The inverse $Z$-polynomial of a matroid

Alice L. L. Gao; Matthew H. Y. Xie; Xuan Ruan

The inverse $Z$-polynomial of a matroid

T0 review · 0 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read For sparse paving matroids, the inverse Z-polynomial equals the uniform-matroid formula minus a Catalan correction, and its coefficients are unimodal and log-concave.

desk verdict A solid, workmanlike paper: first systematic study of the inverse Z-polynomial, clean closed forms for uniform and sparse paving matroids, and a mostly hand-checkable unimodality/log-concavity proof; minor gaps but worth refereeing. read the letter →

arxiv 2507.01332 v1 pith:AQKY2TAY submitted 2025-07-02 math.CO

Alice L. L. Gao , Xuan Ruan , Matthew H. Y. Xie This is my paper · ORCID

classification math.CO MSC 05B3552B4005A20

keywords inverseZ-polynomialsparsepavingmatroiduniformvaluativeinvariantunimodalitylog-concavityCatalannumber

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

The paper establishes that the inverse Z-polynomial of a matroid is a well-behaved companion to the Z-polynomial: its coefficients are non-negative, it is multiplicative under direct sums, and the assignment is a valuation on matroid polytopes. Its main computational result is an explicit formula for sparse paving matroids, the class obtained from a uniform matroid by relaxing $\lambda$ circuit-hyperplanes: the inverse Z-polynomial is the uniform case corrected by a single term built from $(1+t)^k$ and, for even rank, the Catalan number. The paper also proves that for sparse paving matroids the coefficient sequence is unimodal, log-concave, and free of internal zeros, confirming a general conjecture in this class. This matters because sparse paving matroids are conjectured to account for almost all matroids, so the formulas constrain a near-universal family.

What carries the argument

The central object is the inverse Z-polynomial $Y_M(t)$, defined as $(-1)^{\mathrm{rk}(M)}$ times a sum over flats of $M$ in which each flat contributes an inverse Kazhdan–Lusztig polynomial, a power of $t$ from the contracted matroid, and the Möbius invariant of that contraction. Two structural facts carry the argument. First, the map $M \mapsto Y_M(t)$ is a valuative invariant: it is compatible with subdivisions of matroid base polytopes, by way of convolution of valuations, so any decomposition of a matroid's polytope into simpler pieces yields the polynomial by inclusion–exclusion. Second, a sparse paving matroid is exactly a uniform matroid with $\lambda$ circuit-hyperplanes relaxed, and the relaxation changes the valuative invariant by a single computable correction. The Catalan number appears in the even-rank correction because the middle coefficient of $Y_{U_{k,k+1}}(t) - (1+t)Y_{U_{k-1,k}}(t)$ is $\binom{k}{k/2} - C_{k/2}$.

What would settle it

Take any small sparse paving matroid, say rank 4 on 8 elements with the maximum allowed number of circuit-hyperplanes, compute its inverse Z-polynomial by summing over all flats, and compare each coefficient with the formula of Theorem 1.3 and Corollary 5.4; a single negative coefficient, a violation of $a_i^2 \ge a_{i-1}a_{i+1}$, or any disagreement with the formula would falsify the paper's central claim.

Watch

Extended reading notes

Core claim

The central discovery is that a sparse paving matroid is determined, for inverse Z-polynomial purposes, by its rank $k$, cardinality $n$, and the number $\lambda$ of circuit-hyperplanes. Explicitly, $Y_M(t) = Y_{U_{k,n}}(t) - \lambda\bigl((1+t)^k - \tfrac{1+(-1)^k}{2} C_{k/2} t^{k/2}\bigr)$, where $C_{k/2}$ is the Catalan number and the second term is understood to vanish when $k$ is odd; when $\lambda=0$ this reduces to the uniform matroid. From this formula the coefficients are read off as $\binom{n}{k}\binom{k}{i}\bigl(\tfrac{k-i}{n-i} - \lambda^*\bigr)$ for $i$ below the middle (with a slightly different expression at the middle when $k$ is even), where $\lambda^* = \lambda/\binom{n}{k}$. The paper proves these coefficients are non-negative, palindromic, unimodal, log-concave, and have no internal zeros, and it upgrades the general unimodality and log-concavity conjecture to a theorem for sparse paving matroids.

