Pith. sign in

REVIEW 2 major objections 4 minor 28 references

Trapezodial property of the generalized Alexander polynomial

T0 review · 2 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proves that the Alexander polynomial of a special alternating link is a matroid generating function whose coefficients are log-concave for every bipartite graph, and trapezoidal for totally positive matrices, giving a new proof…

desk verdict Solid matroidal framework with new non-planar and totally positive results, but the main log-concavity proof has two compressed steps that need referee attention. read the letter →

arxiv 2504.20967 v1 pith:HBKLJUD2 submitted 2025-04-29 math.CO math.GT

Tamás Kálmán , Karola Mészáros , Alexander Postnikov This is my paper · ORCID
classification math.COmath.GT MSC 05B3505A2057K14
keywords AlexanderpolynomialFoxtrapezoidalconjecturespecialalternatinglinkslog-concavityorientedmatroidsexternalsemi-activitygeneralizedpermutahedratotallypositivematrices
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Fox's 1962 conjecture predicts that the coefficient sequence of the Alexander polynomial of any alternating link, after removing the alternating signs, is trapezoidal. That conjecture remains open in general, but this paper gives a unified matroidal explanation for special alternating links: their Alexander polynomial is a generating function f_A(t) that counts bases of an oriented matroid by a statistic called external semi-activity. The authors prove that the coefficients of f_A are log-concave whenever A arises from a bipartite graph, thereby dropping the planarity assumption used in earlier work, and prove a distinct coefficient property called box-positivity when A is totally positive. As a corollary, they obtain a new proof that special alternating links satisfy Fox's conjecture.

What carries the argument

The main object is the polynomial f_A(t) = \sum_B $t^{{\mathrm{ext}}$_\rho(B)} \operatorname{Vol}(\Pi_B), which sums over bases of the oriented matroid of a flat matrix A and weights each base by its external semi-activity; it is independent of the auxiliary generic vector \rho and is palindromic. The argument then converts f_A into a refined count of integer lattice points of a trimmed zonotope: each tile of a zonotopal tiling contributes one lattice point on a prescribed level, and in the bipartite-graph case the trimmed zonotope is a generalized permutahedron. That identification lets the theory of Lorentzian polynomials show that the level counts, hence the coefficients of f_A, form a log-concave sequence. In the totally positive case, the same polynomial is expanded explicitly as a positive combination of products of q-numbers [m_1]_q \cdots [m_d]_q, which gives trapezoidality directly.

What would settle it

Verify Theorem 7.5 on a non-planar bipartite graph such as $K_{3,3}$: compute f_{M_{\vec G}}(t) by summing $t^{\mathrm{ext}(B)}$ over all spanning trees, or equivalently count integer points on parallel levels of the trimmed zonotope, and test whether the resulting coefficient sequence is log-concave. A single non-log-concave example would refute the theorem; if many examples pass, the remaining risk is to compare the hypotheses of the cited result behind Theorem 7.8 with the claim it is used to prove.

Watch

Extended reading notes

Core claim

The central claim is that the Alexander polynomial of a special alternating link is not an isolated knot invariant but a special case of the polynomial f_A(t), defined for any flat matrix A as f_A(t) = \sum_B $t^{{\mathrm{ext}}$_\rho(B)} \operatorname{Vol}(\Pi_B), where the sum runs over bases of the oriented matroid of A and \mathrm{ext}_\rho(B) is the external semi-activity of a base. The paper proves that for a special alternating link L coming from a plane bipartite graph G, the polynomial \Delta_L(-t) equals f_{M_{\vec G}}(t), where the oriented matroid is both graphic and cographic. It then shows the coefficients of f_{M_{\vec G}} are log-concave for every connected bipartite graph G with edges oriented from one side of the bipartition to the other, not only for planar graphs. For totally positive matrices with last row all 1s, it proves the stronger structural property that f_A(q) is a positive linear combination of products of q-numbers, which implies the coefficients form a trapezoidal sequence. Together these results reprove the log-concavity, and hence trapezoidality, of the Alexander polynomial for special alternating links, and motivate the question whether f_A(t) is trapezoidal for every flat matrix A.

