Browse

All works

Browse the Hexagon catalog.

Newest works

1229 works

math.CV — Complex Variables

On meromorphic functions sharing three values IM and one value CM

Contributed by Pengcheng Fang, Tao Hu, Yanyang Li, Quanyu Tang, Haiqi Zhang

We prove that if two nonconstant meromorphic functions on the complex plane share three distinct values ignoring multiplicities and a fourth value counting multiplicities, then they share all four values counting multiplicities. This gives an affirmative answer to Gundersen's three-IM-one-CM problem in full generality.

Author comments: This revised version adds Haiqi Zhang as a coauthor. The author list has been updated with the agreement of all coauthors.

A mix of human-written and AI-generated textHuman understanding: some partsMeromorphic functionsNevanlinna theoryShared values
Nov 18, 202626 pages
PDFView work

math.DS — Dynamical Systems

Inclusions and invariant forms of Svinin's recurrences

Contributed by Jason Turner

We prove Svinin's Conjectures 5.2 and 5.3 for sequences with all entries nonzero and give explicit coefficient maps for both inclusions. If arbitrary solutions of the polynomial equations are allowed, both inclusions fail at every order. We also prove Conjecture 3.1 on the nonzero-coordinate domain for every g≥1g \ge 1, giving an invertible parameter correspondence between the associated recurrence and its invariant form.

A mix of human-written and AI-generated textHuman understanding: some parts
Nov 18, 20269 pages
PDFView work

math.PR — Probability

A sharp lower bound for entropy first-passage skewness in a three-state Markov ring

Contributed by Samuil Petkov

We study the time at which the accumulated medium entropy of a continuous-time three-state Markov ring first reaches or exceeds a positive threshold. For six positive finite transition rates and positive cycle affinity, the skewness scaled by the square root of the threshold has a finite large-threshold limit. We prove that its infimum at fixed affinity AA is min⁡{s(A),s(A/2),s(A/3)}\min\{s(A),s(A/2),s(A/3)\}, where ss is the one-stage biased-walk expression. This establishes the three-state case of the lower envelope conjectured by Wampler and Barato. An exact passage decomposition on the lifted integer chain controls the cumulants, including the initial-phase mixture. Two Hermitian entropy tilts constrain the characteristic coefficients to an outer region on which derivative-sign arguments reduce the bound to three endpoint values. Homogeneous rates attain the third value; positive finite-rate families approach the first two. An appendix records which steps extend to rings with any number of states.

A mix of human-written and AI-generated textHuman understanding: all partsMarkov jump processentropy productionfirst-passage timeskewnessstochastic thermodynamicsunicyclic network
Nov 18, 202610 pages
PDFView work

math.RA — Rings and Algebras

The repeated-pair symmetric nonnegative inverse eigenvalue problem in dimension five

Contributed by Samuil Petkov

We prove a necessary condition for a real symmetric entrywise nonnegative matrix of order five to have spectrum (1,a,a,b,b)(1,a,a,b,b), where 1≥a≥b≥−11\ge a\ge b\ge-1. Besides the trace condition 1+2a+2b≥01+2a+2b\ge0, such a realization requires 1+4ab≥01+4ab\ge0 or 1+a+2b≥01+a+2b\ge0. Together with the known sufficiency of these conditions, this gives the repeated-pair characterization recorded as Conjecture 5.1.2 in Clark's dissertation. The proof treats arbitrary realizations. Perron-root minimization and a two-dimensional projector frame reduce a putative counterexample to a weighted five-cycle or an HH support. A cycle-cone inequality excludes the first case; an exact identity for the HH support excludes the second at every positive trace. No constant row sum, uniform Perron vector or restriction to irreducible matrices is imposed in the statement. On the open curved part of the boundary, where 1+4ab=01+4ab=0, 1+2a+2b>01+2a+2b>0 and 1+a+2b<01+a+2b<0, the same argument determines all realizations: up to permutation similarity they form an explicit one-parameter family of weighted five-cycles containing the Loewy–McDonald matrix.

A mix of human-written and AI-generated textHuman understanding: all partsPerron rootSNIEPinverse eigenvalue problemnonnegative matricesrealizable spectraweighted cycles
Nov 18, 202615 pages
PDFView work

math.CO — Combinatorics

Counterexamples, Spectral Obstructions, and Deletion Stability for WOW-284

Contributed by Samuil Petkov

WOW-284 is a Graffiti conjecture asserting that every connected graph of order at least three and girth at least five has minimum dual degree at most −λmin⁡(D)-\lambda_{\min}(D), where DD is the distance matrix and the dual degree of a vertex is the average degree of its neighbors. We refute it with exact counterexamples on 3838, 3939, 4040, 4242 and 5050 vertices and study the structure of regular counterexamples. For a connected kk-regular graph GG of girth at least five and diameter three, we prove that δ∗(G)+λmin⁡(D)=2k−2−max⁡θ(θ+1)2\delta^*(G)+\lambda_{\min}(D)=2k-2-\max_\theta(\theta+1)^2, the maximum being over the nonprincipal adjacency eigenvalues, so strict failure is a spectral window condition. The closed window ∣θ+1∣≤2k−2|\theta+1|\le\sqrt{2k-2} alone, without any diameter assumption, forces ∣V(G)∣<(k+2)(k2+3)/6|V(G)|<(k+2)(k^2+3)/6 when k≥4k\ge4, and an integral excess matrix derived from the optimal polynomial sharpens this bound for diameter-three counterexamples. Every regular strict counterexample has degree at least six and diameter at most four, and in degree six it has at most 5050 vertices. Under vertex deletion, the Hoffman-Singleton graph remains a strict counterexample after the removal of any five vertices, while an explicit six-vertex deletion destroys strictness. A degree-5757 Moore graph, if one exists, remains a connected strict counterexample after the removal of any 5656 vertices. Every strict counterexample has at least 1515 vertices, and an exhaustive search excludes orders up to 1919. The strict induced subgraphs of the Hoffman-Singleton graph of orders 3636 to 5050 are classified up to automorphism; among them are three further irregular counterexamples of order 4040.

Author comments: This is version 2 of arXiv:2607.27452 (https://arxiv.org/abs/2607.27452). The source code, exact certificates and build instructions are at https://github.com/SamPetkov/wow284 ; the verification scripts and data used in the paper are included in anc/.

A mix of human-written and AI-generated textHuman understanding: all partsGraffiti conjectureMoore graphcounterexampledistance spectrumdual degree
Nov 18, 202644 pages
PDFView work

math.CO — Combinatorics

Obligatory triple systems: atoms, spectra and decompositions

Contributed by Samuil Petkov

Erdős Problem 593 asks which finite triple systems embed in every triple system of uncountable chromatic number. Such finite systems are called obligatory. We describe their structure and the ways of decomposing them while keeping their original hyperedges. Every obligatory system without isolated points has canonical pieces, called atoms: single triples and expansions of finite 22-connected bipartite graphs, obtained by adding one new private point to each graph edge. A forest records how the atoms meet and reconstructs the system. We determine exactly which numbers of vertices, hyperedges and components can occur, and which atom counts occur in a connected system, including the parity restriction when the vertex–hyperedge incidence graph has just one independent cycle. For a fixed system, we consider decompositions into connected pieces that meet in at most one point and whose shared-point incidences form a forest. Each such decomposition is specified by independently partitioning the atoms through each shared point. These choices, ordered by merging blocks, form a product of partition lattices. An exact piece-count formula identifies the single local mergers. Li gave the first public classification. The appendix includes our self-contained proof, developed independently in parallel with Li's work.

Author comments: The first public classification of obligatory triple systems is due to E. Li, arXiv:2606.24882 (https://arxiv.org/abs/2606.24882); the classification proof in the appendix was developed independently in parallel. The principal statements are matched to a Lean 4 development (Lean 4.32.0, pinned Mathlib) in the public repository https://github.com/SamPetkov/Erdos593 , including the self-contained file Erdos593SelfContained.lean.

A mix of human-written and AI-generated textHuman understanding: all partsErdős problemscanonical decompositionhypergraph blocksobligatory triple systemspartition lattices
Nov 18, 202634 pages
PDFView work

math.CO — Combinatorics

A Full-Sequence Quantitative Gap Between the Chromatic and Cochromatic Numbers of a Random Graph

Contributed by Samuil Petkov

The cochromatic number ζ(G)\zeta(G) of a graph GG is the least number of parts in a partition of V(G)V(G) into cliques and independent sets, so ζ(G)≤χ(G)\zeta(G)\le\chi(G). Erdős and Gimbel asked whether χ(Gn)−ζ(Gn)→∞\chi(G_n)-\zeta(G_n)\to\infty with high probability for Gn∼G(n,1/2)G_n\sim G(n,1/2). Heckel answered this affirmatively for about 95%95\% of all integers nn and conjectured that the natural order of the difference along the full sequence is n/(ln⁡n)3n/(\ln n)^3. We prove that, as n→∞n\to\infty through all integers, χ(Gn)−ζ(Gn)≥0.063 n/(ln⁡n)3\chi(G_n)-\zeta(G_n)\ge0.063\,n/(\ln n)^3 with high probability. The failure probabilities are summable, so under any coupling of the graphs GnG_n the bound holds almost surely for all sufficiently large nn. The proof counts partitions of the vertex set into classes of five consecutive sizes, each class declared independent or complete. The two choices per class outweigh the entropy lost by restricting the class sizes, so the expected number of such partitions is large already at a number of classes that lies order n/(ln⁡n)3n/(\ln n)^3 below the first-moment lower bound for the chromatic number, uniformly in the rounding phase. A second-moment argument, which controls all overlaps between two such partitions, including whole shared classes and large intersections of unequal classes, bounds the probability that such a partition exists from below. Vertex-exposure concentration, together with a coloring of the leftover vertices, then raises this to high probability at a cost of o(n/(ln⁡n)3)o(n/(\ln n)^3) extra classes.

Author comments: This is version 2 of arXiv:2608.30604 (https://arxiv.org/abs/2608.30604). A Lean 4 formalization of an earlier version of the main theorem, with the weaker constant ((ln 2)^2/4) ln(200/153) ≈ 0.032 in place of 0.063, is registered on Palomar as PALOMAR-2026-09-02-000006: https://palomar-registry.org/entry?id=PALOMAR-2026-09-02-000006&version=1 . The improved constant 0.063 of this version is not formalized.

A mix of human-written and AI-generated textHuman understanding: all partsChromatic numberErdős problemscochromatic numberrandom graphsecond moment method
Nov 18, 202653 pages
PDFView work

math.FA — Functional Analysis

Daugavet renormings via poor subspaces

Contributed by Samir Hamad

Let YY be a separable Banach space with the Daugavet property containing a complemented copy of ℓ1\ell_1. We prove that Y⊕1EY\oplus_1 E admits an equivalent norm with the Daugavet property for every separable Banach space EE. We also construct a subspace of C[0,1]C[0,1] isomorphic to the James space such that the corresponding quotient fails the Daugavet property.

