-integer model that certifies global optimality for all $n \le 8$ up to a relative gap of $10^{-4}$. For $n \le 7$ this independently confirms the known exact optima, the case $n = 7$ having been proved by Zeng and Chen with symbolic computation. For $n = 8$, where only bounds were known, it narrows the gap to Cantrell's configuration fourfold in 39 minutes on one machine. The optimum there is conjectured to be the real root of a septic obtained by Chen, Zeng and Zhou, which our reconstruction confirms to 250 digits. We show its Galois group is $S_7$, so on that conjecture no expression in radicals exists.

comments
Accepted for Operations Research Proceedings 2026 (GOR, Passau). 7 pages, 1 figure, 1 table
fields
Optimization and Control, Combinatorics
pdf ↗ page arXiv:2607.15021 ↗

Matroid configurations -- introduced by Geramita, Harbourne, Migliore, and Nagel -- are projective varieties which generalize so-called \textit{star configurations} and whose defining ideals are obtained by appropriately specializing the Stanley-Reisner ideal of a matroid. Motivated by this connection, we study the asymptotic resurgence of the Stanley-Reisner ideals of matroids. A result of Villareal shows that it is equivalent to study the asymptotic resurgence of facet ideals. We prove a formula for the asymptotic resurgence of the facet ideal of a matroid in terms of the Waldschmidt constant of facet ideals of the contractions of the matroid. As a consequence, we show that asymptotic resurgence respects the weak order on matroids of the same rank. Therefore, the asymptotic resurgence of the facet ideal of a given matroid is bounded above by the asymptotic resurgence of the facet ideal of a so-called \textit{almost-uniform} matroid of the same rank, which we compute explicitly. Guardo, Harbourne, and Van Tuyl showed that the asymptotic resurgence of an ideal is bounded below by the ratio of the initial degree of the ideal by its Waldschmidt constant. We prove that this lower bound is an equality for facet ideals of many classes of matroids, including matroids of rank $k$ on a ground set of size $n\ge 2k$ whose dual is paving, perfect matroid designs, and sparse paving matroids arising from Steiner systems. For the latter two classes, we explicitly compute the asymptotic resurgence.

comments
Updated version includes minor changes: corrected statement/proof of Prop. 5.11, shortened proof of Theorem 7.5, correcting some typos. 40 pages (including figures), comments welcome
fields
Commutative Algebra
pdf ↗ page arXiv:2607.20892 ↗

Quantum cellular automata (QCAs) describe locality-preserving quantum dynamics and link quantum information, many-body physics, and topological quantum field theory (TQFT). Constructing a QCA from a TQFT, however, remains challenging: although a topological action yields a Hamiltonian with commuting terms that realizes the desired ground state, it does not by itself specify an automorphism of the full operator algebra. We develop a unified algebraic construction that extends such Hamiltonians to complete separator-flipper algebras on the full tensor-product Hilbert space, thereby defining the corresponding QCAs microscopically. In 3 spatial dimensions, our framework unifies all previously known QCA constructions associated with the $\mathbb{Z}_8\times\mathbb{Z}_2$ subgroup of the Witt group, including the $U(1)_2$ and $U(1)_4$ QCAs. The same algebraic structure yields new infinite families of generalized $U(1)_2$ and $U(1)_4$ non-Clifford QCAs in spatial dimensions $d=4k-1$. We also reformulate the 4-dimensional $w_2w_3$ QCA and use it to construct QCAs from TQFTs associated with arbitrary products of Stiefel–Whitney classes. For example, we explicitly construct two infinite families: $w_2^nw_3^m$ QCAs in dimension $d=2n+3m-1$ and $w_2w_{4k-1}$ QCAs in dimension $d=4k$. For the 5-dimensional $w_3^2$ and $w_2^3$ QCAs, our construction gives finite-depth quantum circuits, consistent with the cobordism classification. More generally, we show that our construction yields a finite-depth quantum circuit whenever the corresponding Stiefel-Whitney number vanishes on every closed oriented manifold. Overall, these results convert invertible TQFTs into microscopic QCAs, provide higher-dimensional constructions beyond the Clifford setting, and open a systematic route to characterizing their stable structures and boundary anomalies.

comments
57+23 pages; expanded discussion in appendix; typo fixes and clarity improvement throughout
fields
Mathematical Physics, Quantum Algebra
pdf ↗ page arXiv:2607.21697 ↗

We extend Toën's derived Brauer theory to the open site of Spivak's derived smooth manifolds. For a compact derived smooth manifold $X$, we define a spectralized structure sheaf $\mathbb{O}_{X}$ and prove that every Picard twist can be represented by a derived Azumaya algebra, whose Morita equivalence class is uniquely determined by the twist. In the open locally Morita trivial setting, this gives a natural bijection \begin{align*} Φ_{X}:\mathbf{dBr}^{\mathrm{open}}(X) \xrightarrow{\cong} H^{1}(|X|;\underline{\mathbb {Z}}) \times π_0Γ(|X|,\mathrm B^2GL_1(\mathbb O_X)). \end{align*} If we consider $C^\infty_X(\mathbb C)$ on ordinary smooth manifolds, the second factor is $H^3(|X|;\underline{\mathbb Z})$, which recovers the Dixmier-Douady invariant. If we consider $C^\infty_X(\mathbb R)$, the second factor is $H^2(|X|;\underline{\mathbb Z/2})$, which recovers the corresponding invariant of Donovan and Karoubi.

comments
30 pages, no figures. The written exposition of v1, v2 and v3 assisted with generative AI; all mathematical ideas, arguments, proofs, and conclusions were developed by the authors. V4 was prepared without AI assistance, corrects errors in v1, v2 and v3, and constitutes the complete and final version of the manuscript
fields
General Mathematics
pdf ↗ page arXiv:2607.22668 ↗

Quantum resource testing is a fundamental primitive of quantum information processing, profoundly connected to resource manipulation. Its goal is to discriminate $n$ copies of a given resourceful state $ρ$ from all free (i.e., resourceless) states; key instances for applications are entanglement testing and quantum magic testing. The asymptotic characterisation relies on the recently proven Generalised Quantum Stein's Lemma (GQSL), which establishes the rate of decay of the false negative error probability for a fixed false positive error probability. This result, however, is intrinsically asymptotic and thus can provide no finite-resource guarantees, which makes its practical implications unclear. Here, we establish the first rigorous finite-$n$ bounds on quantum resource testing and hence quantum resource manipulation, thus strengthening the GQSL and providing explicit estimates on the number of copies needed to achieve a prescribed performance. As notable consequences, we obtain (a) the convergence of the regularised Rényi relative entropies of a resource, which settles the important open problem from [Fang/Hayashi, arXiv:2508.12901, IEEE ToIT 72:6, 2026]; and (b) the first sample-complexity bound for asymmetric resource testing: for any fixed false positive error probability, a false negative error probability of at most $δ$ can be achieved with $n=O\left(\frac{\log(1/δ)}{D^\infty(ρ\|\mathcal F)}\right)$ copies of $ρ$, in the limit where $δ\to 0$.

comments
10+27 pages
fields
Information Theory, Mathematical Physics
pdf ↗ page arXiv:2607.24712 ↗

Let $Γ$ be a Coxeter graph, and let $S$ be its vertex set. An elemental complex over $Γ$ is an abstract simplicial complex $\mathcal D$ on $S$ that contains every pair $\{s, t\}$ with $s \neq t$ and $m_{s,t} \neq \infty$. To each elemental complex $\mathcal D$, we associate a cube complex $Ξ_{\mathcal D}$ on which the virtual Artin group ${\rm VA} [Γ]$ acts. We prove that $Ξ_{\mathcal D}$ is connected, simply connected, and locally regular; that the action of ${\rm VA}[Γ]$ on $Ξ_{\mathcal D}$ is regular, cocompact, and by isometries; and that the stabilizers of cells are parabolic subgroups. We finally prove that $Ξ_{\mathcal D}$ is CAT(0) if and only if ${\mathcal D}$ is a flag complex.

fields
Group Theory
pdf ↗ page arXiv:2607.24840 ↗

We show that the space of Bridgeland stability conditions Stab(Y) on a very general quintic threefold Y has infinitely many connected components. Using similar methods, we also prove that for a particular smooth projective (weak Fano) threefold X the braid group action on the set of full exceptional collections in Dscr^b(X) has infinitely many orbits, in particular, it is not transitive. This provides a counterexample to a conjecture of Bondal and Polishchuk from 1993. The conjecture was first disproved by Chang, Haiden, and Schroll, who constructed a family of partially wrapped Fukaya categories for which the transitivity fails. However, no counterexample of the form Dscr^b(X) for X a smooth projective variety was previously known. In addition, we show that Stab(X) also has infinitely many connected components. These are the first known examples of a smooth projective varieties whose spaces of Bridgeland stability conditions are disconnected.

comments
The preprint has been reorganised and slightly extended, most significantly to include disconnectedness of Stab for very general quintic threefolds, as well as an example of disconnected Stab/Auteq
fields
Algebraic Geometry
pdf ↗ page arXiv:2607.25391 ↗

In this paper, we build on the analysis initiated by Averkov and Schymura (2022) and prove that the shifted Heller constant for matrices with $m$ rows is at most $m^2 + m$ for all sufficiently large $m$. Consequently, the number of distinct columns of a $Δ$-modular matrix $A \in Z^{m \times n}$ of rank $m$ is at most $O(m^2 Δ)$. This improves upon the earlier bounds of $O(m^4 Δ)$ and $O(m^2 Δ^2)$.

fields
Combinatorics, Optimization and Control
pdf ↗ page arXiv:2608.06215 ↗

We investigate positive weak solutions of the critical $p$-Laplace equation $$ -Δ_p u = u^{p^*-1} h(u), \qquad 1 < p < n, $$ where $h$ is a positive, bounded, continuous, and nonincreasing function. Our first main result is a complete classification of normalized, bounded, positive entire solutions for every equation in a compact family determined by $h$: Any such solution must coincide with an Aubin–Talenti profile. Moreover, the existence of an Aubin–Talenti profile as a solution implies that $h$ is constant on the entire interval of values attained by that profile. Subsequently, applying this classification result, we establish the following scale-invariant Schoen-type estimate $$ \left(\sup_{B_R} u\right)\left(\inf_{B_{2R}} u\right)^{p-1} \le C R^{p-n} $$ for nonnegative weak solutions defined in $B_{3R}$. As a direct corollary, we apply the profound result of Ciraolo–Gatti \cite{CG2026} to obtain a fully unrestricted Liouville theorem: Every positive entire solution must coincide with an Aubin–Talenti profile. For the purely critical equation, we also show that the corresponding Liouville classification is equivalent to a Schoen-type Harnack inequality. The arguments in this work give quasilinear versions of the Kelvin transform and the method of moving spheres, which was previously available only in the semilinear setting, and also yield alternative proofs of the classical results. The technique developed here can likely be extended to a wider class of Liouville-type problems for critical equations.

comments
66 Pages. See Appendix B for a concise outline for the pure-power case with p=2. In the abstract, we now explicitly emphasize the important role of Ciraolo--Gatti in the classification result!
fields
Analysis of PDEs, Differential Geometry
pdf ↗ page arXiv:2608.09113 ↗