Load-bearing premise

The load-bearing premise is that a single short citation really covers the full slicing claim: the counts of lattice points on the parallel slices of the graph's trimmed zonotope form a log-concave sequence for every bipartite graph; if the cited result secretly assumes planarity or the special-link setting, the general theorem loses its support.

Editorial extensions

If this is right

  • For every special alternating link L, the coefficients of \Delta_L(-t) are log-concave with no internal zeros and hence trapezoidal, giving a new proof of Fox's conjecture for that class.
  • Log-concavity holds for the entire family f_{M_{\vec G}}(t) where G is any connected bipartite graph, planar or not, so the phenomenon is purely matroidal rather than topological.
  • For any Eulerian digraph D, the Murasugi-Stoimenow polynomial P_D(t) equals f_{M^*_D}(t), placing the Alexander polynomial identity inside a cographic matroid framework.
  • For any flat totally positive matrix with last row all 1s, f_A(q) is a positive combination of q-number products, hence trapezoidal and palindromic, though not necessarily log-concave.
  • The results motivate the paper's broad question whether f_A(t) is trapezoidal for every flat matrix A.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If Theorem B survives direct tests on non-planar bipartite graphs such as $K_{3,3}$, the log-concavity is genuinely independent of planarity; one could then ask whether stronger coefficient properties, such as ultra-log-concavity, also hold for all flat graphic matroids.
  • The identity \Delta_L(-t) = f_A(t) suggests a possible route toward Fox's conjecture for arbitrary alternating links: find, for every alternating link, a flat matrix whose f_A reproduces its Alexander polynomial, a task the paper leaves open.
  • The explicit q-number expansion for totally positive matrices may transfer to flag matroids or valuated matroids, where a similar positive expansion would prove trapezoidality without relying on log-concavity.
  • The contrast between log-concavity for graphic-bipartite arrangements and box-positivity for totally positive arrangements hints at a hierarchy of coefficient properties controlled by the geometry of the associated zonotope and its trimmed polytope.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper studies the family of polynomials f_A(t) introduced by Li and Postnikov for flat vector configurations. The authors prove that the Alexander polynomial of a special alternating link, up to the sign convention, equals f_A(t) for a flat matrix whose oriented matroid is both graphic and cographic (Theorem A, restated as Theorem 5.5). They then prove that for a connected bipartite graph with the standard bipartite orientation, the coefficients of f_A are log-concave (Theorem B, Theorem 7.5), and deduce as Corollary 7.11 the known log-concavity and trapezoidal property for special alternating links, advertised as a new proof of Fox's conjecture in that case. In addition, for flat totally positive matrices they prove a stronger box-positivity property (Theorem C, Corollary 8.9). The proof of Theorem B proceeds by showing that the coefficients of f_A count integer points of a generalized permutahedron on parallel hyperplanes, then invoking Lorentzian polynomial theory.

Significance. If the proof is fully substantiated, the paper gives a clean matroidal explanation for the log-concavity observed in the Alexander polynomials of special alternating links: the polynomial is a generating function for external semi-activity on graphic and cographic matroids, and the polytopal model allows an application of Lorentzian polynomials. The removal of the planarity hypothesis in Theorem B is a genuine generalization of the recent result of Hafner, Mészáros, and Vidinas. The totally positive case introduces box-positivity, a stronger and apparently new property, with an explicit closed formula in Theorem 8.8. The paper is written with concrete, checkable matrix constructions and worked examples (Examples 2.3 and 8.10), which is a strength. However, the central chain from the polytopal model to log-concavity depends on two compressed steps, one of which cites an unstated corollary of an external paper; these need to be made fully explicit before the advertised new proof can be considered established.