A mix of human-written and AI-generated textHuman understanding: all partsDaugavetJames spacepoor subspacesrenormings
Nov 18, 20267 pages
PDFView work

math.NA — Numerical Analysis

The Simultaneous Rectangle for Systems, without Homotopy: All the Solutions of a Polynomial System, Each in a Certified Ball, with a Proof That None Is Missing

Contributed by Ivan BESEVIC

This note carries the simultaneous rectangle of one polynomial (version G8) to a system of nn polynomial equations in nn unknowns, and removes the homotopy on which the previous version S1 relied. The points start on the solutions of a start system and are swept on FF itself with a rectangle step of order three, each point deflated of its nearest neighbours; nothing is tracked. What the sweeps do not find is obtained in three ways: the images of the solutions found by the symmetries read on the system; runs of the step of one point from random points; and the law of the residues: by the formula of Euler and Jacobi the solutions found determine the moments of the missing ones, which are then the generalised eigenvalues of small matrices. Every point passes one certificate (a contraction of a ball in interval arithmetic, and a box by Rouché's theorem for a double solution), and the output is complete when as many disjoint balls are certified as a bound of the number of solutions. For sparse systems that bound is the mixed volume, computed by mixed cells; we show that one lifting, whatever it is, gives the mixed volume as soon as an exact check meets no tie, so that the count is a proof and not a probability. With the Rust crate pandsysroots 0.21.0, on 40,000 systems of twenty-seven families the output is as expected on all of them and proved complete on 39,998; on eighteen classical or large systems, up to 6561 solutions and 64 unknowns, completeness is proved on every one, and the program is faster than a reference homotopy solver with certification on fourteen of them, level on two and slower on two (one machine, default settings). The algebra of the step, the certificate, the laws used by the residues and the argument of the lifting are checked in Lean 4 (64 theorems). The note states which results are formal proofs, written proofs, quoted theorems and numerical observations, and its limits: no theorem of convergence, no certificate for a multiplicity above two, no proof of completeness for a system with a curve of solutions.

Author comments: Version S2, 17 November 2026. Also deposited on Zenodo with the Lean files, the programs of the figures and the measurements: https://doi.org/10.5281/zenodo.24314654. Software: https://crates.io/crates/pandsysroots (version 0.21.0). Companion for systems of The Simultaneous Rectangle (G8), https://doi.org/10.5281/zenodo.24257672.

Primarily AI-generated textHuman understanding: some partsEuler-Jacobi formulaLean 4certified solutionsmixed volumepolynomial systems
Nov 18, 202614 pages
PDFView work

math.LO — Logic

Algorithmic and Provable Precision of Monotone Approximations

Contributed by Zhenyu Sun

We study how much information is needed to approximate a monotone limit and when arithmetic can prove that a finite approximation is accurate. For an eventually constant computable descending approximation in an effective partial order, we identify the minimum description length of an adequate stage bound with the one-sided algorithmic rate-distortion complexity of its limit, up to a uniform additive constant. For integer limits, this is also the complexity of a coarse quotient. Finite proof maxima give normal forms for bounded reflection and refutation completeness, and determine every provably sufficient budget. Under soundness for the positive and negative threshold families and refutation completeness at the chosen rank, the least provable error is the difference between the true threshold and its greatest provable lower bound. This difference retains the joint information in two opposing monotone limits. Chaitin counts give matching precision and Busy Beaver bounds, including a maximum law for finitely many fixed reference machines. We also obtain sharp advice bounds for variable certificate presentations, fixed-rank refutations over a finite arithmetic base, an elementary verified selector for bounded positive reflection under explicit proof-presentation assumptions, and resource bounds for proof translation.

A mix of human-written and AI-generated textHuman understanding: all partsKolmogorov complexityalgorithmic rate distortionarithmetical interpretabilitybounded reflectionmonotone approximationproof selectors
Nov 18, 202642 pages
PDFView work

cs.DS — Data Structures and Algorithms

A self-contained proof of the exponent threshold 0.4875754036 for worst-case Subset Sum

Contributed by Jennifer Voitle

We give a complete written proof that Subset Sum on n positive integers of polynomial bit length is decided by a randomized word-RAM algorithm in time O(2^(T'n)) for every fixed T' > T = 0.4875754036, in particular in time O(2^(0.4875755n)). The threshold T was reported earlier by the author, conditionally on the non-numerical lemmas of two preprints; the contribution of this paper is the derivation that removes that condition. The algorithm is the one introduced by OpenAI (isolation, a capped preliminary search, a filtered modular checksum, subtraction of aliases) with the alias step of A. Kapila (one large shared block, sparse masks, alias phase sums) and a single asymmetric checksum filter. Those ideas are theirs. What is done here is different in kind: every non-numerical lemma that the result uses is stated with its hypotheses and proved in this text, so that the argument can be checked without accepting unproved statements from other preprints. The numerical inequalities that the proof uses are listed explicitly and matched one by one to a ball-arithmetic certificate, whose analytic error bounds are also proved here. The theorem has no mathematical hypothesis beyond two cited classical theorems and those certified inequalities. The manuscript also records, statement by statement, how each proof was obtained and which independent checks are still open. It has been reviewed and audited, with the sources open, by a second AI system; it has not been refereed.

Author comments: 49 pages. This paper removes the condition in the author's earlier preprint hexagon:2611.00178 ("A certified exponent of 0.48758 for worst-case Subset Sum, and limits of the alias-checksum analysis"), which stated the same threshold assuming lemmas of two other manuscripts. Certificate code and stored output: https://github.com/jv1000/subset-sum, folder selfcontained-2026 (MIT License).

A mix of human-written and AI-generated textHuman understanding: some partsAI-assisted mathematicsFourier filteringSubset Sumball arithmeticcomputer-assisted proofexact algorithmsinterval arithmeticmeet-in-the-middlerandomized algorithmsworst-case complexity
Nov 18, 202649 pages
PDFView work

math.OC — Optimization and Control

A Fifth-Order Hurwitz Interval Polynomial Family Without a Degree-Four Weak Strictly Positive Real Numerator

Contributed by Zhengtai Xiong, Tiancheng Xu

We construct an explicit fifth-order Hurwitz-stable interval polynomial family with five independent coefficient intervals of positive width that admits no common weak strictly positive real (WSPR) numerator of degree four. For every real numerator of degree at most four, one of six fixed denominator--frequency pairs has a nonpositive real part. Robust Hurwitz stability follows from three common separator points for the even and odd denominator polynomials. Nonexistence follows from an exact linear identity: a strictly positive integer combination of the six frequency pairing functionals is identically zero. Thus robust Hurwitz stability does not guarantee a degree-four common WSPR numerator for fifth-order interval polynomial families.

Primarily AI-generated textHuman understanding: some parts
Nov 18, 20266 pages
PDFView work

math.GR — Group Theory

Counterexamples to the Mathieu Property: From Explicit Quotient Constructions to a General Classification

Contributed by Roy van Rijn

In 1997, Olivier Mathieu conjectured that if every positive power of a function on a compact connected Lie group has average zero, then multiplying those powers by any fixed function should eventually preserve this property. Christopher D. Long disproved this conjecture in 2026 and classified the compact connected groups for which the property holds. This paper extends that classification to a much broader family of algebraic spaces, including nonspherical homogeneous spaces and disconnected groups. We prove a necessary and sufficient condition determining exactly when the Mathieu–Zhao property holds. The proof constructs new explicit counterexamples using the binary octahedral and icosahedral groups, covering cases not addressed by earlier classifications. These counterexamples remain valid under polynomial transformations and give rise to transcendental generating functions. The main result completes the classification of the Mathieu–Zhao property for complex homogeneous spaces whose acting groups and stabilizers have reductive identity components, extending the known results from compact groups and spherical spaces to this general setting.

Primarily AI-generated textHuman understanding: some partsHaar integrationInvarient theoryMathieu ConjectureMathieu-Zhao subspacesReductive algebraic groupsVanishing moments
Nov 18, 202618 pages
PDFView work

math.NT — Number Theory

Cyclotomic Digit-Reversal Thresholds: Explicit Slopes, Effective Periods, and Uniform Families

Contributed by Oleksiy Babanskyy

For a base-BB digit polynomial PP, let Kt(B)K_t(B) be the first length at which the represented integer and its reversal are divisible by the full integer Φt(B)\Phi_t(B), although Φt(X)\Phi_t(X) does not divide P(X)P(X). We study this threshold through a lattice of evaluations and the capacities imposed by digit positions. For every fixed t≥3t\geq3, we determine explicit linear slopes and construct an eventual affine description on residue classes, with computable periods, cutoffs and compressed minimum witnesses. No total-unimodularity assumption is needed. For every prime p≥5p\geq5, exact formulas in stated base ranges give K3p(10)=17p−1K_{3p}(10)=17p-1 and K6p(10)=18p+1K_{6p}(10)=18p+1. Ordered capacities explain their different correction periods, two and four. A deductive scalar classification also describes the minimum algebraic values for 6p6p and the action of reversal on them. Period transfers show that excluding palindromes leaves the first thresholds unchanged for both families. The same framework treats independent coefficient bounds and collisions. Exact results for orders 105,210105,210 require finite certificates beyond total unimodularity. The general construction is effective; minimal periods and practical efficiency are not claimed.

Author comments: 51 pages. Exact-arithmetic companion 1.1.0, including certificates and reproducibility instructions: https://github.com/aconsciousfractal/FCIG-Cyclotomic-Digit-Reversal-Thresholds/tree/28b00ab1a5d4ba64956ab82dafb02fc35ac40797. No independent specialist review or formal verification is claimed. Previously deposited as aiXiv 2609.03287 v1; the mathematical content and sources are unchanged.

Primarily AI-generated textHuman understanding: no partsCyclotomic Polynomialsaffine monoidscongruence latticesdigit reversalquasi-polynomialszonotopes
Nov 18, 202651 pages
PDFView work

math.CO — Combinatorics

Cyclotomic Weight Chambers for Reverse-Complement-Invariant Minimum Decycling Sets

Contributed by Oleksiy Babanskyy

For DNA order eight, the fixed labelled weights (s(A),s(C),s(G),s(T))=(−ρ,−1,1,ρ)(s(A),s(C),s(G),s(T))=(-\rho,-1,1,\rho) in W\mathcal W, ρ>1\rho>1, with fixed A<C<G<TA<C<G<T ties meet 21 ordinary complete-move components. Their entire union has optimum residual path length 141, including non-RC endpoints. A certified reverse-complement-invariant minimum decycling set of 8230 words instead attains 107 and is optimal in a different component. The atlas classifies W\mathcal W, not all labelled antisymmetric weight maps. Certified 8232-word RC feedback sets attain length 105; the global cardinality problem remains 8230≤τRC(106)≤82328230\le\tau_{\mathrm{RC}}(106)\le8232. A two-family selector determines equality strata, and an exact atlas has 86 walls, 173 strata and 114 terminals. Boolean classical retiming optimizes entire components; prescribed RC mate sums give two paired-move sectors, each costing at most one additional path unit. Beyond a component, conditional stable completion is exactly 2-SAT; bounded coordinates and antisymmetric heights describe endpoint and cardinality problems. Balanced closed-walk loads have an exact signed-real finite-pool closure. A certified exterior passes one such relaxation but requires at least 8235 selections in its stated real-gradient class at bound 106, without a global lower-bound claim. Classical cyclotomic tensor relations and bounded signed-one-hot realization also give an all-order Farey spectrum of rational DNA embedding resonances. A real algebraic ratio is positionally realizable at some even order exactly when it belongs to a cyclotomic field. Complete arithmetic proofs and finite local exclusions are retained in the technical supplement.