In analogy with the notion of Krull dimension, for every commutative ring $R$, we introduce the notion of pure dimension $\pdim(R)$ by considering chains of purely-prime ideals. Then we see that the inequality $\pdim(R)\leqslant\dim(R)$ holds for a very broad class of rings $R$, including pseudo-Noetherian rings (e.g., Noetherian rings), Gelfand rings (e.g., local rings), mp-rings (e.g., integral domains), and rings with finitely many maximal ideals or finitely many minimal primes. In fact, for all of these rings $R$, we always have $\pdim(R)=0$, i.e., every purely-prime ideal of $R$ is purely-maximal. \\ However, we prove a classification theorem which states for each ordinal number $Ω$ the proper pure ideals of the commutative ring $$R_Ω=k[X_α : α<Ω]/\big(X_α(1-X_β) : α<β<Ω\big)$$ are exactly of the form $I_λ=(x_α: α<λ)$ with $λ=0$ or $λ\leqslantΩ$ a limit ordinal (we write $x_α$ for the image of $X_α$ in $R_Ω$). In addition, all of them are purely-prime ideals and form the following unique strictly ascending chain: $$I_{0}=0\subset I_ω\subset I_{ω\cdot2}\subset I_{ω\cdot 3}\subset\ldots\subset I_{ω\cdotα}$$ where $α$ is a unique ordinal with $Ω=ω\cdotα+n$ and $n\inω$. To prove this theorem, we use the full strength of ...

comments
24 pages
fields
Commutative Algebra, Rings and Algebras
pdf ↗ page arXiv:2608.14251 ↗

We study stable equivalences of Brauer graph algebras. We prove that a basic symmetric algebra without semisimple summands which is stably equivalent to a Brauer graph algebra is again a Brauer graph algebra. In particular, Brauer graph algebras are closed up to semisimple summands under stable equivalence of Morita type. As consequences, we reprove the closure under derived equivalence established by Antipov and Zvonareva and obtain a solution of the reconstruction problem posed by Rickard and Rouquier for algebras stably equivalent of Morita type to Brauer graph algebras.

comments
33 pages. All comments are welcome!
fields
Representation Theory, Rings and Algebras
pdf ↗ page arXiv:2608.14253 ↗

In this paper we extend Hitchin-Kobayashi correspondence along holomorphic fibre space and we show that there exists a generalized Hermitian-Einstein metric which is twisted with Weil-Petersson current.In fact we generalise Song-Tian program for vector bundles

fields
Differential Geometry
pdf ↗ page arXiv:2608.15206 ↗

We establish a new local-to-global rigidity phenomenon for the Bergman metric. Namely, under natural geometric hypotheses, a local conformal identification of Bergman metrics determines the underlying complex manifold globally, up to the unavoidable ambiguity of removing Bergman-negligible subsets. More precisely, let $Ω\subseteq\mathbb{C}^n$ be a bounded domain with a complete Bergman metric, and suppose that the Bergman metric of a complex manifold $M$ is locally conformal, via a holomorphic map $f$, to that of $Ω$, with a constant conformal factor when $n=1$. We prove that the given local map $f$ extends to a biholomorphism $F\colon M\to D$ onto a subdomain $D\subseteqΩ$ in two complementary settings. If $M$ is Stein, then $Ω\setminus D$ is a closed pluripolar set. If $M$ is a bounded domain and $Ω$ satisfies a natural symmetry condition expressed in terms of its automorphism orbits, then $Ω\setminus D$ is Bergman-negligible. In particular, this applies when $Ω$ is a bounded homogeneous domain and yields a characterization, up to Bergman-negligible sets, of a class of bounded domains with locally symmetric Bergman metrics, addressing a question raised by Loi–Palmieri and Zimmer. A key ingredient in the proofs is a new Calabi-type extension theorem tailored to Bergman metrics.

comments
This is an updated version of a previous preprint with the same title (originally uploaded August 2026). The main change is a new section added (Section 7)
fields
Complex Variables
pdf ↗ page arXiv:2608.15998 ↗

Equation-free forecast-error analysis is useful when governing equations, Jacobians, tangent dynamics, or full state vectors are unavailable and only scalar time series are observed. In this setting, the multi-horizon forecast-error curve can provide both an instability estimate and nonlinear predictive features, making its numerical reliability important. We show that long finite-precision chaotic records can enter an information-saturation regime in which increasing record length N (the total number of analyzed samples) adds repeated histories rather than new dynamical information. For the float32 logistic map, exact recurrence progressively fills the training catalogue with duplicate histories and identical futures, suppressing forecast-error growth and driving the estimated Lyapunov slope toward zero. Across 10,000 initial conditions, exact recurrence occurred by iteration 7612; across 28 stratified seeds, collapse onset strongly followed the recurrence-occupancy scale. A two-dimensional Hénon-map control confirms that recurrence and estimator failure are separate events: the float32 trajectory enters an exact machine-state cycle after a transient of $μ= 103,039$ samples, with cycle period P = 12,268, yet the forecast-error estimate remains close to the positive Jacobian/QR reference until catalogue saturation near N = 200,000. Casting an already generated float32 record to float64 does not restore lost information, whereas genuine float64 generation, restarted short trajectories, and one-ULP perturbations prevent collapse over the tested range. A recursive local-linear predictor exhibits the same endpoint at a later, predictor-dependent threshold. The results identify a recurrence-to-catalogue-saturation failure mode specific to forecast-error/analogue inference rather than Lyapunov estimation in general.

comments
34 pages, 7 figures, 3 tables. Revised and expanded version. Published in Chaos, Solitons & Fractals
journal
Chaos, Solitons & Fractals 213 (2026) 119363
fields
Dynamical Systems
pdf ↗ page arXiv:2608.16120 ↗

The likelihood of a finite simple undirected graph $G$ on $n$ vertices is the probability that the uniform sequential attachment process, which at each step joins a new vertex to a uniformly random subset of uniformly random size of the vertices already present, outputs a graph isomorphic to $G$. Dervovic, Mocherla and Severini conjectured that the likelihood is minimised by the balanced complete bipartite graph. We prove that, among complete bipartite graphs of a given order, the balanced one uniquely minimises the likelihood. Exact computation shows that it also minimises over all graphs for every order from $6$ through $14$. At $n=15$, the exact global minimisers are $C_5[\bar K_3]$ and its complement, giving the first counterexample. Completed exact searches show that the global minimisers at order $16$ are $K_{8,8}$ and its complement. At order $17$, the unique minimiser is a self-complementary mixed blow-up of $C_5$. At order $18$, the minimisers are $K_{9,9}$ and its complement. We show that the failure is not sporadic by proving that the likelihood of the balanced complete bipartite graph is $2^{-(1/2-1/(8\ln 2)+o(1))n^2}$, whereas the minimum over all graphs of order $n$ is $2^{-(1/2+o(1))n^2}$, so the conjectured minimiser exceeds the minimum by a factor exponential in $n^2$. We also determine the Shannon entropy of the process to leading order, namely $n^2/(4\ln 2)$ bits, which shows that the conjectured minimiser is in fact more likely than a typical output of the process. The proofs rest on a vertex deletion recurrence which evaluates the likelihood in time $O(2^n n)$ and which closes on the blow-ups of any fixed base graph.

comments
22 pages. Certified global minima and all minimizers through n=18; unique self-complementary minimizer at n=17. Improved exact pruning and measured search timings
fields
Combinatorics
pdf ↗ page arXiv:2608.19467 ↗

Let $0\le μ_1\le μ_2 \le \cdots \le μ_n$ and $δ> 0$. Samuels' conjecture states that if $X_1,\dots,X_n$ are independent non-negative random variables with $\mathbb{E}[X_i] = μ_i$, then $$ \mathbb{P}\left( \sum_{i=1}^n X_i < δ+ \sum_{i=1}^n μ_i \right) \ge \min_{1\le i\le n} \prod_{j=i}^n \left(1-\frac{μ_j}{δ+ \sum_{k=i}^n μ_k}\right).$$ This conjecture was recently proved by Ling. In this note, we prove a refinement of Samuels' inequality via an operation called the \emph{cap move}. Our refined inequality is sharp, with equality attained for a strictly larger class of distributions than in the original inequality.

fields
Probability, Combinatorics
pdf ↗ page arXiv:2608.22816 ↗

At an $O(2)$-equivariant Hopf bifurcation a spatially periodic system selects between traveling waves and standing waves, and which branch appears and whether it is stable is decided by two cubic normal form coefficients. Obtaining these coefficients for a given PDE has required a derivation carried out afresh for each model. We remove that step for two-component systems with conservative polynomial differential nonlinearities on a one-dimensional periodic domain, deriving closed form formulas for both coefficients. They are expressed directly in terms of the array of nonlinear PDE coefficients and the spectral data of the linearization, its critical eigenvectors, adjoint eigenvectors, and non-critical resolvents. We give verifiable conditions under which the underlying center manifold reduction is valid. A structural condition on the principal part of the linearization, together with sign conditions on the trace and determinant of its Fourier symbol, yields the resolvent estimate and the spectral gap that the reduction requires. A companion implementation evaluates the two normal form coefficients from the coefficient array and verifies the assumptions for a given system. We specialize the formulas to two applications: a strain-gradient regularization of the nonlinear $p$-system, whose quadratic case contains a previously studied model as a special case, and a first order conservative bilinear family, not previously analyzed, in which every bifurcation scenario allowed by the general classification is realized.

comments
26 pages, 4 figures, 1 table. Revised version: proofs and references updated, shortened. Submitted to Zeitschrift für angewandte Mathematik und Physik (ZAMP)
fields
Dynamical Systems, Analysis of PDEs
pdf ↗ page arXiv:2608.24376 ↗

Geometry of quantum states is a fundamental research area with applications ranging from topological materials in condensed matter to variational algorithms in quantum information. Due to their relative simplicity, pure states are often studied, while mixed states are generally needed, e.g., to have nonzero temperature. Mixed-state geometry, however, is much more challenging and has so far lacked any structure relating distance and curvature, like the Kähler structure does for pure states. This work introduces that structure. We show that it has the physical meaning of fluctuation-dissipation—by identifying distance with fluctuations and curvature with linear response—and that it becomes the known structure for pure states, but is non-Kähler otherwise. This paints a simple picture that unifies fluctuation-dissipation and geometry of quantum states. It also applies on parameter spaces, as encountered in condensed matter or estimation theory. Our picture implies that the geometry contains all linear responses and fluctuations, proves the fluctuation-dissipation relation, explains that quantum geometry appears in any sum rule, and allows one to geometrically characterize generic transport behaviors. Closed and open systems are treated with equal ease.

comments
6 pages + Methods, RevTeX, 2 figures; updated version with improved presentation; added open dynamics
fields
Mathematical Physics
pdf ↗ page arXiv:2608.25006 ↗