major comments (2)
  1. [§7.3, Theorem 7.8] Theorem 7.8 is the unique bridge from the polytopal model to log-concavity, but its proof is a single sentence: 'The statement follows from Theorem 7.6, Lemma 7.7 and [HMV24a, Corollary 3.8].' Since [HMV24a, Corollary 3.8] is not stated, the reader cannot verify that its hypotheses hold for every generalized permutahedron appearing in Corollary 7.9, nor that the specialization argument produces the asserted homogeneous polynomial. Moreover, if that corollary already contains the log-concavity result for special alternating links asserted in Corollary 7.11, then the advertised 'new proof' is circular. Please state [HMV24a, Corollary 3.8] explicitly, verify its hypotheses in the present setting, and clarify the logical relationship between it and Corollary 7.11.
  2. [§6.2, Theorem 6.5] The proof of Theorem 6.5 is compressed to the assertion that in a dissection of a polytope into top-dimensional pieces, the set of integer points generating the l-trimming is the union of the corresponding sets for the pieces. This is the step that turns the zonotopal tiling of Theorem 6.1 into an exact coefficient count, yet it is justified only by 'readily proves.' In particular, the convex hull in (11) could in principle contain integer points that are not among the generating vertices, and the interaction of the trimming operation with the tiling along common faces requires a precise argument. Please give a full proof of the equality of these integer-point counts in the specific case of flat unimodular zonotopes with an m-admissible vector l.
minor comments (4)
  1. [Title and running header] The word 'Trapezodial' in the title and running header appears to be a typo for 'Trapezoidal'.
  2. [§8.2, Lemma 8.3] The proof of Lemma 8.3 says 'We leave it as an exercise for the reader to calculate all minors of J'; since the positivity of all maximal minors is load-bearing for Theorem 8.8, a sentence or two outlining the Cauchy–Binet computation would improve readability.
  3. [§7.2, Lemma 7.3] The notation Δ_{[m+n]} and Δ^j_{[m+n]} is introduced only in the proof of Lemma 7.3; a brief definition at first use would help.
  4. [§7.3, Corollary 7.9] The phrase 'When homogenized' is ambiguous about the homogenization variable; specifying the homogenization explicitly would make the application of Theorem 7.8 easier to check.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the main chain is an independent application of Lorentzian-polynomial and zonotope machinery, with the HMV24a citation in Theorem 7.8 being a completeness gap rather than a demonstrated reduction to the target result.

full rationale

The paper's derivation chain is not circular in the sense of the rubric. The polynomial f_A is defined independently via external semi-activity and volumes of base parallelepipeds (Definition 3.3, from [LP13]); Theorem 5.5 identifies the Alexander polynomial of a special alternating link with f_A using the Murasugi–Stoimenow theorem and the graphic/cographic duality; Theorem 6.5 is a polytopal counting statement about trimmed zonotopes; and Theorem 7.5 derives log-concavity of the coefficients of f_{M_G} from the Lorentzian property of integer-point transforms of generalized permutahedra (Theorem 7.6, [BH20]) together with Proposition 7.10. None of these steps defines its conclusion in terms of the conclusion, and no parameter is fitted to a subset of data and then renamed as a prediction. The one genuinely load-bearing same-author citation is in the one-sentence proof of Theorem 7.8, which invokes [HMV24a, Corollary 3.8] without stating it; since [HMV24a] is the paper whose Theorem 1.2 is exactly the Alexander log-concavity result reproved as Corollary 7.11, a reader cannot fully verify that the cited corollary does not already contain the target conclusion. However, this is a citation/completeness gap, not an exhibited circular reduction: on its face, Corollary 3.8 is a general Lorentzian statement about level sets of generalized permutahedra, and Theorem B is stronger than the planar special-alternating case because it covers all bipartite graphs. The asserted 'readily proves' tiling/trimming equality in Theorem 6.5 is likewise a correctness risk, but it is not circular. The totally positive matrix results in Section 8 are self-contained explicit q-number expansions with positive coefficients, independent of the Alexander-polynomial identification. Overall, the central claims have independent mathematical content and the paper does not reduce its predictions to its inputs by construction.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central claims do not fit numerical parameters; the auxiliary vector l in Lemma 7.1 is existential and the polynomial f_A is independent of its choice. The paper relies on standard results in oriented matroids, zonotopal tilings, Lorentzian polynomials, total positivity, and knot theory. The heaviest unstated input is [HMV24a, Corollary 3.8] behind Theorem 7.8, which should be stated explicitly to make the proof fully checkable.