Load-bearing premise

The coefficient non-negativity and unimodality arguments rely on an earlier bound limiting how many circuit-hyperplanes (special rank-$(k-1)$ flats) a sparse paving matroid can contain; if that bound failed, the positivity and unimodality proofs would need new input.

Editorial extensions

If this is right

For any sparse paving matroid, the inverse Z-polynomial is completely determined by rank, cardinality, and the number of circuit-hyperplanes; no finer matroid data enter.
The coefficients of these polynomials are non-negative, palindromic, unimodal, and log-concave with no internal zeros, so the general unimodality and log-concavity conjectures are true for the whole sparse paving class.
For uniform matroids the coefficient formula $\binom{n}{i}\binom{n-i-1}{n-k}$ gives a closed form up to the middle degree, with the upper half determined by palindromicity.
A new closed expression for the Z-polynomial of uniform matroids follows by inverting the relation between Z-polynomials and inverse Z-polynomials.
The valuative and multiplicative properties mean inverse Z-polynomials can be computed by subdividing base polytopes and by splitting direct sums into products.

Reading between the lines

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

Since sparse paving matroids are conjectured to make up almost all matroids asymptotically, Theorem 1.6 is evidence that the unimodality and log-concavity conjectures hold in full generality; a natural next test is the larger class of elementary split matroids, where the same valuative argument gives a multi-term correction instead of a single Catalan term.
The explicit $\lambda^*$-dependent coefficient formula invites a bijective interpretation: the coefficients appear to count $k$-subsets of $[n]$ with a penalty proportional to how many circuit-hyperplanes they meet, and finding such a bijection could yield an independent proof of log-concavity.
The paper's example of $U_{4,5}$ shows the inverse Z-polynomial need not be $\gamma$-positive even when it is unimodal, so any future geometric or representation-theoretic model for these polynomials would have to work outside the $\gamma$-positive framework that succeeds for Z-polynomials.

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.

Desk Editor's Note

read the letter

Colleague,

Quick take: this is a solid, workmanlike paper that gives the first systematic treatment of the inverse Z-polynomial Y_M. It proves non-negativity, multiplicativity, and valuativity for Y_M, produces explicit closed forms for uniform and sparse paving matroids, and shows the coefficients of Y_M for sparse paving matroids are unimodal and log-concave with no internal zeros. The central formulas hang together. I spot-checked identity (12) for small k and it works, and the sparse-paving correction is consistent with direct sum examples like U_{2,3} ⊕ U_{1,1}.

What is actually new: the valuativity result and the explicit closed forms. The valuativity proof is clean, using Ardila–Sanchez convolution and known valuativity of the Tutte polynomial and inverse Kazhdan–Lusztig polynomial. The uniform matroid formula follows from the authors' earlier inverse KL formula, which is fair game since that is a published independent result. The sparse paving formula is a neat application of the Ferroni–Schröter relaxation framework, and the unimodality/log-concavity proof is a real piece of binomial coefficient analysis.

Soft spots, in proportion. Lemma 2.2 (palindromicity) is cited to a remark in Braden et al. without proof; a referee should ask for a three-line proof or a precise reference, but this is minor. The final log-concavity inequality (24) is verified by Mathematica's Resolve rather than a human-readable derivation; it looks like a low-degree rational inequality that could be cleared in closed form, but as written the paper leaves it to the computer. Also minor. The one genuinely load-bearing external input is the bound λ ≤ C(n,k) min{1/(k+1), 1/(n-k+1)} from Ferroni–Schröter, used in the non-negativity and unimodality proofs. It is a published result and appears correct, but if it ever failed the sparse paving non-negativity and unimodality arguments would need new input. Finally, there are a few cross-reference slips (e.g., "Lemma 2.3" where Proposition 2.3 is meant) — cosmetic.