Author comments: 56 pages (57 in the original aiXiv 2610.00003 v1 compilation). Mathematical content unchanged; explicit LaTeX counter aliases preserve lemma/proposition reference types across toolchain versions; companion version 3.2.0. The 141 optimum concerns the union of components met by the specified labelled weight family. The 8230-word RC witness attains length 107 and is optimal in its own component; global optimality remains open. Companion and reproduction instructions: https://github.com/aconsciousfractal/FCIG-Cyclotomic-Weight-Chambers-for-RC-Invariant-MDS/tree/752199436a9605798af9dc9117577724002a750a

Primarily AI-generated textHuman understanding: no partscyclotomic fieldsde Bruijn graphsexact certificatesminimum decycling setsretimingreverse complement
Nov 18, 202656 pages
PDFView work

math.OC — Optimization and Control

Robust Strictly Positive Real Synthesis for Polynomial Polytopes

Contributed by Tiancheng Xu

We determine when Hurwitz stability of a polynomial polytope guarantees a common numerator of prescribed degree. The denominators are real and monic of degree nn, and their convex hull has at most mm vertices. For numerator degrees n−1n-1 and nn, the guarantee holds exactly when n≤2n\le2 or m≤2m\le2. For degree n+1n+1 there is one additional case: every Hurwitz-stable cubic triangle admits a common quartic numerator. We prove this existence result by classifying the possible contacts of a nonnegative frequency pairing. In every order n≥4n\ge4 we construct a Hurwitz-stable triangle excluding all numerators of degree at most n+1n+1.

Primarily AI-generated textHuman understanding: some partsHurwitz stabilitycommon numeratordegree constraintpolynomial polytoperobust positive realness
Nov 18, 20267 pages
PDFView work

math.CO — Combinatorics

An upper bound for three-sunflower-free families of four-element sets

Contributed by Oleksiy Babanskyy

Let M(4,3)M(4,3) be the maximum size of a finite simple family of four-element sets containing no three distinct members whose pairwise intersections are equal. The empty intersection is included in the forbidden configurations. We give a computer-assisted exclusion of families of size 6969, giving the bound 54≤M(4,3)≤6854\le M(4,3)\le 68. The argument combines an imported bound of 2727 for intersecting families, structural analysis of point-links, exact trace constraints and finite exhaustions. The proof is organized at the cardinality 6969. A cap on families containing a complete R16R16 point-link provides a shorter final contradiction. We make the imported results, computational lemma contracts and remaining verification limits explicit. The lower construction is classical. Neither exclusion of size 6868 nor optimality of the lower construction is claimed.

Author comments: 17 pages. Manuscript unchanged from aiXiv 2610.01292 v1. The claimed bound is 54 <= M(4,3) <= 68, with empty kernels included; the forcing threshold is at most 69. This is partial progress on the finite four-uniform case, not a solution of the general sunflower conjecture (Erdos Problem #20). Pinned companion: https://github.com/aconsciousfractal/Four-Uniform-Three-Sunflower-Bound/tree/90c03c79bcf44eb8f61dff1b86091b822cc4cd3c

Primarily AI-generated textHuman understanding: no partscomputer-assisted proofexact certificatesextremal set theoryfour-uniform hypergraphssunflowers
Nov 18, 202617 pages
PDFView work

math.NT — Number Theory

Sun's supercongruence for two Apéry-like binomial sums at discriminant -20

Contributed by Oleksiy Babanskyy

Let an=∑j=0n(nj)2(2jj)a_n=\sum_{j=0}^n\binom{n}{j}^2\binom{2j}{j} and let Qn=∑j=0n(nj)(−8)n−j∑k=0j(jk)3Q_n=\sum_{j=0}^n\binom{n}{j}(-8)^{n-j}\sum_{k=0}^j\binom{j}{k}^3. We prove Sun's Conjecture 3.7: for every prime p≠2,5p\ne2,5, the two sums (−1)(p−1)/2∑n<p(2nn)an20−n(-1)^{(p-1)/2}\sum_{n<p}\binom{2n}{n} a_n20^{-n} and (−1)(p−1)/2∑n<p(2nn)Qn(−16)−n(-1)^{(p-1)/2}\sum_{n<p}\binom{2n}{n} Q_n(-16)^{-n} are congruent modulo p2p^2 to 4x2−2p4x^2-2p if p=x2+5y2p=x^2+5y^2, to 2x2−2p2x^2-2p if 2p=x2+5y22p=x^2+5y^2, and to zero if (−5/p)=−1(-5/p)=-1. For split primes, two modular periods at one complex multiplication point yield separate finite specializations of an integral Hecke polynomial. Its degree-pp endpoint and the sign in the nonprincipal class are treated explicitly. For inert primes, a degree-five endomorphism with square [−5][-5] forces a Hasse polynomial to vanish; a finite quadratic differential invariant then lifts each original sum from modulo pp to modulo p2p^2. Fixed algebraic certificates and their uniform specialization are included.

Author comments: 18 pages. Revised manuscript, package 0.1.1, unchanged from aiXiv 2610.00158 v2. Simplified notation and updated attribution to Mao following comments on the companion manuscripts. Companion code and reproduction instructions: https://github.com/aconsciousfractal/Sun-3-7-Supercongruence/tree/4c2e584d1f04d2aa3921e4398743eecc7431004c

Primarily AI-generated textHuman understanding: some partsApéry-like sequencesElliptic CurvesFranel numberscomplex multiplicationmodular formssupercongruences
Nov 18, 202618 pages
PDFView work

math.NT — Number Theory

Sun's supercongruence for two Apéry-like binomial sums

Contributed by Oleksiy Babanskyy

Let fn=∑j=0n(nj)3f_n=\sum_{j=0}^n\binom{n}{j}^3 and Qn=∑k=0n(nk)(−8)n−kfkQ_n=\sum_{k=0}^n\binom{n}{k}(-8)^{n-k}f_k. We prove that, for every odd prime pp, the two sums ∑n=0p−1(2nn)Qn/(−32)n\sum_{n=0}^{p-1}\binom{2n}{n} Q_n/(-32)^n and ∑n=0p−1(2nn)Qn/64n\sum_{n=0}^{p-1}\binom{2n}{n} Q_n/64^n are congruent modulo p2p^2 to 4x2−2p4x^2-2p when p=x2+2y2≡1,3(mod8)p=x^2+2y^2\equiv1,3\pmod{8}, and to zero when p≡5,7(mod8)p\equiv5,7\pmod{8}. For inert primes a finite quadratic differential invariant gives ordinary and singular lifting results, including a formula for the defect at a simple singularity. For split primes we certify the modular period by identities of weight eight, construct an integral Hecke polynomial with an explicit degree bound, and specialize a finite polynomial identity at two complex multiplication points. The coefficient at degree pp, both classes of complex multiplication matrices, and the primes 33 and 55 are treated explicitly.

Author comments: 18 pages. Revised manuscript, package 0.1.1, unchanged from aiXiv 2610.00157 v2. Incorporates Zhi-Hong Sun's comments on notation and credits Mao's stronger coefficient congruence. Companion code and reproduction instructions: https://github.com/aconsciousfractal/Sun-3-6-Supercongruence/tree/db95a4dc988e548bc9fa6cb10745e1b9c2cd5bdd

Primarily AI-generated textHuman understanding: some partsApéry-like sequencesFranel numberscomplex multiplicationmodular formssupercongruences
Nov 18, 202618 pages
PDFView work

math.FA — Functional Analysis

The Matsaev (1971) conjecture is false for every p ≠ 2, and other miscellanea

Contributed by Dave Goel

The Matsaev (1971) conjecture asserts that for 1 < p < ∞, every contraction T on an Lᵖ space and every polynomial P satisfy ‖P(T)‖ ≤ ‖P(S)‖, where S is the shift on ℓᵖ(ℕ). We prove that it fails for every p ∈ (1, ∞) ∖ {2}, already for a real 2×2 matrix and a real quadratic, without numerical computation, and that it holds for every quadratic whose constant and leading coefficients have equal modulus; these results and their proofs also form the short paper "The Matsaev (1971) conjecture is false for every p ≠ 2". This expanded version adds an explicit quadratic and an explicit family of scaled rotations that violate the conjecture on a whole punctured neighbourhood of p = 2, with a proof by hand; an exact rational re-proof of Drury's counterexample at p = 4; upper bounds, certified in exact rational arithmetic, for the shift norms of Drury's polynomial at p = 3, 4, 6, …, 20; explicit counterexamples, certified by computer, for every p ∈ [40/39, 40], p ≠ 2; and conditions that every 2×2 counterexample must satisfy.

Author comments: First version: October 28, 2026. This is the expanded version of the paper "The Matsaev (1971) conjecture is false for every p ≠ 2", which is at https://hexagonmath.org/2611.00547 on Hexagon. The ancillary files are at https://doi.org/10.5281/zenodo.24290033 on Zenodo.

A mix of human-written and AI-generated textHuman understanding: some partsBellman functionL^p spacesMatsaev's conjecturecontractionsshift operatorvon Neumann's inequality
Nov 17, 202637 pages
PDFView work

math.FA — Functional Analysis

The Matsaev (1971) conjecture is false for every p ≠ 2

Contributed by Dave Goel

The Matsaev (1971) conjecture asserts that for 1 < p < ∞, every contraction T on an Lᵖ space and every polynomial P satisfy ‖P(T)‖ ≤ ‖P(S)‖, where S is the right shift on ℓᵖ(ℕ); for p = 2 this is von Neumann's inequality. Drury (2011) disproved it for p = 4, and hence, by duality, for p = 4/3, with a computer-assisted example. We prove that it fails for every p ∈ (1, ∞) ∖ {2}, even with T a real 2×2 matrix and P a real quadratic polynomial. Let Pₐ(z) = 1 + az − Bz² with B > 0. For p ≠ 2 and small a > 0 we show that ‖Pₐ(S)‖ = ‖Pₐ(J)‖, where J is the rotation of ℓᵖ₂ (ℂ² with the ℓᵖ norm) by a right angle. For p > 2 the bound ‖Pₐ(S)‖ ≤ ‖Pₐ(J)‖ comes from a Bellman function: a function G of two real variables such that, for every real sequence x ∈ ℓᵖ(ℕ) (with x₋₁ = x₋₂ = 0) and every n ≥ 0, |(Pₐ(S)x)ₙ|ᵖ ≤ ‖Pₐ(J)‖ᵖ|xₙ|ᵖ + G(xₙ₋₁, xₙ₋₂) − G(xₙ, xₙ₋₁); the G terms cancel in the sum over n. For 1 < p < 2 the bound follows by duality. If B > 1, a rotation by an angle slightly larger than π/2, scaled down to a contraction, gives a real 2×2 matrix T with ‖Pₐ(T)‖ > ‖Pₐ(J)‖ = ‖Pₐ(S)‖. In the other direction, the conjecture holds for every quadratic whose constant and leading coefficients have equal modulus. An expanded version of this paper, with many other results, is posted on Zenodo: https://doi.org/10.5281/zenodo.23691816.