In this work, we study guided modes in systems of high-contrast resonators near nonzero interior Neumann frequencies, beyond the subwavelength regime. In the regular exterior regime, where the exterior Dirichlet problem is well-posed at the reference wavenumber, we introduce an infinite-dimensional frequency-dependent capacitance operator obtained by compressing the exterior Helmholtz Dirichlet-to-Neumann map to the traces of the interior resonant Neumann eigenspaces. We prove the norm-resolvent convergence of the continuous problem to this discrete effective operator as the contrast $δ\to0$, and derive first-order asymptotic formulas for compact-defect frequencies and line-defect band functions. We then establish exponential off-diagonal decay of the capacitance coefficients by a Combes–Thomas argument, yielding an exponentially accurate truncation of the discrete operator, and show that its retained coefficients can be computed from local Helmholtz problems. At the physical frequency, this local approximation converges exponentially under a uniform stability assumption for the growing finite-cluster problems. The stability assumption can be removed by introducing a vanishing complex absorption together with a Hermitian symmetrization. In particular, an absorption parameter of order $\sqrtδ$, together with interaction truncation and patch radii of order $|\logδ|$, suffices to preserve the $O(δ^2)$ accuracy of the first-order high-contrast expansion of the defect eigenfrequencies, yielding a fast computational method. Numerical experiments for dipole and quadrupole resonances illustrate the accuracy, exponential locality, and applicability of the discrete model to straight and bent waveguides generated by material or geometric detuning.

comments
41 pages
fields
Numerical Analysis
pdf ↗ page arXiv:2608.26906 ↗

For a finite graph, its energy is the sum of the absolute values of its adjacency eigenvalues. Let n be the product of powers of distinct odd primes, where all exponents are positive integers, the sum of the exponents is odd, and every prime whose exponent is odd is at least five. Among all nonempty sets of proper divisors of n, the energy of the corresponding integral circulant graph is uniquely maximised by the checkerboard divisor set consisting of those divisors for which the sum of the prime-power exponents is even. The maximum energy is one half of the sum of n and the product, over all prime factors, of explicit polynomial quantities. For exponent k and parameter x, the corresponding polynomial is given by two k plus one times x to the k-th power, together with four times the sum from zero to k minus one of j plus one times x to the j-th power with alternating sign determined by the parity of k minus j. For the proof, we use weighted prime-power Ramanujan transforms, which give an exact energy model in tensor form and extend to a polynomial matrix family for real parameters greater than one. A semidefinite parity theorem shows that the matrix obtained from this polynomial family by an explicit positive diagonal normalisation is a real symmetric contraction whose unique eigenvalue of modulus one is the simple eigenvalue determined by the parity of the exponent; this holds for even exponents at least two and parameters at least three, and for odd exponents at least one and parameters at least five. Tensorising this spectral structure and centring the binary divisor tensor yield a sharp one-norm bound for the corresponding tensor operator on sign vectors. In particular, for n equal to p squared times q cubed, with distinct odd primes p and q and with q at least five, this proves Roldan's conjectured maximality in this range.

comments
25 pages,0 figures. The main theorem has been extended from two prime factors to an arbitrary finite number of distinct odd prime factors, under the stated assumptions. The title and proof have been revised accordingly
fields
Combinatorics
pdf ↗ page arXiv:2608.29523 ↗

We study an infinite family of strongly coupled four-dimensional $\mathcal N=2$ superconformal field theories (SCFTs) distinguished by a trivial Higgs branch, known as the $(A_2,D_{3m+1})$ Argyre-Douglas theories. We propose that their associated vertex operator algebras (VOAs) are the doublet algebras $\mathcal A(4m+2)$, an infinite family of non-rational vertex operator superalgebras with a remarkably simple strong generating set of only three fields. We provide several highly nontrivial checks of this proposal. In particular, we reproduce the four-dimensional conformal anomalies $a$ and $c$ from the VOA and analytically prove the exact equality between the Schur index of the SCFT and the supercharacter of the VOA. A key ingredient is a diagonal-gauging realization of the $(A_2,D_{3m+1})$ theories in terms of two simpler building blocks, which makes the Schur-index computation tractable. Remarkably, we find that the resulting Schur index of the $(A_2,D_{3m+1})$ theory coincides with that of $\mathcal N=4$ $SU(2)$ Super-Yang-Mills theory, up to an overall prefactor and an appropriate identification of fugacities. We also discuss a generalization to the two-parameter family $(A_{2s},D_{(2s+1)m+1})$, whose members likewise have trivial Higgs branches and admit diagonal-gauging realizations. A particularly interesting subfamily is $(A_{2s},D_{2s+2})$, for which the four-dimensional conformal anomalies coincide, $a=c$. As a byproduct, we also show that the doublet VOA $\mathcal A(4m+2)$ is not classically free for positive integer $m $, and we discuss its implications for the Schur and Macdonald indices. Our results reveal a systematic connection between Higgsless SCFTs, diagonal gauging, and strongly finite but non-rational vertex operator algebras.

fields
Mathematical Physics
pdf ↗ page arXiv:2609.02437 ↗

We isolate and analyze the geometric PDE governing the evolution of the vorticity direction in the 3D incompressible (unforced) Navier-Stokes equations (NSE), restricted to the case of a critical spatial point singularity where the vorticity magnitude concentrates as $O(|x|^{-2})$, inhabiting the critical Lorentz space $L^{3/2, \infty}$. The PDE consists of the Harmonic Map Heat Flow (HMHF) into the sphere supplemented with the fluid transport, cross-diffusion and tangential strain. The question is whether the NSE mechanics can propagate logarithmic decay of the local mean oscillations of the direction -- the condition $ξ\in \mathrm{bmo}_{1/|\log r|}$ -- which (in this setting) was shown in the companion paper to prevent finite time blow-up. A key observation is that the $O(|x|^{-2})$ concentration, factored out of the viscous cross-diffusion, generates an outward radial drift $4ν\, x/|x|^2$ at the core. Under a suitable condition we prove a barrier lemma for the total drift-diffusion operator and show that the HMHF nonlinearity is harmless for the quantity $\frac12|ξ- e|^2$ which is a subsolution on any hemisphere. This yields a transfer theorem: the $\mathrm{bmo}_{1/|\log r|}$ regularity of the direction at the core, uniformly up to the singular time, is controlled by a logarithmic modulus in time of the direction at the inner scale, together with the physical strain. Crucially, the strain enters the direction equation only through its tangential component which vanishes precisely when the direction is an eigenvector of the strain tensor. Since this includes the eigenvector carrying the maximal stretching, the result is in contrast to the classical geometric regularity theory which is built on depleting the full vortex-stretching term and thus confined to configurations of weak stretching.

comments
23 pp; definition of the critical point singularity is generalized by dropping the lower bound on the shape factor, mostly since divergence-free vector fields living between two critical bounds are likely to get stuck in bmo
fields
Analysis of PDEs
pdf ↗ page arXiv:2609.05720 ↗

We establish local-in-time existence and uniqueness of monotone classical solutions to the two-dimensional nonstationary boundary layer equations for dilatant power-law fluids in a periodic half-space, for every fixed exponent $n>1$. The result holds under Oleinik's monotonicity condition and suitable weighted Sobolev, algebraic decay, and compatibility assumptions. The main difficulties are the tangential derivative loss caused by the normal velocity and the degeneracy at infinity of the principal coefficient $nω^{n-1}$ in the nonlinear vorticity diffusion $\partial_y^2(ω^n)$, where $ω=\partial_y u$. We adapt the Masmoudi-Wong energy method and use a good unknown to control the tangential derivative loss. For the nonlinear diffusion, we separate the estimates near the boundary from the treatment of degeneracy in the far field. Nonlinear boundary reduction and localized trace estimates control the boundary terms, while weighted estimates exploit the compensating decay of vorticity derivatives to control the diffusion commutators. Together with tangential regularization and weighted maximum and minimum principles, these estimates yield bounds uniform in the regularization parameter and preserve weighted monotonicity. Under these assumptions, the result provides an affirmative local-in-time answer to the existence and uniqueness question in Open Problem~11 of Oleinik and Samokhin.

fields
Analysis of PDEs
pdf ↗ page arXiv:2609.06944 ↗

For any \(n\geq 1\), let \((M,θ)\) be a compact strictly pseudoconvex hypersurface type CR manifold with dimension \(2n+1\), \(Q=2n+2\) be its homogeneous dimension. We prove the sharp Folland–Stein Sobolev inequality on \(M\): if \(Q^*=2Q/(Q-2)\), then there exists a constant \(A>0\), depending only on \((M,θ)\), such that \[ \|u\|_{L^{Q^*}(M)}^2 \le K(n)^2\|\nabla_b u\|_{L^2(M)}^2 + A\|u\|_{L^2(M)}^2, \qquad u\in S^{1,2}(M). \] Here \(K(n)\) is the sharp constant in the Folland–Stein Sobolev inequality on the Heisenberg group \(\mathbb H^n\).

fields
Analysis of PDEs
pdf ↗ page arXiv:2609.09841 ↗

Distributed multiple testing asks $N$ sites, each holding p-values for its own hypotheses, to control the false discovery rate (FDR) of the discoveries made across the whole network while communicating only a small number of bits. Greedy interval aggregation (Pournaderi and Xiang, IEEE TSIPN, 2023) meets the communication budget but controls FDR only asymptotically, and its FDR can exceed the target at finite sample sizes; the same interval counts both select the rejection regions and calibrate the stopping rule, which biases the selected interval densities upward. We propose \emph{budgeted BONuS-GA}: each site mixes synthetic uniform p-values into its data, the center ranks candidate p-value intervals across sites using the pooled counts, and each site's false discoveries are estimated from its own synthetic counts under a per-site share of the level. The procedure needs no knowledge of the null proportions, uses every p-value for both selection and inference, and controls $FDR\leα$ at every sample size for any fixed assignment of hypotheses to sites, assuming only that the null p-values are independent uniforms, independent of the non-null p-values. It keeps the original $O(\sqrt m\log m)$ communication, $m$ being the total number of p-values, when $N=O(\sqrt m)$. We also give a sample-splitting alternative, cross-fit greedy aggregation, with finite-sample control under a random-site model, and we quantify the selection bias behind the original procedure's failure. In a monitoring-network simulation the budgeted procedure retains most of the power of its heuristic counterpart at moderate-to-large site sizes.

fields
Information Theory
pdf ↗ page arXiv:2609.14170 ↗

Let $\mathscr L$ denote the set of Liouville numbers. We prove a local arithmetic rigidity theorem for real-analytic functions. If $U\subset\mathbb R$ is an open interval and $f:U\to\mathbb R$ is real-analytic with $f(\mathscr L\cap U)\subseteq\mathscr L$, then $f$ is the restriction to $U$ of a rational function in $\mathbb R(x)$. Quantitatively, there exists an absolute constant $τ>2$ such that, for every nonrational real-analytic function $f$ and every nonempty open subinterval $V\subset U$, the set of $ξ\in V\cap\mathscr L$ satisfying $μ(f(ξ))\leqτ$ contains a Cantor set. As a consequence, every entire function $F:\mathbb C\to\mathbb C$ satisfying $F(\mathscr L)\subseteq \mathscr L$ is a polynomial with real coefficients. This gives a negative answer to a problem posed by Mahler in 1984. The proof develops a two-height counting estimate for rational approximation to analytic graphs, treating source and target denominators separately, and combines it with a nested construction producing Liouville inputs whose images have uniformly bounded irrationality exponent.

comments
40 pages. Retitled as Part I of a two-part work. Includes a Lean 4 formalization
fields
Number Theory
pdf ↗ page arXiv:2609.14202 ↗