Who this is for: anyone working on Kazhdan–Lusztig–Stanley invariants of matroids. It provides the first systematic computational handle on Y_M and strengthens the case for the conjectured unimodality and log-concavity across all matroids. The paper deserves a serious referee. My recommendation: send it to review, with a request to fill the palindromicity proof and ideally convert the computational verification of (24) into a readable derivation.

Referee Report

0 major / 7 minor

Summary. The paper studies the inverse Z-polynomial Y_M(t) of a matroid, a notion introduced by Ferroni, Matherne, Stevens, and Vecchi. The authors prove that Y_M has non-negative coefficients, is palindromic, is multiplicative under direct sums, and is a valuative invariant. Their main results are explicit formulas: Theorem 1.1 gives Y_{U_{k,n}}(t) for uniform matroids, Theorem 1.3 gives Y_M(t) for sparse paving matroids in terms of the number of circuit-hyperplanes, and Theorem 1.6 establishes unimodality and log-concavity with no internal zeros for sparse paving matroids. The proofs combine known results on inverse Kazhdan-Lusztig polynomials and valuative invariants with direct binomial-coefficient algebra; one central inequality is verified by a Mathematica Resolve computation included in the text.

Significance. If valid, this is a substantial contribution to the study of inverse Z-polynomials. It provides the first systematic computation of these polynomials for a broad and combinatorially important class (sparse paving matroids), and it resolves the authors' Conjectures 1.4 and 1.5 for that class. The valuativity result is conceptually important and is applied systematically through the relaxation machinery of Ferroni and Schröter. The paper has several concrete strengths: the formulas are parameter-free and explicit, the main derivation is internally consistent, and the computer-assisted verification of the final inequality is reproducible from the provided code. The only external input that is load-bearing in the unimodality and non-negativity proofs is the bound on λ from [9, Corollary 4.13]; this is a published result, and the extremal cases checked here are consistent with it, so I do not see a circularity or a plausibly false dependency.

minor comments (7)

[Section 2, proof of Proposition 2.3] The sentence 'By the inductive hypothesis, we get...' is misleading: the displayed equality is just the definition (2) applied to M1 ⊕ M2, and the induction hypothesis is not actually used in the way described. Since the equality follows directly from multiplicativity of the inverse Kazhdan-Lusztig polynomial, multiplicativity of the Möbius invariant, and additivity of rank, please rewrite this proof either by removing the induction or by applying the induction hypothesis explicitly to the relevant restrictions.
[Section 2, general notation] The symbol \hat Q is used in the proof of Proposition 2.3, while the definition (2) and the surrounding text use Q for the inverse Kazhdan-Lusztig polynomial. Please standardize the notation to avoid confusion.
[Section 6, inequality (24)] The verification of inequality (24) is performed by Mathematica's Resolve rather than by a human-readable argument. Since this inequality is the final load-bearing step in the proof of Theorem 1.6(2), please either provide a hand-checkable derivation or at least describe the quantifier-elimination setup and the exact assumptions entered, so that the computer-assisted proof is fully transparent and reproducible.
[Section 5, Proposition 5.2] The reference to 'Lemma 2.3' should be to 'Proposition 2.3', which is the multiplicativity result proven earlier in the paper.
[Section 1, Theorem 1.3] In the statement of Theorem 1.3, the Catalan term vanishes when k is odd; stating this explicitly would improve readability, since the notation C_{k/2} is otherwise undefined for odd k.
[Section 2, proof of Proposition 2.1] The proof uses the fact that the inverse Kazhdan-Lusztig polynomial Q has non-negative coefficients without a citation at that point; please add a reference to [13] (or to the relevant result therein) for completeness.
[Section 4, Corollary 4.2] The formula in Corollary 4.2 is dense and the change of summation limits is not immediately transparent; a short derivation of the interchange of sums or an example would help the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the inverse Z-polynomial formulas and unimodality proofs are derived from definitions and external cited theorems, with no fitted parameter renamed as a prediction.