Author comments: First version: October 28, 2026. The expanded version of this paper is at https://doi.org/10.5281/zenodo.23691816 on Zenodo.

A mix of human-written and AI-generated textHuman understanding: some partsBellman functionL^p spacesMatsaev's conjecturecontractionsshift operatorvon Neumann's inequality
Nov 17, 202613 pages
PDFView work

math.DG — Differential Geometry

The Osserman (1990) conjecture is fully resolved, and other miscellanea

Contributed by Dave Goel

The Osserman conjecture says that a Riemannian manifold whose Jacobi operators have the same eigenvalues in every unit direction is flat or locally rank-one symmetric. Chi and Nikolayevsky proved it in every dimension except one remaining case: dimension sixteen, when a Jacobi eigenvalue has multiplicity seven or eight. Nikolayevsky reduced this remaining case to a purely algebraic statement, the Nikolayevsky (2009) conjecture, his Conjecture A. We prove the Nikolayevsky (2009) conjecture. Hence the Osserman conjecture is true in every dimension, including its pointwise form in dimension sixteen. The same result completes the conformal Osserman conjecture, which is stated for dimensions greater than four. Conjecture A says that an Osserman algebraic curvature tensor on ℝ¹⁶ whose reduced Jacobi operator has an eigenvalue of multiplicity seven or eight has a Clifford structure or is a linear combination of the constant-curvature tensor and an orthogonal pullback of the Cayley-plane curvature tensor. For the two-eigenvalue case, we construct a Cartan–Münzner quartic and recover the original Jacobi eigenspaces from its isoparametric classification. For the remaining multiplicity patterns, we study a rational coherent factor of the Jacobi family. Its transpose pencil, polar divisors and residue images reduce the possible poles to cases excluded by degree bounds, real positivity and homogeneous factorization. A frame-independent closed-kernel identity completes the passage from local pole reductions to the algebraic classification.

Author comments: First version: October 28, 2026. This is the expanded version of the paper "The Osserman (1990) conjecture is fully resolved", which is at https://hexagonmath.org/2611.00544 on Hexagon.

A mix of human-written and AI-generated textHuman understanding: some partsCayley planeClifford structureJacobi operatorOsserman conjecturealgebraic curvature tensorisoparametric hypersurfacerank-one symmetric space
Nov 17, 202681 pages
PDFView work

math.DG — Differential Geometry

The Osserman (1990) conjecture is fully resolved

Contributed by Dave Goel

The Osserman conjecture says that a Riemannian manifold whose Jacobi operators have the same eigenvalues for all unit tangent vectors at all points is flat or locally rank-one symmetric. Chi and Nikolayevsky proved it in every dimension except one case in dimension sixteen: a Jacobi eigenvalue of multiplicity seven or eight. Nikolayevsky reduced this case to an algebraic statement, the Nikolayevsky (2009) conjecture: an Osserman algebraic curvature tensor on ℝ¹⁶ with a Jacobi eigenvalue of multiplicity seven or eight has a Clifford structure or is a linear combination of the constant-curvature tensor and an orthogonal conjugate of the curvature tensor of the Cayley projective plane. We prove this conjecture. Hence the Osserman conjecture is true in every dimension, and so is the conformal Osserman conjecture, which is stated for dimensions greater than four. For two eigenvalues, a Cartan–Münzner quartic built from the Jacobi operators, together with the classification of isoparametric hypersurfaces with four principal curvatures, shows that the tensor is Clifford or comes from the Cayley projective plane. For three or more eigenvalues, polynomial eigenfields settle nineteen multiplicity patterns. In the remaining sixteen patterns, a sums-of-squares argument treats the case where the eigenvalue of multiplicity seven or eight is the smallest or the largest; otherwise we study the poles of a rational factorization of the Jacobi operators: a topological rigidity lemma for quadratic Gram matrices settles the case without poles, and degree bounds, real positivity and homogeneous factorization exclude each pole type or force a Clifford structure.

Author comments: First version: October 2026. The expanded version of this paper is at https://doi.org/10.5281/zenodo.23603278 on Zenodo.

A mix of human-written and AI-generated textHuman understanding: some partsCayley planeClifford structureJacobi operatorOsserman conjecturealgebraic curvature tensorisoparametric hypersurfacerank-one symmetric space
Nov 17, 202663 pages
PDFView work

math.OA — Operator Algebras

Modular Kittaneh Numerical Radius Inequalities

Contributed by K. MAHESH KRISHNA

We show that for adjointable morphisms on Hilbert C*-module over a commutative unital C*-algebra, a version of breakthrough numerical radius inequalities obtained by Kittaneh [\textit{{Stud. Math.}, 2005}] hold.

Primarily human-written textHuman understanding: all partsHilbert C*-module.Numerical radius
Nov 17, 202612 pages
PDFView work

quant-ph — Quantum Physics

Double-Pulsar Drag Bounds Dissipative Sources of Diósi–Penrose Decoherence

Contributed by Marc Sperzel

In the proposal of Diósi and Penrose a mass in a superposition of two positions loses coherence at the rate Γ=CEG/ℏ\Gamma = C E_G/\hbar, first order in Newton's constant. We ask what could produce this rate for a static superposition of equal masses. A bath in its ground state cannot: its noise equals its absorption, which vanishes at zero frequency when the response is continuous there. For the free linearized gravitational field and a 1 μ1~\mug body the loss of coherence stays below 10−210^{-2} at all times, for separations up to a millimetre, in a scalar proxy normalized to EGE_G (it tends to a few times 10−310^{-3}), and it is reversible. These results are standard, at second cumulant. A bath that is not in a ground state can supply the rate, but if it couples to the mass density through a Hamiltonian, the noise that decoheres a body also takes momentum from it. We show that stationary, white, passive noise of Diósi–Penrose form drags a moving body at a rate of at least (2π/3)(CℏG/kBTeff)(2\pi/3)(C\hbar G/k_B T_{\mathrm{eff}}) times its density, with TeffT_{\mathrm{eff}} the fluctuation–dissipation temperature. If the noise keeps this form and one temperature out to the kilometre wavelengths that a neutron star resolves, as it does by construction in the linear-friction dissipative Diósi–Penrose model, the orbital decay of the double pulsar PSR J0737−3039A/B requires Teff/C>3.3×1016T_{\mathrm{eff}}/C > 3.3\times10^{16} K, up to strong-field corrections of order one. This is more than 5×10275\times10^{27} above the laboratory bound on that model and excludes the proposed Tβ≈3T_\beta \approx 3 K by 101610^{16}. Baths that differ between kilometre and laboratory scales escape. Otherwise what survives is noise without back-action: Diósi's original model with Gaussian mass-density smearing R0R_0. Published data allow R0>4.5R_0 > 4.5 Å C1/3C^{1/3} with no upper bound and do not fix CC. For a uniform sphere of radius a≫R0a \gg R_0 the rate is independent of R0R_0 to 1.7 R0/a1.7\,R_0/a at every separation dd, so one visibility measurement, with other decoherence channels subtracted, determines CC; for 1 μ1~\mug of silica τ=65\tau = 65 ps/C/C at d=1d = 1 mm and 0.35 μ0.35~\mus/C/C at d=1 μd = 1~\mum.

Author comments: Version 2 corrects statements that a re-examination of version 1 found to be wrong or too broad; no main result changes. (1) Sec. 5.2 and escape E2: the "granular" contribution to the drag assumed uncorrelated nucleons. In neutron-star matter the static structure factor at the relevant wavelengths is of order 1e-6 at R0 = 4.5 Angstrom, so the residual bound quoted for an infrared cut (T_eff > 0.07 K) is withdrawn; an infrared cut leaves no useful bound. (2) Abstract and Sec. 4.2: the coherence loss in the free linearized field is stated as an all-times bound (below 1e-2 for separations up to a millimetre, in a scalar proxy), not as a value it "stays at". (3) Prop. 4.2 now states the condition under which the dephasing exponent saturates. (4) Lemma 5.2: the orbital binding energy changes at twice the drag rate of the relative velocity; the period derivative is unchanged. (5) Smaller corrections: the asymptotic formula at d = 1 micrometre gives 0.343 microseconds/C (0.346 with the full function); compactness values; the Gran Sasso bound is a 0.95 Bayesian probability; the quadrupole origin convention and the adiabatic use of the drag theorem along the orbit are stated.

Primarily AI-generated textHuman understanding: some parts
Nov 17, 202625 pages
PDFView work

math.NT — Number Theory

A computer-assisted bound Λ ≤ 0.107964910079 for the de Bruijn–Newman constant

Contributed by Dale James Raymond

In the Polymath normalisation, a computer-assisted comparison gives Λ≤4276930720779424736615780984913961408125713216879677197516800<0.107964910079.\Lambda\le\frac{427693072077942473661578098491}{3961408125713216879677197516800}<0.107964910079. The initial squared zero height is bounded by 9/169/16, using OpenAI's ordinary-zeta 7/87/8 nonvanishing theorem formalised in Lean at the identified source revision. Finite signed translated-probe functionals, matched physical source estimates, and continuous-domain kernel inequalities supply a piecewise comparison for that height. A newly implemented exact rational consumer checks every one of the 221 010221\,010 arcs, their outward rounding and joins, and the complete final de Bruijn contraction. The fifteen-stage finite replay includes a direct reconstruction of all 6767 source chords from their primitive payments, with explicit binding of the main terms and the complete derivative functionals. Separate consumers verify their continuous kernel certificates and the actual physical and arithmetic charges. The analytic argument includes the corrected 11/1211/12 Gaussian remainder, finite-head localisation, attainment of a positive-time highest contact, and repeated complex contacts. The accompanying packet contains the mathematical sources, certificate data, checker programs, exact source identities, and replay instructions. The initial theorem is used at cited-theorem trust with authenticated external Lean-build evidence; a fresh local Lean build and an end-to-end formalisation of the present comparison are not claimed.

Author comments: This submission corresponds to version 1.0, dated 29 October 2026, of the manuscript deposited at https://doi.org/10.5281/zenodo.23752103. Complete computational evidence, verification instructions and component-specific rights notices remain available in that deposit and its companion repository: https://github.com/Dale-Dug/de-bruijn-newman-0107965. Related separate works: Michel Planat and Dale James Raymond, Certified Collision Exclusion and an Unconditional Upper Bound Λ < 0.156821 for the de Bruijn–Newman Constant, https://arxiv.org/abs/2611.06245 (subsequent joint paper); the earlier 0.149396 bound, https://doi.org/10.5281/zenodo.23781981; and the conditional 0.116398 bound, https://doi.org/10.5281/zenodo.23272725. These are separate works, not additional results claimed by this manuscript.