assumptions (5)
  • standard math The Alexander polynomial of a special alternating link L_G equals the Murasugi-Stoimenow polynomial P_D(t), used as the definition in Definition/Theorem 2.2 from [MS03, Theorem 2].
    This bridges knot theory to combinatorics; it is an external prior theorem taken as the definition here.
  • standard math The polynomial f_A(t) is independent of the generic vector rho, Theorem 3.4, proven in [LP13].
    Central to the definition of f_A; the paper gives only a proof idea and cites LP13.
  • standard math The zonotopal tiling theorem, Theorem 6.1 from [LP13], and the trimming identity in Theorem 6.5, asserted as 'readily proves'.
    Used to identify coefficients of f_A with integer point counts on parallel hyperplanes.
  • standard math Integer point transforms of generalized permutahedra are denormalized Lorentzian polynomials, Theorem 7.6 from [BH20, Theorem 3.10], together with the slice theorem 7.8 relying on [HMV24a, Corollary 3.8].
    The key tool for log-concavity; the HMV24a corollary is not stated in the paper.
  • standard math Every totally positive matrix has a network parametrization, Lemma 8.5, following [Pos06].
    Used to prove that totally positive matrices with last row all 1s are flat max-positive.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Trapezodial property of the generalized Alexander polynomial." pith.science (2026). https://pith.science/paper/HBKLJUD2

@misc{pith2026250420967,
  author       = {Pith},
  title        = {Pith review of: Trapezodial property of the generalized Alexander polynomial},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HBKLJUD2}},
  note         = {Machine review of arXiv:2504.20967}
}
read the original abstract

Fox's conjecture from 1962, that the absolute values of the coefficients of the Alexander polynomial of an alternating link are trapezoidal, has remained stubbornly open to this date. Recently Fox's conjecture was settled for all special alternating links. In this paper we take a broad view of the Alexander polynomials of special alternating links, showing that they are a generating function for a statistic on certain vector configurations. We study three types of vector configurations: (1) vectors arising from cographic matroids, (2) vectors arising from graphic matroids, (3) vectors arising from totally positive matrices. We prove that Alexander polynomials of special alternating links belong to both classes (1) and (2), and prove log-concavity, respectively trapezoidal, properties for classes (2) and (3). As a special case of our results, we obtain a new proof of Fox's conjecture for special alternating links.

Figures

Figures reproduced from arXiv: 2504.20967 by the authors.

Figure 1
Figure 1. A plane bipartite graph G, medial graph M(G), the associated special alternat￾ing link LG, and the dual graph G∗ of G endowed with an Eulerian orientation that makes it into an alternating dimap D. The edges of M(G) also receive orientations by stipulating that the faces of M(G) that surround vertices of G be consistently oriented, while around the boundaries of the other faces of M(G) the orientations alternate. No… view at source ↗
Figure 2
Figure 2. On the left is a directed graph D. On the right is its signed incidence matrix I(D). v1 v2 v3 v4 v5 e1 e2 e3 e4 e5 e6 e7 AD,T =   e1 e2 e3 e4 e5 e6 e7 e1 1 0 0 0 −1 0 0 e2 0 1 0 0 1 −1 −1 e3 0 0 1 0 0 −1 −1 e4 0 0 0 1 0 0 1   [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. On the left is a directed graph D along with a spanning tree T. On the right is the matrix AD,T of Definition 4.1, containing the highlighted block M. same linear dependences among its columns as I(D) does. (This is because it is in fact obtained from I(D) by Gaussian elimination followed by deleting a trivial row.) As such, the oriented matroids M[I(D)] and M[I| V(G)|−1 | M] are isomorphic and the polynomials fI(D)… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: On the left is a directed graph D along with a fixed spanning tree T. On the right is the matrix BD,T of Definition 4.8, with its block K highlighted. D. We fix an arbitrary root vertex r ∈ V(D) and start by determining the flows on the leaves of T that are farthest aw…
Figure 5
Figure 5. Figure 5: The flow values on the edges of G⃗ with netflow vector l coincide with the coef￾ficients αj from Definition 6.2. 7. Log-concavity of fA in the flat graphic case In this section we prove Theorem B (see Theorem 7.5 below). We accomplish this by exhibiting an admissible v…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 23 canonical work pages