We extend Behrend's theory of complementary polyhedra to include Jordan-Hölder theory by introducing the notion of Jordan-Hölder facets. As a consequence, we can construct Jordan-Hölder parabolic subgroup schemes for reductive group schemes over a smooth projective curve. We use these results to provide a uniform approach to describing Jordan-Hölder filtrations and S-equivalence of moduli stacks of semistable bundles, including vector bundles, principal bundles, parabolic bundles, parahoric torsors, and parahoric Higgs torsors, following Zhang. We explicitly give a description of Jordan-Hölder filtrations and S-equivalence in the semistable locus of parahoric torsors and parahoric Higgs torsors, coming from their good moduli spaces, as constructed by Alper-Halpern-Leistner-Heinloth and Réga. This application uses the Rees construction, as given by Heinloth, to identify filtrations of moduli stacks of parahoric torsors with reductions of parahoric torsors to parabolic subgroups, inducing Jordan-Hölder stratifications of the moduli stacks.

comments
Improvements to Sections 2 and 3, 35 pages, 5 figures. Comments welcome
fields
Algebraic Geometry
pdf ↗ page arXiv:2609.17231 ↗

This article presents proofs of seven conjectural series identities for $π$ proposed by Z.-W. Sun. Our proofs rely mainly on transformations involving the Beta function. We also use symbolic integration with the software \texttt{Mathematica} to reduce the amount of computational work.

comments
12 pages
fields
General Mathematics
pdf ↗ page arXiv:2609.22304 ↗

This article is the second part of a review devoted to spacetime singularities in classical General Relativity. While Primer I developed the mathematical framework for defining, classifying, and proving the existence of singularities, the present work examines their concrete realization in spatially homogeneous cosmological models. After reviewing the geometrical foundations of homogeneity and the Bianchi classification, we derive the field equations for orthogonal and tilted perfect fluid cosmologies and use the Kasner and Bianchi I solutions as elementary prototypes of anisotropic singular behaviour. Raychaudhuri focusing provides general conditions under which the homogeneous Cauchy development reaches a finite proper time boundary. We then distinguish matter singularities, characterized by divergent density and Ricci curvature, from more subtle non-scalar or whimper singularities associated with Cauchy horizons and extreme fluid tilt. The classical Ellis-King analysis is reconsidered in the light of later developments, including rigorous curvature-blow-up results for Bianchi VIII and IX models, kinematic singularities in geodesically complete tilted spacetimes, and the recent reappearance of whimper behaviour in dipole cosmology. The resulting picture shows that even spatially homogeneous cosmologies exhibit a remarkably rich hierarchy of singular boundaries. The detailed dynamical approach to these boundaries - Hamiltonian cosmology, Kasner transitions, Mixmaster dynamics and the BKL conjecture - is reserved for Primer III.

comments
43 pages, 4 figures, Appendix
fields
Mathematical Physics
pdf ↗ page arXiv:2609.25112 ↗

We obtain the relation between non-hermite skew information $NI_{ρ,α}(A)$, $NU_{ρ,α}(A)$, variance $NV_ρ(A)$ and $NJ_{ρ,α}(A)$ by using the maximal eigenvalue and the minimal eigenvalue of density $ρ$. And also we obtain their refinements.

fields
Functional Analysis
pdf ↗ page arXiv:2609.27426 ↗

We study affine reflection transforms of convex polygons with parallel opposite sides. An opposite-side involution sends a pencil through an interior point to a piecewise smooth family of lines. Its completed envelope consists of conic arcs and transition segments. The completed transform has non-positive oriented area. Maximising the negative oriented area and dividing by the polygonal area gives an affine invariant that vanishes precisely for centrally symmetric polygons. Combining an identity for reflected chord vectors with the Rogers–Shephard inequality shows that this invariant is strictly less than $1$, with optimal universal constant $1$. The energy extends continuously to the boundary and vanishes there. We also compare chord counts under compatible smoothing and prove Hausdorff convergence to the smooth transform.

comments
24 pages, 5 figures; Version 1 has been split into two papers. This version (v2) treats the polygonal case
fields
Differential Geometry, Metric Geometry
pdf ↗ page arXiv:2609.27931 ↗

We prove nearly quadratic lower bounds for randomized algorithms for linear optimization and uniform sampling over convex bodies in the membership oracle model. For linear optimization, this matches the known nearly quadratic upper bound up to a polylog factor in the dimension. For uniform sampling, this improves on the previous linear lower bound. Our construction also implies the same lower bound for volume estimation.

comments
toc, figures, fewer logs
fields
Functional Analysis, Optimization and Control
pdf ↗ page arXiv:2609.30215 ↗

In the planar $n$-body problem, Moeckel and Montgomery \cite{MM25} showed that there is no infinite spin for total collision solutions, when the reduced and normalized configuration converges to an isolated central configuration. Following their approach, generalizations have been obtained for partial collision solutions and parabolic solutions in the planar case, and for total collision solutions in the spatial cases by various authors, all under the same assumption that the reduced and normalized configuration converges to an isolated central configuration. Here we prove there is no infinite spin for any collision or parabolic solution in all dimensions without such an assumption.

comments
13 pages with minor changes and typos fixed. Suggestions or comments are welcome!
fields
Dynamical Systems
pdf ↗ page arXiv:2609.33250 ↗

Disruptions can leave logistics networks short of capacity, yet outside capacity is hard to use: incumbent models are private, provider records are incomplete, and offers stay hidden until a costly request for quotation (RFQ). We introduce open capacity pooling as a budgeted provider-discovery problem that links which capacity need to pursue, whom to contact, and which verified offer to admit. The alternating direction method of multipliers (ADMM) coordinates incumbents without sharing models, and the resulting capacity gaps direct discovery. A machine-learning layer fitted before search ranks providers by predicted capability and a fleet-size proxy from directory records and large language model (LLM) readings of self-descriptions. Verification reveals offers, reduced-cost screening admits them, and re-optimization updates the gaps and prices. Under fixed offers and exact coordination, a full-revelation reference reaches the finite-universe optimum, while budgeted discovery remains heuristic. We evaluate the mechanism on 4,509 real firms in three US regions excluded from predictor training, using scenarios calibrated to seven historical disruptions and simulated availability, offered quantities, and prices. With 80 RFQs, about 1% of exhaustive search, the mechanism recovers 19-43% of the full-information value of outside transport, against 2-6% for random search, and saves USD 73,000-104,000 per event before search and contracting costs. At tight RFQ budgets, gap-guided need selection and learned ranking are complements in these testbeds. LLM reading helps rank capabilities that records omit, such as refrigerated service, but its operating value varies by region. The results support combining record-based learning with coordination-derived gaps under limited search budgets.

comments
52 pages, 5 figures. v2: substantially revised and retitled. Adds real provider records from three US regions, a learned provider-ranking layer trained outside these regions, and scenarios calibrated to seven historical disruptions
fields
Optimization and Control
pdf ↗ page arXiv:2609.40296 ↗

The Hilbert matrix operator $H$ is bounded on the standard weighted Bergman spaces $A^p_α,\, p>α+2>1$. We examine the limiting case when $α=p-2$, and for $p\leq 2$, we characterize the logarithmic modification of the standard weight needed for the domain $A^p_{v_{γ,p}}$ and codomain $A^p_{v_{β,p}}$ to ensure boundedness. In particular, we prove that $H:A^2_{\log^γ} \to A^2_{\log^β}$ is bounded if and only if $γ-β\geq 2$ and $γ>1$, solving an open problem posed by B. Karapetrović and M. Jevtić. We also settle the situation for $p>2$ in all cases, except when the logarithmic exponents $γ$ and $β$ satisfy $γ-β=p$. Moreover, we completely solve the same problem when the domain and codomain are certain weighted reflexive $\ell^p$ spaces. All results are extended to the generalized Hilbert operator $H_g$.

comments
A generalization of the results presented in v1
fields
Functional Analysis, Complex Variables
pdf ↗ page arXiv:2610.00252 ↗

Given a prime $p$, the Nottingham group $G(p)$ is the group of formal power series $x+a_2x^2+a_3x^3+\cdots$, $a_i\in{\bf Z}/p{\bf Z}$, under substitution. For $n\geq 1$, we write $G_n(p)$ for the quotient of $G(p)$ by its normal subgroup $K_n=\{x+a_{n+1}x^{n+1}+a_{n+2}x^{n+2}+\cdots\,|\, a_i\in {\bf Z}/p{\bf Z}\}$, namely the group of truncated polynomials $x+a_2x^2+\cdots+a_nx^n$, $a_i\in {\bf Z}/p{\bf Z}$, under substitution, which is a $p$-group of order $p^{n-1}$ generated by $x+x^2,\dots,x+x^n$. In this paper, we determine the power-commutator presentation of $G_{15}(p)$ relative to $x+x^2,\dots,x+x^{15}$. This, in turn, automatically gives the power-commutator presentation of $G_{n}(p)$ relative to $x+x^2,\dots,x+x^{n}$ for any $1\leq n<15$, as well as the initial segments of the power and commutator relations in $G_{n}(p)$ corresponding to $x+x^2,\dots,x+x^{15}$ for any $n>15$.

fields
Group Theory
pdf ↗ page arXiv:2610.01469 ↗

Let $X$ be a smooth proper integral variety of dimension at most four over a field of characteristic zero. We construct a finite sequence of blowups along smooth centers with projective endpoint, preserving the intersection of all maximal quasi-projective opens of $X$. The construction depends only on a fixed functorial principalization procedure and is natural under isomorphisms, including those over different ground fields. The proof has two parts. A colength argument on regular local surfaces gives, in every dimension, a finite preparation after which some maximal quasi-projective open has complement mapping into codimension at least three in $X$. In dimensions at most four these images are points or curves. Ample numerical classes on the normalized components of that complement give finite bounds for summing base ideals of linear systems. The resulting ideal descends to the original field; geometric integrality is not required.

comments
10 pages. Added attribution to Wlodarczyk's 1999 finiteness theorem; mathematical results unchanged
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.05427 ↗

Let $G=(G_1,...,G_n)$ be a graph system on a common $n$-vertex set $V$. For $v\in V$, set \[ δ(v)=\min_{i\in[n]}d_{G_i}(v), \qquad E_{\cap}=\bigcap_{i=1}^n E(G_i), \] and let $d^c(u,v)$ denote the length of a shortest rainbow $u$--$v$ path. We study Hamiltonicity conditions imposed only on pairs at small rainbow distance. Suppose that every $G_i$ is $2$-connected and that \[ δ(u),δ(v)<\frac n2, \qquad d^c(u,v)\le2 \quad\Longrightarrow\quad uv\in E_{\cap}. \] Our main result shows that any system satisfying this local Fan-type condition but containing no rainbow Hamilton cycle must satisfy \[ E_{\cap}=\varnothing, \qquad \sum_{v\in V}δ(v)\le\frac{n^2-1}{2}. \] Consequently, either $E_{\cap}\ne\varnothing$ or $\sum_{v\in V}δ(v)\ge\lceil n^2/2\rceil$ guarantees a rainbow Hamilton cycle. We further prove that the local Ore-type condition \[ δ(u)+δ(v)<n, \qquad d^c(u,v)\le2 \quad\Longrightarrow\quad uv\in E_{\cap} \] already forces a rainbow Hamilton cycle. Along the way, the Fan-type hypothesis yields a rainbow Hamilton path and a rainbow $(n-1)$-cycle. For diagonal systems, our conditions reduce to the classical Fan and local Ore conditions. The proofs combine rainbow rotation–extension with a double-counting argument in an auxiliary digraph. Keywords: graph system; rainbow Hamilton cycle; rainbow distance; Fan-type condition; Ore-type condition; rotation–extension.