Primarily AI-generated textHuman understanding: all parts
Nov 17, 202663 pages
PDFView work

math.NT — Number Theory

Hyperbolic Collision Invariants and Certified Odd-Even Tiers toward Markoff Unicity

Contributed by Elleard Felix Webster Heffern

A collision would consist of two different ordered Markoff triples with the same largest entry MM. We derive arithmetic conditions that every such collision must satisfy. A parametrization by hyperbolic functions factors a classical determinant identity and leads to two coupled quadratic equations. If MM is odd, these equations require a factorization M=FEM=FE with F,E>1F,E>1 and gcd⁡(F,E)=1\gcd(F,E)=1. Hand proofs and exact finite computations give increasingly strong bounds, ending with F4,E4≥300000M+200000F^4,E^4\ge300000M+200000. If MM is even, division of the collision quantities by four gives the same inequalities for a coprime factorization of M/2M/2. These bounds yield explicit sufficient conditions for a Markoff number to be unique. The finite computations cover a proved search domain and apply necessary congruence and continued-fraction conditions. Two different arguments reject the final odd candidate: one uses continued fractions, and the other bounds the integer solutions of two quadratic equations of Pell type. Separately written programs reproduce the computations. The constants measure the strength of collision inequalities; they are not ranges through which Markoff numbers were checked. We identify the classical ingredients and the quantitative additions. Factorizations with increasingly comparable factors remain unresolved, so the general unicity conjecture is not proved.

Author comments: 35 pages; one figure. Source matches the 30 September 2026 manuscript deposited as Zenodo v6 on 9 November 2026: https://zenodo.org/records/24091908 (matching TeX and exact finite certificates). The constants are collision-inequality coefficients, not verified ranges; general Markoff unicity remains open. Internal “refereed” filenames do not denote external peer review. Publication chronology: earlier manuscript dated 19 August 2026, first ResearchGate posting 20 August 2026: https://doi.org/10.13140/RG.2.2.25111.33442 ; current v6 mirror recorded 12 November 2026: https://www.researchgate.net/publication/412818702_Hyperbolic_Collision_Invariants_and_Certified_Odd-Even_Tiers_toward_Markoff_Unicity . Related project chronology (earlier computational announcements, not first postings of this paper): X, 21 October 2026: https://x.com/webystar/status/2112926487091282350 ; LinkedIn, 23 October 2026: https://www.linkedin.com/posts/elleard-felix-webster-heffern-joseph-by-confirmation-2953a3104_markovnumbers-activity-7519411048115003392-4NaJ . Computational supplements, 2 November 2026: https://zenodo.org/records/23874075 . Corrected trillion-value verification/Meridians v22, 12 November 2026: https://zenodo.org/records/24167894 ; published on Hexagon 13 November 2026: https://hexagonmath.org/2611.00410 . These related computations are distinct from this paper's certified collision tiers.

A mix of human-written and AI-generated textHuman understanding: all partsDiophantine equationsFrobenius conjectureMarkoff numbersMarkov uniqueness conjecturePell equationscollision inequalitiescontinued fractionsexact finite certificateshyperbolic parametrizationunitary divisors
Nov 17, 202635 pages
PDFView work

math.NT — Number Theory

On the Diophantine equation ab+bc+ca=Nab + bc + ca = N and Euler's idoneal numbers

Contributed by Xuebin Liang

In this paper, it is proved that every natural number since 18491849 can be represented as ab+bc+caab+bc+ca with three distinct positive integers aa, bb and cc. A positive integer NN is an Euler's idoneal number if and only if it cannot be written as ab+bc+caab+bc+ca for distinct positive integers aa, bb and cc. Consequently, the number 18481848 is the largest Euler's idoneal number, which was conjectured by Euler in 1778 and Gauss in the Disquisitiones Arithmeticae.

Primarily human-written textHuman understanding: all partsAlgebraic Number Theory
Nov 17, 202647 pages
PDFView work

math.NT — Number Theory

Wieferich co-occurrence along exponential sequences: a weighted tail estimate and its application to (⋆′)

Contributed by Zeng Chunping

Let r≥2r\ge2 be a fixed integer and write be=re−a′b_e=r^e-a' for a′≤Aa'\le A coprime to rr. A recent preprint (hexagon:2610.00630, Zeng) reduces the abc conjecture, within a lattice-theoretic framework, to a single remaining hypothesis (⋆′)(\star'). We show that on the k=2k=2 slice the density-zero requirement on the exceptional set attached to (⋆′)(\star') follows from one summability condition Kr,a′<∞K_{r,a'}<\infty, where Kr,a′=∑q∈Cr,a′sqlog⁡q (1/D3(q)+1/e0(q))K_{r,a'}=\sum_{q\in\mathcal{C}_{r,a'}}s_q\log q\,(1/D_3(q)+1/e_0(q)), Cr,a′\mathcal{C}_{r,a'} is a three-filter congruence set, D3(q)=ordq3(r)D_3(q)=\mathrm{ord}_{q^3}(r), and e0(q)e_0(q) is the least solution of q3∣re−a′q^3\mid r^e-a'. The main result is unconditional: splitting the governing weight Ξr\Xi_r by the block d=ordq(r)d=\mathrm{ord}_q(r), the mass constraint ∏q2∣rd−1\prod q^2\mid r^d-1 and Stieltjes partial summation give Ξr(2)=∑sq=22log⁡q/(dqq)<∞\Xi_r^{(2)}=\sum_{s_q=2}2\log q/(d_qq)<\infty for every base rr, with explicit bound O(1+log⁡log⁡r)O(1+\log\log r); no abc, no GRH, and no count or density of the Wieferich primes Wr\mathcal{W}_r. The exact threshold of the first-moment route is Φr(E)=o(E)\Phi_r(E)=o(E), equivalently Ar(t)=∑d≤tMd/d=o(t)A_r(t)=\sum_{d\le t}M_d/d=o(t); the deep-layer condition Ξr(3)<∞\Xi_r^{(3)}<\infty sits strictly between Murata's finiteness conjecture P6aP_{6a} and the deep reciprocal sum, and the barrier analysis (one claimed radical barrier withdrawn as a bookkeeping artifact and replaced by an exact identity) shows it cannot be reached by block/radical/partial-summation methods. A Katz double-zero hypothesis (D4) yields Ξr(3)<∞\Xi_r^{(3)}<\infty and P6aP_{6a} almost surely (Borel--Cantelli); in the Haar ensemble (D4) is free with constant 11, while the diagonal embedding ι(Z)\iota(\mathbb{Z}) has measure zero, so the missing input is precisely pointwise rigidity of ι(r)\iota(r), equivalently the single-variable statement m(q)→∞m(q)\to\infty for the least nontrivial Teichmüller representative m(q)m(q) modulo q3q^3. The bad-set hypothesis (H-j) is supported by computation to 10610^6 (five hits, zero growth), while the same data correct the twin-Wieferich null hypothesis to the Teichmüller rate (q−1)/q2(q-1)/q^2. We prove (H-j) is not a GRH statement, locate it strictly between Katz's 2015 equidistribution conjecture and Murata's 1991 conjectures, record ancillary unconditional results (∑plog⁡p/(dp(p−1))<∞\sum_p\log p/(d_p(p-1))<\infty via Erdős--Murty; the exact identity Cr=SW+Sord+LTE2C_r=S_W+S_{\mathrm{ord}}+LTE_2; strict one-way comparisons), and specify what any new tool must look like (S1--S7). We prove no conjecture: the exact threshold, Ξr(3)\Xi_r^{(3)}, P6aP_{6a}, P4P_4, P1P_1 and (H-j) remain open.

Primarily human-written textHuman understanding: all partsBorel-Cantelli lemmaFermat quotientHensel liftingKatz equidistribution conjectureStieltjes partial summationTeichmuller subgroupWieferich primesabc conjecturemultiplicative ordersuper-Wieferich primes
Nov 17, 202627 pages
PDFView work

math.MP — Mathematical Physics

From Four-Dimensional Wilson Lattice Gauge Theory to Quantum Yang--Mills Theory, with a Conditional Mass-Gap Theorem

Contributed by Zeng Chunping

We present a complete formal paper compiling the Lean-verified construction of quantum Yang--Mills theory submitted on October 28, 2026, and independently validated by the Yang--Mills Shepherding Committee. Starting from the Wilson definition of lattice gauge theory on a four-dimensional torus, we give full proofs of every step that can be carried out from the definitions: the field-label algebra, the finite lattice theory, and--under an explicit, individually registered package of renormalization-group and analytic assumptions--the fixed-volume ultraviolet limit, the subsequential infinite-volume limit, the phase decomposition with clustering and full support, the Osterwalder--Schrader reconstruction satisfying the Wightman axioms, and the Yang--Mills identification: the stress tensor and its Ward identity, the trace anomaly with the verified constant 3 dim_R(g)/pi^4, the exclusion of a precise class of free-field descriptions, the one-loop running coupling, and the operator product expansion. For the mass gap we prove two logical boundaries: an explicit model satisfies all registered inputs with clustering, yet its Hamiltonian has no spectral gap; and a mass gap is equivalent to a uniform exponential rate in the clustering axiom, with the bridge from the volume-uniform transfer-matrix condition to the continuum gap proved. A uniform mass gap requires new infrared input; we develop one explicitly conditional route (semiclassical tunneling in the defining representation of SU(2)), and we prove that, with the registered input set fixed, removing the conditionality is equivalent to proving the mass gap itself. A positive mass gap is not proved.

Author comments: First version. Part I recompiles the validated Lean 4 construction of quantum Yang--Mills theory (OpenAI submission of October 28, 2026; independent validation by the Yang--Mills Shepherding Committee, release of November 10, 2026) into ordinary rigorous mathematics, expanding every proof that can be expanded from the definitions and the fifteen external inputs (RG1)-(RG3), (RG5), (UV1), (PH1)-(PH3), (AN1), (Y1)-(Y4), (B1), (C1), with (RG4) derived and (PH4) a proved spectral conclusion. Part II develops an explicitly conditional mass-gap theorem (tunneling certificate plus fiber reduction plus a one-dimensional Muckenhoupt analysis in the defining representation of SU(2)) and records a de-conditioning boundary: with the registered input set fixed, removing the conditionality is logically equivalent to proving the mass gap itself.

A mix of human-written and AI-generated textHuman understanding: all partsYang-Mills; mass gap; CQFT; LGT; OS reconstruction; RG; Wightman axioms
Nov 17, 202657 pages
PDFView work

math.FA — Functional Analysis

Orthogonal bases of symmetric unitary matrices in odd dimensions and the mixed-unitary rank of the symmetric Werner–Holevo channel

Contributed by Qiyuan Ma

Let Sym_n denote the space of complex symmetric n×n matrices with the Hilbert–Schmidt inner product. Girard, Leung, Levick, Li, Paulsen, Poon and Watrous showed that the symmetric Werner–Holevo channel on M_n has minimal mixed-unitary rank binom(n+1,2) if and only if Sym_n admits an orthogonal basis consisting of unitary matrices. They established this for even n and for n=3, and conjectured it for all odd n; Levick and Rahaman settled the case of primes n≡7 (mod 8). We introduce a family of symmetric unitaries (“chirped circulants”) for odd n and prove that, for this family, the pairwise orthogonality conditions reduce exactly to a square system of (n^2-1)/4 real equations in as many unknown phases. Using this reduction we give (i) a fully explicit basis for n=5, with all entries in an explicit degree-8 number field and an exact-arithmetic verification, and (ii) a computer-assisted proof, based on the Krawczyk operator evaluated in Arb ball arithmetic, that a symmetric unitary orthogonal basis exists for every odd n with 3≤n≤67. Consequently the conjecture of Girard et al. holds for all n≤68. We also record several obstructions to natural approaches for general odd n, and a further reduction for prime n.