  1. [1]

    Log-concave polynomials, I : entropy and a deterministic approximation algorithm for counting bases of matroids

    Nima Anari, Shayan Oveis Gharan, and Cynthia Vinzant. Log-concave polynomials, I : entropy and a deterministic approximation algorithm for counting bases of matroids. Duke Math. J. , 170(16):3459--3504, 2021

    work page 2021

  2. [2]

    On F ox's trapezoidal conjecture, 2024

    Soheil Azarpendar, Andr\'as Juh\'asz, and Tam\'as K\'alm\'an. On F ox's trapezoidal conjecture, 2024. https://arxiv.org/pdf/2406.08662

    arXiv 2024

  3. [3]

    Alexander

    James W. Alexander. Topological invariants of knots and links. Trans. Amer. Math. Soc. , 30(2):275--306, 1928

    work page 1928

  4. [4]

    Log-concave polynomials II : H igh-dimensional walks and an FPRAS for counting bases of a matroid

    Nima Anari, Kuikui Liu, Shayan Oveis Gharan, and Cynthia Vinzant. Log-concave polynomials II : H igh-dimensional walks and an FPRAS for counting bases of a matroid. In S TOC '19--- P roceedings of the 51st A nnual ACM SIGACT S ymposium on T heory of C omputing , pages 1--12. ACM, New York, 2019

    work page 2019

  5. [5]

    Log-concave polynomials III : M ason's ultra-log-concavity conjecture for independent sets of matroids

    Nima Anari, Kuikui Liu, Shayan Oveis Gharan, and Cynthia Vinzant. Log-concave polynomials III : M ason's ultra-log-concavity conjecture for independent sets of matroids. Proc. Amer. Math. Soc. , 152(5):1969--1981, 2024

    work page 1969

  6. [6]

    Lorentzian polynomials

    Petter Br \"a nd \'e n and June Huh. Lorentzian polynomials . Ann. of Math. (2) , 192(3):821 -- 891, 2020

    work page 2020

  7. [7]

    Lower bounds for contingency tables via lorentzian polynomials

    Petter Br \"a nd \'e n, Jonathan Leake, and Igor Pak. Lower bounds for contingency tables via lorentzian polynomials. Israel J. Math. , pages 1--48, 2022

    work page 2022

  8. [8]

    orner, Michel Las Vergnas, Bernd Sturmfels, Neil White, and G\

    Anders Bj\"orner, Michel Las Vergnas, Bernd Sturmfels, Neil White, and G\"unter M. Ziegler. Oriented matroids , volume 46 of Encyclopedia of Mathematics and its Applications . Cambridge University Press, Cambridge, 1993

    work page 1993

Show all 28 references
  1. [9]

    Submodular functions, matroids, and certain polyhedra

    Jack Edmonds. Submodular functions, matroids, and certain polyhedra. In Combinatorial S tructures and their A pplications ( P roc. C algary I nternat. C onf., C algary, A lta., 1969) , pages 69--87. Gordon and Breach, New York-London-Paris, 1970

    1969

  2. [10]

    Some problems in knot theory

    Ralph Hartzler Fox. Some problems in knot theory. In M.K. Fort, editor, Topology of 3-manifolds and related topics , pages 168--176. Prentice Hall, 1962

    1962

  3. [11]

    Alternating links and definite surfaces

    Joshua Evan Greene. Alternating links and definite surfaces. Duke Math. J. , 166(11):2133--2151, 2017. With an appendix by Andr\'as Juh\'asz and Marc Lackenby