fields
Combinatorics
pdf ↗ page arXiv:2610.13977 ↗

We define a class of flows on compact metric spaces that we call $C^α$ metric Anosov flows, $0 < α< 1$ or $α= \mathrm{Lip}$. We give verifiable criteria for a metric Anosov flow to be $C^α$. We show that $C^α$ metric Anosov flows have Hölder continuous local product structure, and in particular Hölder continuous holonomies. We deduce that topological conjugacies between $C^α$ metric Anosov flows are automatically Hölder continuous. We show that any $C^α$ metric Anosov flow admits a strong (i.e. Hölder continuous) Markov coding. This yields strong ergodic properties for equilibrium states for Hölder potentials for transitive $C^α$ metric Anosov flows. The regularity of the local product structure is proved in a non-compact setting which applies for the geodesic flow on the universal cover of a locally CAT(-1) space. This allows us to obtain Hölder regularity results for the holonomy maps for pinched negative curvature manifolds with respect to the Sasaki metric. This contrasts with the result of Ballmann, Brin and Burns that there are examples in this class for which the dynamical distributions are not Hölder.

comments
v2, 33 pages, 2 figures
fields
Dynamical Systems
pdf ↗ page arXiv:2610.14340 ↗

This paper explores topological phase transitions in the momentum space of the $κ$-deformed $φ^4$ lattice model within the mean-field approximation. We show that the double-well structure of the momentum space gives rise to degenerate momentum states at high energies and that this change is marked by a phase transition. We investigate this transition using both numerical and analytical methods, with the analytical approach based primarily on singularity theory. Our numerical analysis shows that the energy at which the topological phase transition occurs is related to the noncommutative parameter by the simple power law $E \sim 1/α^{4}$. By mapping the constant-energy-level surface to the $A_{3}$ singularity, we obtain the same result analytically. We also present results unique to noncommutative systems, showing that the fraction of particles in degenerate momentum states has a nonzero commutative limit. As a nonperturbative effect, this result provides a testable prediction for the existence of noncommutative spacetime.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.15644 ↗

We study Galois coinvariants of the geometric étale fundamental group and étale cohomology of varieties over a finitely generated field $k$. We establish finiteness properties of these coinvariants for (possibly infinite) Galois extensions $L/k$ such that $\operatorname{Gal}(L/k)^{\mathrm{ab}}$ has finite exponent. As applications, we obtain finiteness results for $L$-rational torsion points on abelian varieties and étale fundamental groups of arithmetic schemes over number fields.

comments
17 pages, added subsection 3.1 and corrected minor typos
fields
Algebraic Geometry, Number Theory
pdf ↗ page arXiv:2610.16760 ↗

Let $p$ be an odd prime and let $Σ$ be a nonempty finite set of rational primes different from $p$. In the cyclotomic tower $\mathbb{Q}(μ_{p^{n+1}})$, we determine the kernels of the natural maps from the minus $p$-primary ray class groups, with modulus the product of the primes above $Σ$, to the corresponding ordinary minus $p$-primary class groups. We give their character components and elementary divisors at every finite layer, including the layers before splitting stabilizes. In the norm inverse limit, we obtain an explicit integral cyclic decomposition of the local module modulo the single diagonal global Tate module. Its coinvariants recover the finite local kernels. Together with ordinary class-group control, this gives exact control for the minus ray class groups. We identify the local Frobenius factors with Bernoulli regularization factors. For nonexceptional characters, the classical main conjecture then yields regularized measure generators of the ray class characteristic ideals. The exceptional Teichmüller component is treated without the main conjecture: the single Tate correction corresponds to the single pole of the $p$-adic zeta function. At the bottom layer, the local calculation and Herbrand–Ribet yield characterwise nonvanishing criteria involving regularized Bernoulli values. For a single auxiliary prime, the exceptional component detects the Wieferich condition. We also deduce Eisenstein congruence consequences from existing modular-form theorems.

comments
36 pages
fields
Number Theory
pdf ↗ page arXiv:2610.17649 ↗

It is shown that the random prime Dirichlet series $ X(s)=\sum_{p\in \mathcal{P}}\frac{ξ_p}{p^{s+1/2}} $ can be almost surely approximated by a Brownian motion, where $ \{ξ_p\}_{p\ge 1} $ are independent and identically distributed random variables with zero mean and finite variance, and $ \mathcal{P} $ is the set of prime numbers. As a consequence, the functional law of the iterated logarithm, limit laws of the iterated logarithm, and Chung's law of the iterated logarithm for $ X(s) $ are obtained. It is also shown that the random Dirichlet series $ Y(s)=\sum_{k=2}^{\infty} \frac{(\log k)^α}{k^{s+1/2}} $ with $ α=-1/2 $ admits the same approximation as $ X(s) $, and a direct relationship between $ X(s) $ and $ Y(s) $ with $ α=-1/2 $ is established. The strong invariance principle for $ Y(s) $ with $ α>-1/2 $ is also obtained.

comments
28 pages
fields
Probability
pdf ↗ page arXiv:2610.18353 ↗

We study reversible Higgs bundles on a compact Riemann surface $Σ$ of genus $g\geq 2$, that is, $G$-Higgs bundles $(E,Φ)$ with $(E,Φ)\cong(E,-Φ)$. For a complex reductive group $G$ realized as a self-adjoint subgroup of $GL(n,\mathbb{C})$, we show that if $(E,Φ)$ corresponds to a representation $ρ$ of $π_1(Σ)$, then $(E,-Φ)$ corresponds to $(\overlineρ^{\,t})^{-1}$. As a consequence, the moduli space of reversible Higgs bundles is the fixed locus of an involution of the character variety and carries the structure of a real affine algebraic set. For a connected semisimple group and every topological type, we prove that the reversible locus in the smooth moduli space is a $(B,A,A)$-brane. We then describe reversibility through spectral data: in even rank, for a smooth spectral curve, reversible Higgs bundles correspond to line bundles invariant under the involution $η\mapsto-η$; in odd rank, under natural genericity assumptions for which the spectral curve is nodal, we give a gluing criterion together with an explicit polystability condition. Finally, for $SL(2n,\mathbb{C})$ we construct a relative Fourier–Mukai transform over the regular locus of the Hitchin base and prove that it maps the structure sheaf of the reversible locus to an object whose support is exactly the regular degree-zero $PSp(2n,\mathbb{C})$-locus in the Langlands dual $PGL(2n,\mathbb{C})$-moduli space. This support is of type $(B,B,B)$.

comments
Added references, part of introduction revised
fields
Algebraic Geometry, Mathematical Physics, Differential Geometry
pdf ↗ page arXiv:2610.19037 ↗

We study finite two-player zero-sum stochastic games in which the stream of expected stage payoffs is evaluated by a translation-invariant charge on the set of stages. Every patient valuation, that is, every diffuse charge that obeys the time value of money principle, is translation-invariant. By Mertens and Neyman (1981), every finite two-player zero-sum stochastic game has a value; in particular, each player has, for every $\varepsilon>0$, a strategy that is $\varepsilon$-optimal for all patient valuations simultaneously. We show that this robustness breaks down for translation-invariant charges, even when the charge is fixed in advance. In the Big Match with $0$--$1$ payoffs, for every charge, Player~2 can guarantee that Player~1's payoff is at most $1/2$, but cannot guarantee any smaller bound. Nevertheless, we construct a fixed translation-invariant charge such that, for every strategy of Player~1 and every $δ>0$, Player~2 has a pure reply yielding payoff less than $δ$. Consequently, the Big Match has no value under this charge. More generally, the supremum of the payoffs that Player~1 can guarantee ranges over $[0,1/2]$ as the translation-invariant charge varies. This contrasts with finite Markov decision processes, where every stationary Blackwell optimal strategy is optimal for every translation-invariant charge.

fields
Optimization and Control
pdf ↗ page arXiv:2610.19646 ↗

We propose an event-triggered zero-order hold implementation of the funnel controller for output reference tracking of unknown nonlinear systems. The control input is updated only when the deviation between the current and the last transmitted input exceeds a prescribed threshold, reducing the number of input updates while maintaining closed-loop performance guarantees by incorporating an auxiliary integral state to compensate for the error introduced by the zero-order-hold input. We establish a uniform lower bound on the inter-event intervals excluding Zeno behavior, and derive a growth condition on the nonlinear dynamics that ensures global existence of the solution for arbitrary, predefined sampling schedules. The theoretical results are illustrated by numerical simulations.

fields
Optimization and Control
pdf ↗ page arXiv:2610.22153 ↗

In the thirties of the last century, I.\,M. Vinogradov proved that the inequality $\|pα\|\le p^{-1/5+\varepsilon}$ has infinitely many prime solutions $p$, where $\|\cdot\|$ denotes the distance to a nearest integer. This result has subsequently been improved by many authors. In particular, Vaughan (1978) replaced the exponent $1/5$ by $1/4$, using his celebrated identity for the von Mangoldt function and a refinement of Fourier analytic arguments. The current record is due to Matomäki (2009), who showed that the inequality $\|pα\|\le p^{-1/3+\varepsilon}$ has infinitely many prime solutions. This exponent $1/3$ is considered the limit of current technology. Baier and the author established an analogue of Matomäki's result for the function field $k=\Fq(T)$ and its imaginary quadratic extensions. Moreover, for real quadratic extensions of $k$ of class number one, they obtained an analogue of Vaughan's result (exponent $1/4$). In this paper, we establish the analogue of Matomäki's result for all real quadratic extensions of $k$ without any restriction on the class number.

comments
24 pages
fields
Number Theory
pdf ↗ page arXiv:2610.23742 ↗

This paper presents a proof of the long-standing Yau Geometric Conjecture.

comments
24 pages; added a comparison with related work and streamlined the exposition. Main results unchanged
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.24636 ↗

We study the asymptotic behaviour of families of fractional energies normalized according to the Bourgain–Brezis–Mironescu formula and defined on functions satisfying a pinning condition on possibly aperiodic families of small obstacles. The result of the simultaneous limit depends crucially on the critical rescaling chosen for the obstacles.

comments
Added: Convergence of minimizers for linearly perturbed functionals, Gamma-convergence for homogeneous Dirichlet boundary conditions. Expanded bibliography
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.25515 ↗

In this paper, we study the three-dimensional incompressible stationary Navier–Stokes equations with the Coriolis force in the half space with the inhomogeneous Dirichlet boundary condition. We show the existence of a unique small solution in the scale-critical Besov spaces. In particular, for any boundary data with a vanishing normal component, and for any external forces, we establish the well-posedness of the inhomogeneous Dirichlet boundary value problem when the speed of rotation is sufficiently large. In the proof, to exploit the stabilizing effect of the Coriolis term, we employ two explicit formulas depending on the relation between the horizontal frequency and the magnitude of the rotation speed.