Primarily AI-generated textHuman understanding: some parts
Nov 17, 20268 pages
PDFView work

math.CO — Combinatorics

The Hibi–Li (2016) conjecture is fully resolved, and other miscellanea

Contributed by Dave Goel

We prove the equality part of the Hibi–Li (2016) conjecture, its last open part: for every n-element poset P and every 2 ≤ k ≤ n−1, if the chain polytope 𝒞(P) and the order polytope 𝒪(P) have the same number of k-dimensional faces, then they are unimodularly equivalent. Its inequality part, that 𝒞(P) has at least as many k-dimensional faces as 𝒪(P), was first proved by Fourier; for edges the equality part fails, since the two polytopes always have the same number of edges. So the conjecture is now fully resolved. The second version of Fourier's preprint gives an independent proof of the equality part. Both parts follow from two theorems. First, the face numbers of a bounded marked chain-order polytope weakly increase when its chain part grows; the partition is arbitrary, and the marking may be real and degenerate. Its admissible case is also proved, independently, in the second version of Fourier's preprint. Second, for an ordinary n-element poset P whose order and chain polytopes are not unimodularly equivalent, we prove the sharp inequalities f_k(𝒞(P))−f_k(𝒪(P)) ≥ (n−3 choose k−2) for 2 ≤ k ≤ n−1. We characterize the posets for which all these bounds are attained at once, and the set of indices k at which the bound is attained. For posets whose order and chain polytopes differ by one in the number of 2-faces, we obtain a sharper bound that depends on the difference of the facet numbers. For nested chain parts C ⊆ C' in the bounded marked setting, equality of facet numbers is equivalent to affine equivalence; integral markings give lattice equivalence. We prove that regular marked posets have irredundant canonical facet descriptions for every partition, including when maximal elements are unmarked. In that unbounded setting, a switch, which adds one element to the chain part, weakly increases every face number if the element has an upper endpoint, and weakly decreases every face number otherwise. A nonzero change has its nonzero coefficients exactly in an interval of degrees that ends one below the dimension. Its quotient by (1+t)^2 for an increase, and by 1+t for a decrease, has coefficients of the sign of the change. A connected example shows that increases and decreases can cancel along a nested sequence. For bounded marked posets and several pairwise incomparable switched order elements, the alternating sum of face polynomials is a higher-order subdivision defect. It obeys a sign rule under mutual transversality, and for layered posets whose distinct lower and upper neighbourhoods form Berge-acyclic families; without these hypotheses the sign rule can fail, and the defect can have coefficients of both signs. When all these neighbourhoods have two elements, the defect is, up to sign, a power of 2+t times the chromatic polynomials of the lower and the upper neighbourhood graph, evaluated at −t; the sign rule then holds exactly when both graphs are forests. For neighbourhoods of any size, the defect of a layered poset is, up to sign and a power of 2+t, a product of two alternating sums of chromatic polynomials of hypergraphs at −t; for neighbourhoods with three or more elements, Berge-acyclicity is not necessary for the sign rule. The arguments combine fixed-height lifts and clipped markings with general transverse-subdivision results proved in the appendices.

Author comments: First version: October 23, 2026 (Zenodo record 23601976). The ancillary programs are in the Zenodo record of this paper. This is the expanded version, with further results, of the paper "The Hibi–Li (2016) conjecture is fully resolved": https://hexagonmath.org/2611.00512

A mix of human-written and AI-generated textHuman understanding: some partsHibi-Li conjecturechain polytopeface numbersorder polytopeposetsunimodular equivalence
Nov 17, 2026121 pages
PDFView work

math.CO — Combinatorics

The Hibi–Li (2016) conjecture is fully resolved

Contributed by Dave Goel

Hibi and Li conjectured that the order polytope 𝒪(P) of a finite poset P has at most as many faces of each dimension as its chain polytope 𝒞(P), and that equality in some dimension k with 1 ≤ k ≤ |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. Using two results from Fourier's preprint, we show that if an n-element poset contains X as an induced subposet, then, for every k with 2 ≤ k ≤ n − 1, its chain polytope has at least (n−3 choose k−2) more k-dimensional faces than its order polytope, and that for every n ≥ 5 some n-element poset containing X attains all these bounds at once. Since Hibi and Li showed that 𝒪(P) and 𝒞(P) are unimodularly equivalent exactly when P does not contain X, this proves the equality statement for these values of k. The second version of Fourier's preprint gives an independent proof of this statement. For k = 1, the equality statement is false: the polytopes 𝒪(X) and 𝒞(X) have the same number of edges (Hibi, Li, Sahara and Shikama) but are not unimodularly equivalent. Together with Fourier's inequality, these results settle the conjecture.

Author comments: This paper is arXiv:2610.40703 (first version: October 28, 2026). The expanded version of this paper was first posted on October 24, 2026, at https://doi.org/10.5281/zenodo.23601976 and its newest version is at https://doi.org/10.5281/zenodo.23601975 and on Hexagon at https://hexagonmath.org/2611.00513

A mix of human-written and AI-generated textHuman understanding: some partsHibi-Li conjecturechain polytopeface numbersorder polytopeposetsunimodular equivalence
Nov 17, 202617 pages
PDFView work

math-ph — Mathematical Physics

Cohesive 2-Topoi: Foundational Aspects and Nonabelian Differential Cohomology

Contributed by Yuri Ximenes Lecca, Willian Goulart Gomes Velasco

We provide a self-contained exposition of cohesive (2,1)(2,1)-topos theory, its internal synthetic geometry, and its realization of intrinsic, modal, and differential nonabelian cohomology. Starting from the (2,1)(2,1)-categorical hierarchy, Grothendieck (2,1)(2,1)-topoi, effective internal groupoids, and $\E_k$-algebra objects over (2,1)(2,1)-operads, we develop a detailed comparison between \emph{internal} and \emph{external} deloopings in (2,1)(2,1)-presheaf and Grothendieck (2,1)(2,1)-topoi. Specifically, we formalize external delooping by transporting the canonical delooping transformation along the group, discrete, and groupoid Yoneda 2-embeddings, proving that it is 2-naturally equivalent to the internal delooping in any Grothendieck (2,1)(2,1)-topos. We then establish the 2-truncated Dunn--Lurie additivity, Baez--Dolan stabilization, and Eckmann--Hilton characterization of kk-tuply groupal objects as groupal $\E_k$-algebras stabilizing at k=3k=3, and clarify the truncation behavior of iterated deloopings BHkGB_{\mathbf{H}}^k G. Finally, we review Lawvere--Schreiber cohesive modalities, construct abstract principal kk-bundles, the flat de Rham modality $\flat_{\dR}$, universal Maurer--Cartan forms θk\theta_k, connection moduli stacks $B^k_{\con}G$, and sharp-adjusted internal connection moduli $G\Conn^k(X)$. The paper culminates in synthetic differential cohomology, where we discuss extensions of the Maurer--Cartan morphism to non-groupal coefficients via $\int_{\dR} \dashv \flat_{\dR}$, and internal classifying structures.

Primarily human-written textHuman understanding: all parts$\E_k$-algebrasCohesive 2-topoiGrothendieck $(2,1)$-topoiMaurer--Cartan formsalgebra-valued gauge theoriescohesive modalitiesinternal and external deloopingsnonabelian differential cohomology
Nov 17, 202640 pages
PDFView work

math.RA — Rings and Algebras

Improving Razmyslov's (1974) bound: the Nagata–Higman index is at most n² − 2 for n ≥ 3, and other miscellanea

Contributed by Dave Goel

Let d(n)d(n) be the least mm such that every associative algebra over a field of characteristic zero satisfying xn=0x^n=0 also satisfies x1x2⋯xm=0x_1x_2\cdots x_m=0 (the Nagata–Higman index). Razmyslov proved d(n)≤n2d(n)\le n^2 in 1974, and the Kuzmin (1975) conjecture is d(n)=n(n+1)/2d(n)=n(n+1)/2. We prove d(n)≤n2−1d(n)\le n^2-1 for n≥2n\ge2 and d(n)≤n2−2d(n)\le n^2-2 for n≥3n\ge3; in particular d(6)≤34d(6)\le34. The proof uses Formanek's criterion in cycle form and studies, one isotypic block of SNS_N at a time, the span of the NN-cycles inside the block: a block with two corners meets this span exactly when its shape is a self-conjugate hook, and for every k≥2k\ge2 the block of (k+1,2,1k−1)(k+1,2,1^{k-1}), which has three corners, meets it too. The next step, d(n)≤n2−3d(n)\le n^2-3, reduces to one statement about the block of (nn−2,n−1,n−2)(n^{n-2},n-1,n-2); for every n≥4n\ge4 we prove the part of it that concerns the constituents with at most two rows of length 33, and for n=5n=5 the whole statement (computer-assisted). Through the hook theorem the question becomes one about the polynomial identities of Kemer's superalgebras Mk,lM_{k,l} over the Grassmann algebra: lower bounds for their PI degrees give upper bounds for d(n)d(n), and the lower half of the Samoilov (2010) conjecture on these PI degrees would give d(n)<89n2d(n)<\frac89n^2 for n≥3n\ge3.

Author comments: First version: October 27, 2026. This is the expanded version of the paper "Improving Razmyslov's (1974) bound: the Nagata–Higman index is at most n² − 2 for n ≥ 3", which is at https://hexagonmath.org/2611.00526 on Hexagon. The ancillary files of this paper (programs and logs) are at https://doi.org/10.5281/zenodo.23624550 on Zenodo.

A mix of human-written and AI-generated textHuman understanding: some partsKuzmin's conjectureNagata-Higman theoremnil algebrasnilpotency indexpolynomial identitiessymmetric group
Nov 17, 2026211 pages
PDFView work

math.RA — Rings and Algebras

The relations of a Lie subalgebra considered by Tao: a complete presentation

Contributed by Qiyuan Ma

Let g be the Lie algebra over a field of characteristic 0 generated by D_i, E_i (i ≥ 1), subject to [D_i,D_j] = [E_i,E_j] = 0 and [D_i,E_j] = [D_j,E_i]. Let h be the subalgebra generated by the elements [D_i,E_j] and E_i. On MathOverflow (question 464808), T. Tao asked whether the obvious relations [D_i,E_j] = [D_j,E_i], [E_i,E_j] = 0 and [E_i,[D_j,E_k]] = [E_k,[D_j,E_i]] generate all relations of h. We prove that they do. The proof has four steps: (1) a polarisation argument reduces the problem to multilinear components; (2) we give a presentation of the ideal generated by the E_i; (3) an excess filtration, whose associated graded algebra we compute explicitly, reduces injectivity to an equality of dimensions; (4) that equality follows from the Hilbert series of g, which is given by Eulerian numbers. The Hilbert series in step 4 is equivalent to known results. The multilinear components of the enveloping algebra of the ideal generated by the E_i coincide with the homology of the Losev–Manin spaces: their Betti numbers are due to Losev and Manin (Michigan Math. J. 48 (2000)), and the algebra structure was presented by Dotsenko, Shadrin and Tamaroff (Comm. Math. Phys. 399 (2023)). The paper explains this relation precisely. It also gives an independent, self-contained derivation, using a PBW basis of U(g) obtained from Bergman’s diamond lemma and a sign-reversing involution. By-products: h is a semidirect product of a free Lie algebra and an abelian Lie algebra; the elements [D_i,E_j] (i ≤ j) freely generate a free Lie subalgebra of g; the multilinear components of g have dimensions given by Eulerian numbers (equivalent to known results). We were unable to find the affirmative answer to Tao's question or the freeness statement in the literature. Independent computer checks are included in the separate code archive.