    2017

  4. [12]

    Richard I. Hartley. On two-bridged knot polynomials. J. Aust. Math. Soc. , 28(2):241--249, 1979

    1979

  5. [13]

    Arithmetic aspects of symmetric edge polytopes

    Akihiro Higashitani, Katharina Jochemko, and Mateusz Micha ek. Arithmetic aspects of symmetric edge polytopes. Mathematika , 65(3):763--784, 2019

    2019

  6. [14]

    Log- C oncavity of the A lexander P olynomial

    Elena S Hafner, Karola M\' e sz\' a ros, and Alexander Vidinas. Log- C oncavity of the A lexander P olynomial. Int. Math. Res. Not. IMRN , (13):10273--10284, 2024

    2024

  7. [15]

    Hafner, Karola Mészáros, and Alexander Vidinas

    Elena S. Hafner, Karola Mészáros, and Alexander Vidinas. On the A lexander polynomial of special alternating links, 2024. https://arxiv.org/pdf/2401.14927

    2024 arXiv

  8. [16]

    Joshua A. Howie. A characterisation of alternating knot exteriors. Geom. Topol. , 21(4):2353--2371, 2017

    2017

  9. [17]

    Alexander polynomials of alternating knots of genus two

    In Dae Jong. Alexander polynomials of alternating knots of genus two. Osaka J. Math. , 46(1):353--371, 2009

    2009

  10. [18]

    Root polytopes and J aeger-type dissections for directed graphs

    Tam\' a s K\' a lm\' a n and Lilla T\' o thm\' e r\' e sz. Root polytopes and J aeger-type dissections for directed graphs. Mathematika , 68(4):1176--1220, 2022

    2022

  11. [19]

    Slicing zonotopes

    Nan Li and Alexander Postnikov. Slicing zonotopes. preprint, 2013

    2013

  12. [20]

    The A lexander polynomial of planar even valence graphs

    Kunio Murasugi and Alexander Stoimenow. The A lexander polynomial of planar even valence graphs. Adv. in Appl. Math. , 31(2):440--462, 2003

    2003

  13. [21]

    On the A lexander polynomial of alternating algebraic knots

    Kunio Murasugi. On the A lexander polynomial of alternating algebraic knots. J. Aust. Math. Soc. , 39(3):317–333, 1985

    1985

  14. [22]

    Heegaard F loer homology and alternating knots

    Peter O zs v\'ath and Zolt\'an Sz ab\'o. Heegaard F loer homology and alternating knots. Geom. Topol. , 7:225--254, 2003

    2003

  15. [23]

    Matroid theory , volume 21 of Oxford Graduate Texts in Mathematics

    James Oxley. Matroid theory , volume 21 of Oxford Graduate Texts in Mathematics . Oxford University Press, Oxford, second edition, 2011

    2011

  16. [24]

    Total positivity, G rassmannians, and networks, 2006

    Alexander Postnikov. Total positivity, G rassmannians, and networks, 2006. https://arxiv.org/pdf/math/0609764

    2006 arXiv

  17. [25]

    Permutohedra, associahedra, and beyond

    Alexander Postnikov. Permutohedra, associahedra, and beyond. Int. Math. Res. Not. IMRN , (6):1026--1106, 2009

    2009

  18. [26]

    Knots and links , volume No

    Dale Rolfsen. Knots and links , volume No. 7 of Mathematics Lecture Series . Publish or Perish, Inc., Berkeley, CA, 1976

    1976

  19. [27]

    Newton-like polynomials of links

    Alexander Stoimenow. Newton-like polynomials of links. Enseign. Math. (2) , 51(3-4):211--230, 2005

    2005

  20. [28]

    A geometric proof for the root-independence of the greedoid polynomial of E ulerian branching greedoids

    Lilla T \'o th m\'er\'esz. A geometric proof for the root-independence of the greedoid polynomial of E ulerian branching greedoids. J. Combin. Theory Ser. A , 206:Paper No. 105891, 21, 2024

    2024

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.