comments
60 pages
fields
Analysis of PDEs
pdf ↗ page arXiv:2610.25694 ↗

We study three-dimensional $BF$ theory as the large-level limit of the doubled Chern-Simons theory $G_k\times G_{-k}$ for simply connected, compact, simple $G$, a limit that connects the symmetry topological field theories of finite and continuous symmetries. Lines whose two highest weights grow linearly with $k$ at fixed difference become $BF$ lines, with magnetic flux set by their average and electric charge equal to their difference. Representing level-$k$ torus states as normalized flux-charge wave packets, we prove that every fixed modular transformation of any fixed normalizable state with regular flux converges in the strong-$*$ topology to its $BF$ counterpart. The Dehn twist gives each line exactly its Aharonov-Bohm phase at every level, while $S$ converges to a Fourier transform exchanging flux and charge, without escaping to singular holonomies or large charges. States that approach singular fluxes at large $k$ are not covered. The natural boundary 2d theory is speculated to be a decompactified boson in the abelian case $G=U(1)$, and $\dim G$ non-compact free scalars for general $G$. For $SU(2)$, we discuss the gapped interfaces between adjacent levels, which must shift some flux or charge by a finite amount. In the $k\rightarrow\infty$ limit the interface is conjectured to be the $c=1$ Runkel-Watts theory.

comments
85 pages, 1 figure
fields
Mathematical Physics, Quantum Algebra
pdf ↗ page arXiv:2610.26735 ↗

We present FOMTools, an open-source Python package for developing, analyzing, and deploying first-order algorithms for parametric linear and convex quadratic optimization. The user specifies an iteration in Python, with syntax close to its mathematical description. The package represents the iteration as a computation graph of affine and piecewise affine operations. To analyze worst-case performance, FOMTools formulates a mixed-integer verification problem that maximizes a convergence residual over a specified set of problem instances and initial iterates. For probabilistic performance analysis, a batched GPU simulator collects algorithm trajectories and computes statistical bounds on the probability of exceeding a specified convergence tolerance. From the same algorithm description, FOMTools generates specialized solver code for embedded hardware, multicore CPUs, and GPUs. The package first generates the numerical data, factorizations, and matrix-vector routines without any agent. A coding agent then translates the iteration into C or CUDA code and iteratively optimizes the implementation. After each phase, the package runs the generated code on sampled inputs and compares it with a reference execution. FOMTools also serves as a CVXPYgen backend, generating and optimizing solvers for parametric CVXPY problems. We demonstrate FOMTools on sparse coding, network optimization, optimal control, and portfolio rebalancing. The verification bounds reveal worst-case behavior that sampling misses. The generated solvers run up to two orders of magnitude faster than NumPy implementations, and they outperform hand-written GPU code and the OSQP solvers generated by CVXPYgen.

fields
Optimization and Control
pdf ↗ page arXiv:2610.26743 ↗

Filtering by spatial convolution underlies coarse-graining and Large Eddy Simulation, the analysis of multiscale dynamics, and data processing across the sciences. Yet, as commonly practiced on the sphere, it violates a basic consistency requirement: commutation with differential operators, which is necessary for deriving coarse-grained dynamics and preserving the physics across lengthscales. The root of the problem is nonlocality of filtering, which compares directions in different tangent planes. The same reason underlies the Foucault pendulum's precession, and it is why the commutation failure is intrinsically geometric rather than being due to the coordinate system or numerical grid. A recently proposed generalized spherical convolution is, for a given scalar kernel, the unique linear operator commuting with the radial projection, divergence, and curl. Any consistent linear operator, be it spectral truncation, wavelet transform, diffusion, Green's-function inversion, or a machine-learned convolution, necessarily takes this form. From a simulation of an unstable jet, we demonstrate commutation to within numerical round-off, whereas common filtering that ignores curvature produces significant spurious divergence where none exists and gradient errors comparable to the gradients themselves.

comments
Submitted to Europhysics Letters: Focus Issue on Complex Flows and Complex Fluids associated with the StatPhys29 satellite conference
fields
Mathematical Physics
pdf ↗ page arXiv:2610.26896 ↗

We give a computer-assisted proof that at most seven congruent infinite circular cylinders in $\mathbb{R}^3$ can touch pairwise. This answers a question of Littlewood from 1968. For cylinders of arbitrary radii we prove the upper bound ten. Whether ten can be attained remains open.

fields
Metric Geometry, Combinatorics
pdf ↗ page arXiv:2610.26940 ↗

We study turn-based deterministic zero-sum games with discounted and long-run average costs. For discounted costs, we establish exponential and superpolynomial iteration lower bounds for the simple strategy-iteration method with Dantzig's rule under unrestricted costs and unit weights, respectively. To the best of our knowledge, we establish for the first time that the method is not strongly polynomial, and this remains true even when all action costs belong to $\{-1,0,1\}$. For long-run average costs, we extend the Dantzig counterpart in~\cite{zhongye2026} to the two-player setting and prove that it terminates at an optimal strategy profile. However, a similar counterexample construction establishes an exponential iteration lower bound for termination with unrestricted costs. This construction does not extend to unit weights, for which we prove that the algorithm admits a strongly polynomial implementation.

fields
Optimization and Control
pdf ↗ page arXiv:2610.27386 ↗

In this work, we establishes results of existence and regularity for solutions to nonlocal equations of the form $f(Δ)u = U(x,u(x)) ~\text{and}~f(Δ_{g})u=U(x,u(x))$ in the Euclidean context and Riemannian context respectively. We extend the results of \cite{Gorka 6} for the Euclidean space $\mathbb{R}^{n}$ under cylindrically symmetry and compact Riemannian manifolds $(M,g)$ under action of a closed Lie subgroup $G \leq \text{Isom}_{g}(M)$. Our approach combines spectral theory, complex interpolation and fixed-point methods.

comments
27 pages
fields
Analysis of PDEs, Mathematical Physics
pdf ↗ page arXiv:2610.28546 ↗

We study an analogue of the permutohedron for the dihedral group, which we call the dihedral permutohedron. We determine the facet structure of the dihedral permutohedron, use the facet structure to construct a triangulation, and determine a nice combinatorial formula for the volume of the dihedral permutohedron. Lastly, we find a family of interesting unimodular affine symmetries of the odd dihedral permutohedron.

comments
21 pages, 4 figures
fields
Combinatorics
pdf ↗ page arXiv:2610.28672 ↗

This paper is the first in a series of two which aims at introducing a new family of groups called $k$-shuffle braid groups, denoted $\mathcal{B}_{n}(k)$. These groups have two intrinsically related stories. They were first observed by Calegari in arXiv:2105.11265 by looking at spaces of complex monic polynomials of degree $n$ whose critical values lie in $\mathbb{C} \setminus \{1, e^{\frac{2πi}{k}}, \dots, e^{\frac{2πi (k-1)}{k}}\}$. The $k$-shuffle braid groups are the fundamental groups of those spaces, which correspond geometrically to complements of $k$ translates of the discriminant variety. The second story starts from the poset of $k$-shuffle partitions, which were introduced by Drew Armstrong and explored in depth in his thesis arXiv:math/0611106. In this paper, we focus on the combinatorial aspects of $k$-shuffle braid groups starting from the poset of $k$-shuffle non-crossing partitions. We build a CW-complex on which the groups act freely, properly discontinuously and cocompactly, which we show is CAT(0) for dimension up to 7. For higher dimension, we show that the CAT(0) property is implied by the conjecture of Brady and McCammond that the dual braid complex is CAT(0) (arXiv:0909.4778).

comments
This is the first paper in a series of two, the second one is available at arXiv:2610.33890
fields
Geometric Topology, Combinatorics, Group Theory
pdf ↗ page arXiv:2610.28986 ↗

Semantic communication jointly optimizes inference accuracy, channel use, and computation, yet the underlying communication program is typically specified before training. This paper presents ComRSI (Communications-Oriented Recursive Self-Improvement), a framework for recursively evolving that program through large-language-model-assisted code edits. An agent modifies the encoder, channel adapters, decoder, loss, and training policy; a protected evaluator screens candidates under prescribed budgets and resource constraints. Accepted code and textual feedback inform subsequent proposals, while candidate weights are trained from scratch. In the original ten-generation CIFAR-100 lineage, eight accepted revisions produce an attention-pooled code achieving 74.3% test accuracy over an Additive White Gaussian Noise (AWGN) channel at 32 complex symbols per image, compared with 67.2% at 256 symbols for the initial design. The program also improves fading-channel accuracy and transfers to Tiny-ImageNet after retraining, where a separate ComRSI search demonstrates further recursive improvements. A repeat study with ten searches per method compares ComRSI against non-recursive agents with and without textual history. Mean final search scores balancing accuracy, transmission rate, and computational cost are 0.7587, 0.7376, and 0.7089, respectively; the mean numbers of accepted revisions per search are 6.0, 5.0, and 2.1. The advantage over the memoryless agent is supported by a two-sided rank test, whereas the difference from the history-equipped agent remains statistically inconclusive at this sample size. Archived proposals show that pooling can yield a large improvement, while retained history supports subsequent refinement. These results support recursive program optimization within the evaluated setting and expose its remaining rate-robustness and compute-placement trade-offs.

fields
Information Theory
pdf ↗ page arXiv:2610.29641 ↗

A non-zero $m \times n$ (0, 1)-matrix $A$ is called a {\em nullspace matrix} if each entry $(i, j)$ of $A$ has an even number of $1's$ in the set of entries consisting of $(i, j)$ and its rectilinear neighbors. These matrices correspond to even dominating sets of $m \times n$ grid graphs. We study these matrices and consider those that cannot be constructed from smaller nullspace matrices.

fields
Combinatorics
pdf ↗ page arXiv:2610.31475 ↗

We prove the all-exp property for the Traveling Salesman Problem with the $2$-opt neighborhood, and for $(3,2,2)$- and $(2,3,2)$-Maximum Constraint Assignment with the flip neighborhood. The all-exp property states that infinitely many instances have an initial solution from which every sequence of improving moves to a local optimum, and hence every run of local search, has exponential length. For the Traveling Salesman Problem, the all-exp property was previously known for every $k$-opt neighborhood with $k \ge 3$, so our result for $2$-opt completes the picture for all $k$-kopt neighborhoods. In $(p,q,r)$-Maximum Constraint Assignment, each constraint involves at most $p$ variables, each variable occurs in at most $q$ constraints, and each domain has size $r$. Decreasing any of the parameters of $(3,2,2)$ or $(2,3,2)$ yields a problem in which every sequence of improving flips has polynomial length, so both Maximum Constraint Assignment results are tight. All three results settle long-standing open problems. Our proofs build on a common construction idea, which reveals a close connection between the local-search dynamics of the three problems.

comments
36 pages
fields
Combinatorics
pdf ↗ page arXiv:2610.31513 ↗

Let $RS_q(k)$ denote the full-length Reed–Solomon code over $\mathbb{F}_q$. Every word $u=(u_x)_{x\in \mathbb{F}_q} \in \mathbb{F}_q^q$ has a unique interpolation polynomial $U(X)\in \mathbb{F}_q[X]$ of degree less than $q$, satisfying $U(x)=u_x$ for all $x\in \mathbb{F}_q$. It is well known that every word with $°U=k$ is a deep hole. The Cheng–Murray conjecture asserts that these are the only deep holes for $3\le k\le q-2$, except when $q$ is even and $k=q-3$. Previous results established the conjecture for $k\le p-1$ and for $k\ge\lfloor(q-1)/2\rfloor$, where $p$ is the characteristic of $\mathbb{F}_q$. In this paper, we prove the Cheng–Murray conjecture in the remaining range $3\le k\le\lfloor q/2\rfloor$, thereby resolving it completely. Our proof combines two complementary approaches. The tangent envelope of a marked arc yields differential constraints that establish the conjecture for $3\le k\le q/2$ when $q$ is even, and for $3\le k\le\lceil q/3\rceil$ when $q$ is odd. For the remaining cases with $q$ odd, we combine local syndrome compatibility with combinatorial arguments to rule out additional deep holes.