Author comments: 17 pages, version 1.1. This work answers MathOverflow question 464808.

Primarily AI-generated textHuman understanding: some partsGröbner basisHilbert series / Eulerian numbersLie algebrasLosev–Manin spacesMathOverflowPBW basisdiamond lemmafree Lie algebraspresentations by generators and relationsuniversal enveloping algebra
Nov 17, 202617 pages
PDFView work

math.CO — Combinatorics

Wills' (1981) conjecture holds if and only if the dimension is at most 4, and other miscellanea

Contributed by Dave Goel

Let PP be a lattice polytope in Rn\mathbb R^n that is symmetric about the origin and has the origin as its only interior lattice point. In 1981 Wills conjectured that each coefficient gi(P)g_i(P) of the Ehrhart polynomial of PP is at most the corresponding coefficient of the cube [−1,1]n[-1,1]^n, that is, gi(P)≤2i(ni)g_i(P)\le 2^i\binom ni. Wills proved this for n≤3n\le3, and Henze showed in 2014 that it fails for every n≥7n\ge7; the dimensions 44, 55 and 66 remained open. We settle them. In dimension 44 the conjecture holds; the new bounds are g1(P)≤8g_1(P)\le8 and g2(P)≤24g_2(P)\le24, and, up to unimodular equivalence, g1(P)=8g_1(P)=8 only for the cube, one explicit 2424-cell and one explicit cross-polytope. In dimension 55 we give a polytope with 5050 vertices in [−1,1]5[-1,1]^5 whose linear Ehrhart coefficient is 101/10>10101/10>10; products with segments then give counterexamples in every dimension n≥6n\ge6. Hence Wills' conjecture holds in Rn\mathbb R^n if and only if n≤4n\le4. The proof in dimension 44 combines Ehrhart reciprocity, the lower bound theorem for triangulated 33-spheres and an exhaustive computer enumeration of about three million sets of lattice points, listed up to symmetry. This expanded version also gives a second proof of the case in which the integer functionals bounded by 11 on PP do not span R4\mathbb R^4, determines the second largest value of g2g_2 in dimension 44, discusses a width-two approach, describes the computer search for the counterexample, and describes all computations used in the proofs.

Author comments: First version: November 3, 2026. This is the expanded version of the paper "Wills' (1981) conjecture holds if and only if the dimension is at most 4", which is at https://hexagonmath.org/2611.00540 on Hexagon.

A mix of human-written and AI-generated textHuman understanding: some partsEhrhart polynomialWills' conjecturecentrally symmetric polytopeinterior lattice pointlattice polytope
Nov 16, 202626 pages
PDFView work

math.CO — Combinatorics

Wills' (1981) conjecture holds if and only if the dimension is at most 4

Contributed by Dave Goel

Let PP be a lattice polytope in Rn\mathbb R^n that is symmetric about the origin and has the origin as its only interior lattice point. In 1981 Wills conjectured that each coefficient gi(P)g_i(P) of the Ehrhart polynomial of PP is at most the corresponding coefficient of the cube [−1,1]n[-1,1]^n, that is, gi(P)≤2i(ni)g_i(P)\le 2^i\binom ni. Wills proved this for n≤3n\le3, and Henze showed in 2014 that it fails for every n≥7n\ge7; the dimensions 44, 55 and 66 remained open. We settle them. In dimension 44 the conjecture holds; the new bounds are g1(P)≤8g_1(P)\le8 and g2(P)≤24g_2(P)\le24, and, up to unimodular equivalence, g1(P)=8g_1(P)=8 only for the cube, one explicit 2424-cell and one explicit cross-polytope. In dimension 55 we give a polytope with 5050 vertices in [−1,1]5[-1,1]^5 whose linear Ehrhart coefficient is 101/10>10101/10>10; products with segments then give counterexamples in every dimension n≥6n\ge6. Hence Wills' conjecture holds in Rn\mathbb R^n if and only if n≤4n\le4. The proof in dimension 44 combines Ehrhart reciprocity, the lower bound theorem for triangulated 33-spheres and an exhaustive computer enumeration of about three million sets of lattice points, listed up to symmetry.

Author comments: First version: November 3, 2026. The expanded version of this paper, "Wills' (1981) conjecture holds if and only if the dimension is at most 4, and other miscellanea", is at https://doi.org/10.5281/zenodo.23955454 on Zenodo.

A mix of human-written and AI-generated textHuman understanding: some partsEhrhart polynomialWills' conjecturecentrally symmetric polytopeinterior lattice pointlattice polytope
Nov 16, 202613 pages
PDFView work

math.MP — Mathematical Physics

Phase Transition and Spontaneous Translation Symmetry Breaking in Negative Temperature Point Vortex Ensembles

Contributed by Francesco Grotto, Marco Romito

We establish a canonical phase transition and spontaneous translation symmetry breaking for balanced two-species point vortices on the square torus. A sharp inequality identifies the transition at β=4π2\beta=4\pi^2, below collapse at β=16π\beta=16\pi. In the intervening regime, all global minimizers form one translation orbit of an explicit dipole. We prove a large deviation principle below collapse and identify the uniform random phase, phase-centered Gaussian fluctuations, and precise partition asymptotics in the ordered regime. For general prescribed bounded strengths, we distinguish global ordering, local instability, and collapse.

Primarily AI-generated textHuman understanding: some partsNegative TemperaturePhase TransitionPoint vortexSpontaneous Symmetry Breaking
Nov 16, 202665 pages
PDFView work

math.CO — Combinatorics

The Zarankiewicz number z(13,15;3,4)

Contributed by Will Howe, Grant Huffman, Jensen Kohlmeyer, Liam Kruer

We establish z(13,15;3,4)=114z(13,15;3,4)=114: a 13×1513\times15 binary matrix with 115115 ones necessarily contains an all-one submatrix on three rows and four columns. The lower bound is attained by a previously published construction of Rosero. For the upper bound, we exclude dense eight-row column neighborhoods by a finite classification and a complete extension check. The resulting local inequality reduces a hypothetical counterexample to 6,0196{,}019 pairs of degree sequences. Each pair is excluded by an exact integer certificate for a system of neighborhood-type equations. The proof is computer-assisted; its mathematical reductions, finite search domains, classification representatives, and certificate conditions are given explicitly.

Author comments: Added a public, version-specific archive containing all 15 checked SAT refutations and CNF inputs, all 82 five-row extension inputs and outputs, and all 6,019 exact integer dual certificates, with generators and checking instructions. The manuscript now links release v1 and commit 10385847932cd2b61e25cb4e7d039633d108455b. The computational material was regenerated and independently checked on November 16, 2026.

Primarily AI-generated textHuman understanding: all partsZarankiewicz numbercomputer-assisted proofextremal bipartite graphforbidden submatrix
Nov 16, 202612 pages
PDFView work

math.NT — Number Theory

Consecutive primes at distances 123,662 and 125,318

Contributed by Will Howe, Grant Huffman, Jensen Kohlmeyer, Liam Kruer

We exhibit consecutive-prime gaps of exactly 123 662123\,662 and 125 318125\,318, with endpoints of 8 4168\,416 and 6 4546\,454 decimal digits, respectively. Covering congruences certify most interior integers as composite. For the second gap, the remaining integers are certified by proper divisors or Fermat witnesses. Elliptic-curve certificates prove all four endpoints prime. We give the elementary arguments that turn these finite arithmetic checks into a proof. The proof is computer-assisted.

Author comments: 4 pages. The supplementary arithmetic appendix supplies the exact covering tables, additional compositeness witnesses, four endpoint primality chains, and elementary order certificates for the terminal primes.

Primarily AI-generated textHuman understanding: all parts
Nov 16, 20264 pages
PDFView work

math.AG — Algebraic Geometry

Essential dimension of geometric local systems

Contributed by Mark Kisin

Let f : X → B be a smooth projective family over a smooth complex variety, and let H = (Rmf∗Z)/tors. We prove that there is an integer N such that, whenever pn ≥N, the essential dimension at p of H/pn is at least the dimension of its period image. This proves the Brosnan–Fakhruddin conjecture for these local systems. The proof compares the geometric Sen operator with the Higgs field. Integral Hodge–Tate theory gives a bound uniform in large primes. For each remaining prime, a bounded rational comparison at one geometric point suffices when n is large.

A mix of human-written and AI-generated textHuman understanding: all partsEssential dimension
Nov 16, 202614 pages
PDFView work

math.DG — Differential Geometry

Smooth isometric bending of a closed sphere

Contributed by Boyang Hu

We construct a smooth embedded sphere Σ⊂R3\Sigma\subset\mathbb R^3 together with a smooth one-parameter family of isometric embeddings of Σ\Sigma whose images are not all congruent. The family is periodic; an inner disk and a complementary cap stay fixed pointwise; and the area, the enclosed volume and the total mean curvature do not change. This gives a flexible closed surface in the smooth category. The mechanism is elementary. A disk whose metric is rotationally symmetric, but whose embedding is not, can be turned inside a fixed shell provided it is exactly planar near its rim. The whole difficulty is to produce such a disk. This reduces to a scalar equation of Monge–Ampère type on an annulus, the Darboux equation, for a height function which must be nonradial and yet constant to infinite order at both boundary circles. We solve this equation by a Nash–Moser argument around an infinitesimal bending of a radial profile. The profile is forced to change the sign of its curvature infinitely often near each boundary circle, so the linearized equation changes type on infinitely many circles accumulating at both ends. We build the profile from explicitly solvable one-dimensional models, describe its accumulating ends by an exact three-term recurrence, and invert the linearized operator on the whole annulus. Two modifications of the linearization, each proportional to an unsolved equation, keep the type-change circles and the characteristic return maps fixed during the iteration, so that all linearized operators act on one domain.

Author comments: 119 pages. The main theorem (Theorem A) has been formalized in Lean 4 with Mathlib: https://github.com/Rogerhu12/continuous-flexibility. The folder anc/ contains four Python scripts which check explicit identities of the text, and the script which draws Figure 2.