full rationale

The derivation chain is self-contained in the relevant sense. Theorem 1.1 computes Y_{U_{k,n}} from the defining equation (2)-(3), the Möbius invariant formula (7) cited to Zaslavsky, and the inverse Kazhdan-Lusztig formulas (9)-(10) cited to Gao and Xie [13]; the target Y is not used to define Q, and [13] is a separately published result rather than a restatement of the present theorem. The valuation proof (Theorem 1.2) uses Ardila-Sanchez Theorem C and [9, Theorem 7.17] for the Tutte polynomial; again these are independent inputs. Theorem 1.3 reduces the sparse-paving case to Theorem 1.1 and the valuative expansion [9, Theorem 5.3], with identity (12) proved directly from Theorem 1.1. The unimodality and log-concavity proof uses the coefficient formula (14) and the external bound λ ≤ C(n,k) min{1/(k+1), 1/(n-k+1)} from [9, Corollary 4.13]; this is a stated external hypothesis, not a fitted quantity, and no prediction is claimed from a parameter fitted to the same data. There are no self-definitional normalizations, no unique-choice theorem imported from the authors, and no ansatz smuggled in by citation. The only overlap with the authors' prior work is the cited formula for inverse Kazhdan-Lusztig polynomials of uniform matroids [13], which is logically prior to and distinct from the inverse Z-polynomial, so it does not make the derivation circular.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new entities or free parameters; it rests on established KL-Stanley theory, published formulas for inverse KL polynomials, and valuation results from the matroid combinatorics literature. No assumption here includes the target result.

assumptions (6)

standard math Kazhdan-Lusztig and inverse Kazhdan-Lusztig polynomials of matroids exist with standard properties (non-negative coefficients, multiplicativity under direct sums).
Used throughout; e.g., Proposition 2.1 and proof of Proposition 2.3. Results from [5], [13], [19].
standard math Q_{B_n}(t)=1 and the closed formula for Q_{U_{k,n}}(t) from Gao-Xie [13, Corollary 3.2 and Theorem 3.3].
Used in the proof of Theorem 1.1, equation (9) and (10).
standard math The assignment M -> Q_M(t) is a valuative invariant, and the Tutte polynomial is valuative (Ardila-Sanchez [1, Theorem 8.8], Ferroni-Schroter [9, Theorem 7.17]).
Used in the proof of Theorem 1.2.
standard math The convolution of two valuative invariants is a valuative invariant (Ardila-Sanchez [1, Theorem C]).
Used in the proof of Theorem 1.2.
domain assumption For a sparse paving matroid of rank k on n elements with lambda circuit-hyperplanes, lambda <= C(n,k) min{1/(k+1), 1/(n-k+1)} (Ferroni-Schroter [9, Corollary 4.13]).
Used in the second proof of Proposition 2.1 (inequality (18)) and in the unimodality proof of Theorem 1.6.
standard math The inverse Z-polynomial Y_M(t) is palindromic (Braden-Huh-Matherne-Proudfoot-Wang [3, p.5]).
Stated as Lemma 2.2 without proof; used in Theorem 1.1 and Section 6.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The inverse $Z$-polynomial of a matroid." pith.science (2026). https://pith.science/paper/AQKY2TAY