comments
33pages
fields
Information Theory
pdf ↗ page arXiv:2610.31531 ↗

Multi-robot coordination of nonlinear systems is naturally formulated as joint-space optimal control subject to joint constraints. Existing methods that approach joint optimality remain largely centralized and offline, while decentralized, real-time counterparts target safe navigation without pursuing the joint optimum. This paper develops a decentralized, real-time planner that approaches the joint optimum as an anytime process. We extend a recent single-robot framework that searches dispersive trees of forward-propagated commands to coordinate a multi-robot team near-optimally. We first show that the inter-robot coupling can be effectively replaced by some local constraints, under which locally grown dispersive trees contain a near-optimal agreement. It is this structure that the robots pursue by a negotiation driven by decentralized column generation. In the quiescent limit, this negotiation is shown to settle on a near-optimal agreement. Our real-time protocol, WarpWeaver, runs the negotiation asynchronously over a receding horizon and closes on quiesced agreements even when deadlines cut it short. In simulation, WarpWeaver reduces total travel time by up to 22% against centralized offline planners on the Dynobench suite and by up to 16% against a decentralized real-time baseline, and outperforms a task-specific planner on networked navigation. Hardware experiments on six robots planning onboard demonstrate real-world implementability.

comments
18 pages, 5 figures, 7 tables
fields
Optimization and Control
pdf ↗ page arXiv:2610.31540 ↗

In this paper, we classify profinite groups whose closed subgroups are subnormal of defect at most two, providing profinite versions of theorems due to Heineken and to Mahdavianary. We also establish that if G is a finitely generated periodic pro-π group, where π is a finite set of odd primes, and all its procyclic subgroups are subnormal of defect at most 2, then G is a finite π-group in which every abelian subgroup is subnormal of defect at most 2.

comments
v2: updated references and notation, text corrections
fields
Group Theory
pdf ↗ page arXiv:2610.31878 ↗

For two-weight fractional difference energies with inner and outer exponents $1<p<q$, we study the optimality of the Hardy constant generated by a homogeneous power. We prove nontrivial sharpness intervals for every sign of $D=d+2β-sp$, including a two-sided interval when $D=0$, and strict loss of optimality near both finite-energy endpoints. For $q\ge2$, sharpness extends beyond the elementary pointwise scalar criterion through an integrated remainder estimate: after opposite logarithmic displacements are combined, a positive contribution at bounded displacements absorbs the negative contribution at larger displacements. A single quadratic reference energy controls this comparison and the angular fluctuations of near-extremizers. When $1<p<q<2$, optimization of the scalar supporting-line inequality gives the exact power exponent for that criterion and a larger explicit sharpness interval. Angular variations provide explicit strict-loss examples. The method extends to flat singular sets and regional half-space energies, with separate finiteness ranges for test energies and the homogeneous candidate. Complementary results for $p>q$ include flat-subspace inequalities. The constants are optimal on the stated smooth test classes and extend unchanged to their difference-norm completions whenever they are positive.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.32671 ↗

We prove Wang's planar $D$-coloring conjecture. A $D$-coloring of a graph is a proper edge coloring in which every subgraph isomorphic to $K_4$ minus an edge is rainbow. Wang conjectured that every planar graph of maximum degree $Δ$ has a $D$-coloring with $9$ colors if $Δ=4$, $10$ colors if $Δ=5$, and $2Δ-1$ colors if $Δ\ge6$. Hu, Kong and Wang proved the cases $Δ\le5$ and $Δ\ge33$. We prove the sharp bound $2Δ-1$ for every $Δ\ge6$, settling the remaining cases. The proof uses discharging; the case $Δ=6$ is computer-assisted.

comments
22 pages, 3 figures. Ancillary files contain the certificates and verification scripts for the computer-assisted case of maximum degree 6
fields
Combinatorics
pdf ↗ page arXiv:2610.33322 ↗

We compute derived products and inverse limits in Grothendieck categories by exact coresolutions. For the product of $r$ doubled affine lines over $B[s]$, where $B$ is any nonzero commutative ring,the exact dimensions of products, epimorphic countable limits, and arbitrary countable limits are $r$, $r$, and $r+1$. We identify the top derived functors for every $B$-module of coefficients. Both obstruction modules contain the countable reduced product as a natural $B$-module direct summand, and a formal-series map relates them by an explicit cokernel sequence. A relative cofree coresolution gives degree shifts over an arbitrary Grothendieck category and computes the localization defect for arbitrary locally nilpotent diagrams. We also construct transfinite scalar extensions preserving a prescribed kernel and its derived limit;these give nonzero completion defects beyond any fixed ordinal.The general epimorphic top-degree vanishing question is not settled by these constructions.

comments
Withdrawn for substantial revision by authors
fields
K-Theory and Homology
pdf ↗ page arXiv:2610.34022 ↗

The triangle map is an area-preserving, piecewise-parabolic map of the torus with vanishing Lyapunov exponent. We study the symbolic process $X_n=\pm1$ given by the two halves of the torus, whose partial sums $S_n$ are the transport of the lifted momentum, with an integrator free of accumulated rounding errors and exact computations on polygonal cylinder sets. We prove that the Koopman operator has no non-constant $H^1$ eigenfunctions, that the symbol is not an $H^1$ coboundary, and that runs of equal symbols are uniformly bounded. For generic parameters the walk is normally diffusive and Gaussian, but its diffusion coefficient is not that of a coin: away from the edges of the parameter square it varies between $0.2$ and $14$ times the Bernoulli value, and for a benchmark pair, with $D=3.09\pm0.01$, the excess is produced by the order of the bounded inter-event intervals, not by their distribution. The first-passage exponent is $1/2$ with a non-Bernoulli amplitude, and worst-case Birkhoff averages decay as for a uniquely ergodic system. For $β=0$ the Green-Kubo sum cancels, $D=0$, yet $\mathrm{Var}(S_n)$ keeps growing slowly, sign changes are rare and the correlations of Fourier modes decay as $n^{-0.2}$. We identify the zero-sum words, which become pure twists at $β=0$, as resonances that trap orbits with a computable $1/\ell$ law, set the crossover time $β^{-1/2}$ at small $β$, and, in an asymmetric deformation of the symbol, produce a logarithmic term in the variance whose coefficient we compute exactly. We formulate conjectures on unique ergodicity, mixing, the central limit theorem and the cohomology of the symbol.

comments
v2: added declarations and AI-use statement, restructured for journal submission
fields
Mathematical Physics
pdf ↗ page arXiv:2610.34179 ↗

A state over time (stote) pairs a state $ρ$ with a channel $\mathcal{E}$ as the Hermitian, generally non-positive operator $(ρ\otimes\mathbf{1})\star J_\mathcal{E}$. Characterizing the convex cone $\widehat{\mathcal{Q}}$ these operators generate was posed as an open problem in Hoogsteder-Riera, Calsamiglia and Winter (2026). We exhibit a linear copy of the bipartite separable cone as an exposed face of $\widehat{\mathcal{Q}}$: for output systems of the form $B = A'R$ with $A'$ a copy of the input, there is an explicit witness $W_0\in\widehat{\mathcal{Q}}^*$ and an explicit injective linear map $L$ with $σ\in \mathrm{SEP}_{+}(A{:}R)\iff L(σ)\in\widehat{\mathcal{Q}}(A{:}A'R)$. Composing with the NP-hardness of separability shows that exact membership, weak membership, and weak optimization are NP-hard for $\widehat{\mathcal{Q}}$ and its dual. A quantitative stability bound makes the weak-membership reduction valid at inverse-polynomial accuracy. Thus polynomial-time membership tests do not exist unless $\mathrm{P}=\mathrm{NP}$. Independently of this complexity assumption, Fawzi's theorem implies that $\widehat{\mathcal{Q}}(A{:}A'R)$ admits no finite semidefinite representation when $d=\dim A\ge2$, $r=\dim R\ge2$, and $d+r>5$.

comments
23 pages, 2 Figures
fields
Mathematical Physics
pdf ↗ page arXiv:2610.34972 ↗

We determine all Borel-reducibility comparisons within the family of benchmark equivalence relations in Gao's monograph, enlarged by the universal treeable relation $E_{\infty T}$ and its countable power $E_{\infty T}^ω$. We prove $E_\infty^ω\leq_B c_0$, $E_\infty\nleq_B\ell_p$ for $2<p<\infty$, $\ell_p\nleq_B{\equiv_m}$ for $2<p<\infty$, $c_0\nleq_B{\equiv_m}$, $c_0\nleq_B\ell_\infty^+$, and $\equiv_m\nleq_B\ell_\infty^+$. Together with the classical result, these give $\ell_p\leq_B{\equiv_m}$ exactly when $p\leq2$. The same methods show that $E_{\infty T}$ is incomparable with every $\ell_p$, $1\leq p<\infty$. The comparisons answer questions recorded in Gao's monograph and Kanovei's monograph. We also prove that unitary equivalence of all bounded linear operators is not Borel reducible to $\equiv_m$, answering a question of Ding and Gao. Equivalently, unitary equivalence of arbitrary representations of $F_\infty$ is strictly more complicated than that of arbitrary representations of $\mathbb Z$, answering a question of Thomas.

comments
45 pages
fields
Logic, Functional Analysis
pdf ↗ page arXiv:2610.35025 ↗

To date, a Leray's weak solution to the Navier-Stokes equations enjoys an energy inequality. We prove the existence of a weak solution that enjoys the energy equality. The existence of the weak solution is achieved as limit of solutions to the mollified Navier-Stokes system (Leray's approach). Since no uniqueness result is known for weak solutions we are not able to claim that any weak solution in the sense of Leray-Hopf enjoys the energy equality.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.35517 ↗

We describe a family of convex continuous functions $E_{F,\varepsilon}$, $\varepsilon\in(0,1]$, on the set of states of a bipartite quantum system (consisting of finite-dimensional or infinite-dimensional subsystems) obtained by applying the Mora-Piani-Briegel construction of $\varepsilon$-measure (proposed in [Phys. Rev. A 109, 052413]) to the Entanglement of Formation (EoF) with the use of the trace norm distance in the role of basic metric. These functions are non-increasing in $\varepsilon$ and converge pointwise to the EoF as $\varepsilon\to0^+$. We conducted a detailed analysis of properties of the functions $E_{F,\varepsilon}$, using (among others) the relationship between the family $\{E_{F,\varepsilon}\}_{\varepsilon\in[0,1]}$ and the family $\{E^λ_F(ρ)\}_{λ>0}$ of Moreau-Yosida approximations of the EoF explored in [arXiv:2609.30246]. We obtain a variational expression for the functions $E_{F,\varepsilon}$ (which for $\varepsilon=0$ turns into the standard variational expression for the EoF). We also describe the weak subadditivity property of the functions $E_{F,\varepsilon}$ and analyse their continuity properties. The semicontinuity bounds for the EoF (presented in [Lob.J.Math., 46, 2632-2658],[arXiv:2404.07978]) allow us to obtain easily computable upper bounds on the rate of uniform convergence of the function $E_{F,\varepsilon}$ to the EoF as $\varepsilon\to0^+$ on the sets of states with bounded rank/energy of one of the marginal states. Applicability of the obtained results to other entanglement measures, in particular, to the relative entropy of entanglement, is discussed. As an auxiliary result, it is shown that for any state of an infinite-dimensional bipartite quantum system, there exists a nearest separable state w.r.t. the trace norm distance. The non-uniqueness of such a state is also shown.