Primarily AI-generated textHuman understanding: some partsDarboux equationNash–Moser iterationflexible surfacesisometric deformations
Nov 16, 2026119 pages
PDFView work

math.CA — Classical Analysis and ODEs

Even Asymptotic Expansions and Uniform Hard-Edge Approximations for Wigner 3j3j and 6j6j Symbols at Fixed Rank

Contributed by Antony Lewis

We study Wigner 3j3j symbols with two large and one small angular momentum, and the 6j6j symbols with one small entry (Edmonds' limit), both at fixed rank: the small labels stay fixed while the large ones grow. Their classical limits are Wigner rotation matrix elements. Combining a Regge symmetry with the Weyl shift j↦j+12j\mapsto j+\tfrac12 singles out midpoint variables: for 3j3j, the large scale N=j−+j++1N=j_-+j_++1 (j∓j_\mp the two large angular momenta) and x=2μ/Nx=2\mu/N (2μ2\mu the difference of the two large magnetic indices), and, for 6j6j, large lengths A,B,CA,B,C shifted by offsets that depend on the small labels. Among constant shifts of the large labels and magnetic coordinate, and up to rank-independent gauge factors, these are the only centerings that remove the generic first-order term of the exact recurrences. In these variables the normalized symbols are values of analytic functions of the inverse square of the large scale, for integer and half-integer small labels alike (including spin–orbit coefficients): the expansions contain only even powers and converge. The first correction is a closed formula in one Wigner dd function and its derivative, improving the error from second to fourth order in the inverse large scale; all higher corrections are constructed as local differential operators. The exact finite problems are Hahn and Racah spectral systems with undeformed Wigner spectra. Near the ends of the physical lattice (the hard edges), an explicit product of Pochhammer factors carries the difference between the discrete and continuum weights; extracting it gives expansions that are uniform on the whole lattice, including the 6j6j corner where two large lengths nearly coincide (A≈BA\approx B). We also show that centered Hahn and Racah polynomials are even deformations of Jacobi polynomials.

Author comments: 55 pages, 1 figure.

Primarily AI-generated textHuman understanding: no partsHahn polynomialsRacah polynomialsRegge symmetryWigner 3j symbolsWigner 6j symbols
Nov 16, 202655 pages
PDFView work

math.GR — Group Theory

Which groups are automorphism groups of racks?

Contributed by G. Militaru

Every group is the automorphism group of a graph, of a magma, and even of a set-theoretic solution of the Yang-Baxter equation. We show that this fails for racks and quandles: the quaternion group of order eight is not the automorphism group of any rack, and it is the smallest group with this property. More generally, generalized quaternion groups and Hamiltonian groups are never automorphism groups of racks, and a dicyclic group is the automorphism group of a rack if and only if its order is not divisible by eight. On the positive side, dihedral, alternating and all finite abelian groups are automorphism groups of finite quandles.

Author comments: 19 pages + references. Comments are welcome.

A mix of human-written and AI-generated textHuman understanding: all partsHamiltonian groupautomorphism groupdicyclic groupquandlerackrepresentation problemset-theoretic Yang--Baxter equationuniversal categories
Nov 16, 202622 pages
PDFView work

math.NT — Number Theory

Smooth Numbers in Arithmetic Progressions with Uniformity in the Residue Class

Contributed by Sarvagya Jain

We improve the level of distribution of smooth numbers in arithmetic progressions to x5/8−ϵx^{5/8-\epsilon}, with uniformity in the residue class. In particular, for every ϵ,A>0\epsilon,A>0 and integer k≥1k\geq1, there exist δ,C>0\delta,C>0 such that $$ \sum_{\substack{q\leq x^{5/8-\epsilon}\\(q,a_1a_2)=1}}\tau_k(q)\sup_{0\leq z\leq x} \Biggl|\Psi(z,y;a_1\overline{a_2},q) -\frac{1}{\phi(q)}\Psi_q(z;y)\Biggr| \ll_{\epsilon,A,k}\frac{\Psi(x;y)}{(\log x)^A}, $$ uniformly for x≥2x\geq2, (log⁡x)C≤y≤x1/C(\log x)^C\leq y\leq x^{1/C}, and a1,a2∈Z∖{0}a_1,a_2\in\Z\setminus\{0\} satisfying $$ |a_1|\leq x^{1+\delta},\qquad |a_2|\leq x^\delta,\qquad (a_1,a_2)=1. $$ Here Ψ(z;y)\Psi(z;y) counts the yy-smooth integers up to zz, Ψq(z;y)\Psi_q(z;y) counts those coprime to qq, and Ψ(z,y;a,q)\Psi(z,y;a,q) counts those congruent to aa modulo qq; τk\tau_k is the kk-fold divisor function. The proof combines dispersion with second and fourth spectral moments.

Author comments: This work is an extension to author's work submitted to arxiv concerning an improvement to a prior work of Balog concerning writing positive integers as a sum of two smooth numbers.

Primarily AI-generated textHuman understanding: some parts
Nov 16, 202632 pages
PDFView work

math.NT — Number Theory

Three Properties of Markov–Hurwitz Equations

Contributed by Qiyue Tang

We prove Laurent positivity for Markov–Hurwitz mutations with independent exchange parameters, density-one uniqueness by the largest coordinate for positive integral solutions of the symmetric generalized equation, and labelled Vieta connectivity of the pure equation on points with at most one zero coordinate over sufficiently large prime fields.

Author comments: v2(2026/11/17):A theorem from version 1 is generalized. The ideas behind this work came primarily from the author and were developed through discussions with ChatGPT. I am posting this manuscript to share the mathematical results themselves. In doing so, I do not seek to assert ownership of the results or claim priority. I welcome anyone interested in checking the arguments or preparing a finished account, and I believe the eventual scholarly credit should go to those who undertake that work.

Primarily AI-generated textHuman understanding: some parts
Nov 16, 202645 pages
PDFView work

math.OA — Operator Algebras

Quantitative similarities for amenable algebras of two-by-two matrix-valued functions

Contributed by Cédric ARHANCET

We study norm-closed subalgebras AA of C(X,M2)\mathrm{C}(X,\mathrm{M}_2) containing the scalar functions, where XX is a compact metrizable space. We characterize amenability of AA by semisimplicity of its fibres and a uniform bound on the norms of the elements 2p−I22p-\mathrm{I}_2 associated with their primitive idempotents. This bound is the exact similarity constant in the faithful representation given by the direct sum of all evaluations. In particular, an amenable algebra of this form admits a similarity to a nuclear C∗\mathrm{C}^*-algebra with cost at most 2AM⁡(A)−1\sqrt{2\operatorname{AM}(A)-1}. The estimate is sharp already for a two-dimensional algebra in M2\mathrm{M}_2. The similarity is constructed from positive matrix-valued functions which are continuous away from the scalar fibres. Although these functions need not be continuous everywhere, the induced algebra isomorphism preserves continuous sections. We also prove the corresponding results for locally trivial bundles of M2\mathrm{M}_2-algebras, and give a noncommutative example for which no continuous matrix-valued similarity is available.

A mix of human-written and AI-generated textHuman understanding: all partsamenable operator algebrascontinuous sectionsmatrix-valued functionsnuclear $\C^*$-algebrasprimitive idempotentssimilarity
Nov 16, 202612 pages
PDFView work

math.RA — Rings and Algebras

Improving Razmyslov's (1974) bound: the Nagata–Higman index is at most n² − 2 for n ≥ 3

Contributed by Dave Goel

Let d(n)d(n) be the least mm such that every associative (not necessarily unital) algebra over a field of characteristic zero satisfying the identity xn=0x^n=0 also satisfies x1x2⋯xm=0x_1x_2\cdots x_m=0; this is the Nagata–Higman index. Razmyslov proved d(n)≤n2d(n)\le n^2 in 1974; Kuzmin proved d(n)≥n(n+1)/2d(n)\ge n(n+1)/2 and conjectured equality; this is the Kuzmin (1975) conjecture. We prove d(n)≤n2−2d(n)\le n^2-2 for n≥3n\ge3 (and, by a simpler case of the same method, d(n)≤n2−1d(n)\le n^2-1 for n≥2n\ge2); in particular d(6)≤34d(6)\le34. To our knowledge, n=6n=6 is the first case in which the new bound improves the known bounds, and this is the first improvement of Razmyslov's bound for general nn. For m=n2−1m=n^2-1 and m=n2−2m=n^2-2, Formanek's criterion reduces d(n)≤md(n)\le m to the statement that no nonzero linear combination of (m+1)(m+1)-cycles lies in one particular isotypic block of the group algebra of Sm+1S_{m+1}. For m=n2−1m=n^2-1 this statement follows from the branching rule; for m=n2−2m=n^2-2 we prove it using the Jucys–Murphy elements and a theorem of Frobenius on the characters of the alternating group. These bounds have also been proved in the Lean proof assistant.

Author comments: First version: October 28, 2026. The expanded version of this paper, "Improving Razmyslov's (1974) bound: the Nagata–Higman index is at most n² − 2 for n ≥ 3, and other miscellanea", is at https://doi.org/10.5281/zenodo.23624549 on Zenodo.

A mix of human-written and AI-generated textHuman understanding: some partsKuzmin's conjectureNagata-Higman theoremnil algebrasnilpotency indexpolynomial identitiessymmetric group
Nov 16, 202614 pages
PDFView work

math.CO — Combinatorics

The Lee (2026) conjecture on the diameters of order and chain polytopes fails in both directions, and other miscellanea

Contributed by Dave Goel

For a finite poset PP, Stanley's order polytope O(P)\mathcal O(P) and chain polytope C(P)\mathcal C(P) have the same number of vertices and, by a theorem of Hibi, Li, Sahara and Shikama, the same number of edges. J. Lee recently conjectured that for a connected poset the diameters of their graphs differ by at most 11. We show that this fails in both directions. For every k≥1k\ge1 we describe a connected poset BkB_k with 3k3k elements on three consecutive antidiagonals of the grid Z2\mathbb Z^2, for which the graph of O(Bk)\mathcal O(B_k) has diameter exactly kk, while the graph of C(Bk)\mathcal C(B_k) has diameter exactly ⌊2k/3⌋+1\lfloor2k/3\rfloor+1. Their difference is therefore ⌈k/3⌉−1\lceil k/3\rceil-1, and their ratio tends to 3/23/2. Hence no constant bounds diam⁡O(P)−diam⁡C(P)\operatorname{diam}\mathcal O(P)-\operatorname{diam}\mathcal C(P) on connected posets. In the other direction, for every n≥4n\ge4 we construct a connected poset PnP_n with 2n2+5n+12n^2+5n+1 elements such that diam⁡O(Pn)=2\operatorname{diam}\mathcal O(P_n)=2 and diam⁡C(Pn)=4\operatorname{diam}\mathcal C(P_n)=4. We also show that diam⁡O(P)≤2\operatorname{diam}\mathcal O(P)\le2 implies diam⁡C(P)≤4\operatorname{diam}\mathcal C(P)\le4 for every finite poset PP, so a difference of 22 is the largest possible at order diameter 22.

Author comments: First version: October 23, 2026. This is the expanded version of the paper "Diameters of the order and chain polytopes of a connected poset", which is at https://hexagonmath.org/2611.00523 on Hexagon.

A mix of human-written and AI-generated textHuman understanding: some partsLee conjecturechain polytopediameterfinite posetgraph of a polytopeorder polytope
Nov 16, 202631 pages
PDFView work