comments
38 pages, preliminary version, one comments and references are welcome
fields
Mathematical Physics, Functional Analysis
pdf ↗ page arXiv:2610.35797 ↗

We extend the recently introduced e-closure principle for multiple testing to general state and action spaces and to arbitrary, including continuously indexed, families of risks. A collection of e-values indexed by states and risks certifies a menu of actions, and every data-dependent selection from this menu controls each expected loss at its prescribed level. Conversely, every valid procedure is certified by its own normalized losses, which often leave part of the risk budget unused. We introduce a method called recertification, which spends this unused budget. Using known distributional structure, it enlarges the certifying e-values while preserving validity and selects a less conservative action from the enlarged menu. One particularly compelling application is to the popular framework of conformal risk control (CRC) with a known loss family. Here, recertification redistributes unused risk budget among the held-out points of each hypothetical augmented sample. The resulting controller is valid under every finite exchangeable law and less (usually strictly) conservative than CRC. For convex losses on an interval of actions, repeated recertification converges geometrically to an admissible controller. As a second example involving false discovery rate control with e-values, recertifying randomized e-BH yields a closed-form procedure that rejects a superset of randomized e-BH for every realization of the randomization and maximizes the expected number of rejections under the same constraints. Recertifying with the larger e-collection of closed e-BH gives a second closed-form procedure that rejects a superset of the first. Experiments on CIFAR-100 hierarchical classification and simulations demonstrate improved budget utilization and gains in informativeness and power over the corresponding baselines.

fields
Statistics Theory
pdf ↗ page arXiv:2610.35821 ↗

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 these obstructions and some recent examples of \cite{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 $(X,H)$, no ordinary instanton bundle with bounded quantum number, no non-ordinary $μ$-semistable instanton bundle with bounded quantum number and no bundle with supernatural cohomology and bounded roots.

comments
v2: added a new result (Cor. 2) on instanton and supernatural bundles and a new section on instanton bundles; changed accordingly title and abstract
fields
Algebraic Geometry
pdf ↗ page arXiv:2610.36231 ↗

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 ↗

Place closed Euclidean balls of radius r at the points of (1/2, 1/2, ..., 1/2) + Z^n. The view-obstruction problem for spheres asks for the least r such that every ray from the origin into the open positive orthant meets a ball. Equivalently, it asks for D_n = sup_v min_{t in R} sum_i ||t v_i - 1/2||{Z}^2 over positive integer vectors v, where ||x||{Z} denotes the distance from x to the nearest integer. The values for n <= 5 are classical. We prove D_6 = 107/312, attained only by permutations of (12, 9, 7, 5, 3, 2), and D_7 = 73/180, attained only by permutations of (11, 10, 8, 7, 4, 3, 1). The proof reduces the supremum to a finite search. The key facts are an exact lower bound d_n(v) >= delta^2_U for every rational subspace U perpendicular to v, a matching fiber upper bound, and Banaszczyk's transference theorem, applied along the Korkine-Zolotarev flag of the dual lattice v^perpendicular intersection Z^n. The search is quotiented by signed coordinate permutations and carried out entirely in exact rational arithmetic. It terminates with a certificate: no node of the search is left unresolved, and every surviving candidate is evaluated exactly. The method also recovers the known values for n = 3, 4, 5.

fields
Number Theory, Combinatorics
pdf ↗ page arXiv:2610.37711 ↗

Let $\mathcal{S}_χ$ be a cuspidal representation of $\mathrm{GL}_2(q)$. A subset $S \subseteq \mathrm{GL}_2(q)$ is {$t$-intersecting} if $\dim_{\mathbb{C}}\{v \in \mathcal{S}_χ: g_1v = g_2v\} \geqslant t$ for any $g_1, g_2 \in S$. In this paper, we determine $\max{|S|}$ where $S$ is $t$-intersecting.

fields
Combinatorics
pdf ↗ page arXiv:2610.38472 ↗

We study the mean first-reaction time (MFRT) of a Brownian particle in a bounded three-dimensional domain $Ω$ with a reflecting boundary $\partialΩ$, that contains a thin reactive filament. The filament is a tube of cross-sectional diameter $\varepsilon\ll 1$ around a smooth closed curve $Γ$, and its surface is partially reactive with reactivity $b$. As $\varepsilon\to 0$ the filament reduces to a line source in the outer problem. Its cross-section enters only through a logarithmic reactive capacitance $d_c(b)$, so that the MFRT is O$(-\log[\varepsilon d_c])$. This contrasts with the $O(\varepsilon^{-1})$ scaling for spatially localized traps of radius $\varepsilon$. Using matched asymptotic expansions, we derive a Fredholm integral equation on $Γ$ for the line density of the reactive flux. It sums all powers of $ν\equiv -1/\log[\varepsilon d_c]$ and involves two geometric quantities: the regular part $R$ of the Neumann Green's function, which encodes the interaction of $Γ$ with $\partialΩ$, and an intrinsic self-energy of $Γ$ that depends only on its shape. The framework extends to multiple filaments, splitting probabilities, a spatially heterogeneous diffusivity, and the principal eigenvalue of the Laplacian. For the unit sphere we solve the integral equation with a simple trapezoidal Nyström scheme, and we validate the predictions for rings, linked rings and torus knots against full 3-D finite-element solutions.

fields
Analysis of PDEs
pdf ↗ page arXiv:2610.38715 ↗

These six lectures, delivered at the University of New Brunswick 6-21 October 2026, are an introduction to instantons in quantum mechanics and quantum field theory for senior undergraduate and beginning graduate students. The first four lectures deal solely with particle mechanics, building the necessary background for the last two that cover applications in semi-classical field theory.

fields
Mathematical Physics
pdf ↗ page arXiv:2610.38717 ↗

Let $G$ be a quasi-split reductive group over a finite extension $F$ of $\mathbb{Q}_p$. We study the Fargues-Scholze categorical Langlands correspondence at generic parameters $φ$, namely those whose semisimplification does not coincide with that of any parameter with nontrivial monodromy. Extending the theory of quasi-exceptional sets on the nilpotent cone developed by Bezrukavnikov to the setting of disconnected reductive groups which arise as stabilizers of semisimple generic parameters, to any irreducible representation $ρ$ of such a group we construct sheaves labeled $\overline{\nabla}_ρ$, $\overlineΔ_ρ$, $IC_ρ$, and $δ_ρ$ on the moduli stack of L-parameters for $G$, which were conjectured to exist by Hansen. Using forthcoming work of Hamann, Hansen, and Mann on geometric Eisenstein series compatibility, the (conjectural, in general) behavior of the correspondence at generous parameters, and the (conjectural, in general) agreement between Fargues-Scholze and classical semisimple L-parameters we prove that these sheaves correspond under the Fargues-Scholze functor to the sheaves $i_{b!}π$, $i_{b\sharp}π$, $i_{b\sharp!}π$, and $i_{b!*}π$ on $\text{Bun}_G$, where $π$ is a smooth irreducible representation with Fargues-Scholze parameter $φ$. As applications we prove t-exactness properties of the Fargues-Scholze functor specialized at a generic parameter, and that Hecke operators are perverse t-exact on the generic locus, as conjectured by Hansen and Hamann-Lee. We also construct and study foundational properties of Hecke eigensheaves at generic parameters, compute non-semisimplified Fargues-Scholze parameters on the generic locus, and construct a Fargues-Scholze-theoretic version of the BM-O bijection at generic parameters.

comments
50 pages, comments welcome
fields
Representation Theory, Number Theory
pdf ↗ page arXiv:2610.38954 ↗

We estimate the Lévy distance to the symmetric Bernoulli distribution for normalized Boolean convolution powers of a probability measure with mean zero and variance one. If the absolute moment of order $r$ is finite, we obtain the upper bounds $o(n^{-(r-2)/6})$ for $2<r<5$ and $O(n^{-1/2})$ for $r\ge5$. The proof relies on an inequality that controls the Lévy distance to the Bernoulli law by the values of the reciprocal Cauchy transform at two points. For each $r\ge2$, we also construct a single fixed distribution whose absolute moments are finite exactly up to order $r$, and show that the exponent $\min\{(r-2)/6,1/2\}$ cannot be improved. Under only a finite second moment assumption, the convergence can be arbitrarily slow.

comments
19 pages
fields
Probability, Operator Algebras
pdf ↗ page arXiv:2610.39743 ↗

Military installation energy readiness depends on having sufficient resources to sustain mission-critical loads throughout a prolonged utility outage. This paper develops a method to help operators manage energy resources during normal grid-connected operation while preserving islanded readiness. A robust viability kernel identifies combinations of battery energy, fuel, and load-shifting capability sufficient to sustain critical load under uncertain demand and photovoltaic generation. A probability bound translates stochastic generator start failures and recovery into a deterministic generator count selected to meet a specified availability requirement at islanding. Case studies for a fourteen-day islanding requirement at 0.999 islanding-entry availability show that advance warning of an outage can be used to prepare resources to meet a higher readiness requirement before islanding begins. The case studies further demonstrate that economic-only operation and the evaluated fixed battery reserve policy can fail to sustain critical load for the required duration, whereas enforcing the resource-dependent requirement preserves readiness at a modest increase in modeled grid-connected operating cost. These results demonstrate how active resource management can preserve long-duration energy readiness. The framework also helps commanders understand how readiness changes with operating conditions, even when the available resources remain the same.

comments
32 pages, 5 figures, 3 tables, submitted to Military Operations Research
fields
Optimization and Control
pdf ↗ page arXiv:2610.40078 ↗

We prove a uniform form of the dynamical Manin-Mumford conjecture for split endomorphisms $Φ= \left(f_1,\ldots,f_n\right): \left(\mathbb{P}^1\right)^n \to \left(\mathbb{P}^1\right)^n$ over $\mathbb{C}$, where each $f_i$ has degree $d \geq 2$ and at most one $f_i$ is exceptional. Specifically, we bound the number of irreducible components of $\overline{\operatorname{Prep}\left(Φ\right)\cap V\left(\mathbb{C}\right)}^{\operatorname{Zar}}$ uniformly, depending only on $\left(n,d,\operatorname{deg} V\right)$. We also obtain uniform statements for the dynamical Bogomolov conjecture. Our approach combines a degeneration argument via ultrafilters with a new non-archimedean rigidity theorem, along with the quantitative dynamical Zhang's inequality by Mavraki and the second author.

comments
Comments welcome!
fields
Dynamical Systems, Number Theory
pdf ↗ page arXiv:2610.40152 ↗

Get each issue by email →