@misc{pith2026250701332,
  author       = {Pith},
  title        = {Pith review of: The inverse $Z$-polynomial of a matroid},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AQKY2TAY}},
  note         = {Machine review of arXiv:2507.01332}
}

abstract

Motivated by the $Z$-polynomials of matroids, Ferroni, Matherne, Stevens, and Vecchi introduced the inverse $Z$-polynomial of a matroid. In this paper, we prove several fundamental properties of the inverse $Z$-polynomial, including non-negativity and multiplicativity, and show that it is a valuative invariant. We also provide explicit formulas for the inverse $Z$-polynomials of uniform matroids and a broader class of matroids, namely sparse paving matroids, which include uniform matroids as a special case. Furthermore, we establish the unimodality and log-concavity of these polynomials in the case of sparse paving matroids. Based on the properties of the $Z$-polynomial, we conjecture that the coefficients of the inverse $Z$-polynomial are unimodal and log-concave.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 21 canonical work pages

[1]

Ardila and M

F. Ardila and M. Sanchez. Valuations and the Hopf monoid of generalized permutahe- dra. International Mathematics Research Notices , 2023, 2023(5): 4149-4224. 8 21

work page 2023
[2]

B´ erczi, T

K. B´ erczi, T. Kir´ aly, T. Schwarcz, Y. Yamaguchi, and Y. Yokoi. Hypergraph characteri- zation of split matroids. Journal of Combinatorial Theory, Series A . 2023, 194: 105697. 12

work page 2023
[3]

Braden, J

T. Braden, J. Huh, J. Matherne, N. Proudfoot, and B. Wang. Singular Hodge theory for combinatorial geometries. arXiv:2010.06088, 2020. 1, 4, 5, 17

arXiv 2010
[4]

Braden and A

T. Braden and A. Vysogorets. Kazhdan-Lusztig polynomials of matroids under deletion. The Electronic Journal of Combinatorics , 2020, #P1.17. 1

work page 2020
[5]

Elias, N

B. Elias, N. Proudfoot, and M. Wakefield. The Kazhdan-Lusztig polynomial of a ma- troid. Advances in Mathematics , 2016, 299: 36-70. 1, 5, 11

work page 2016
[6]

Ferroni and M

L. Ferroni and M. Larson. Kazhdan-Lusztig polynomials of braid matroids. Communi- cations of the American Mathematical Society . 2024, 4(02): 64-79. 1

work page 2024
[7]

Ferroni L, J

L. Ferroni L, J. Matherne, M. Stevens, and L. Vecchi. Hilbert-Poincar´ e series of matroid Chow rings and intersection cohomology. Advances in Mathematics. 2024, 449: 109733. 1, 2

work page 2024
[8]

Ferroni, G

L. Ferroni, G. Nasr, and L. Vecchi. Stressed hyperplanes and Kazhdan-Lusztig gamma- positivity for matroids. International Mathematics Research Notices . 2023, 2023(24): 20883-20942. 1, 12, 13

work page 2023

Show all 22 references

[9]

Ferroni and B

L. Ferroni and B. Schr¨ oter. Valuative invariants for large classes of matroids. Journal of the London Mathematical Society . 2024, 110(3): e12984. 1, 6, 8, 12, 13, 16, 21

2024
[10]

Ferroni and L

L. Ferroni and L. Vecchi. Matroid relaxations and Kazhdan-Lusztig non-degeneracy. Algebraic Combinatorics. 2022, 5(4): 745-769. 1, 12

2022
[11]

A. Gao, L. Lu, M. Xie, A. Yang, and P. Zhang. The Kazhdan-Lusztig polynomials of uniform matroids. Advances in Applied Mathematics . 2021, 122: 102117. 1, 10

2021
[12]

A. Gao, N. Proudfoot, A. Yang, and Z. Zhang. The combinatorics behind the leading Kazhdan-Lusztig coefficients of braid matroids, The Electronic Journal of Combina- torics, 2024, 31(3): #P3.21. 1

2024
[13]

Gao and M

A. Gao and M. Xie. The inverse Kazhdan-Lusztig polynomial of a matroid. Journal of Combinatorial Theory, Series B . 2021, 151: 375-392. 1, 5, 9, 10

2021
[14]

K. Gedeon. Kazhdan-Lusztig polynomials of thagomizer matroids. The Electronic Journal of Combinatorics , 2027, 24(3): #P3.12. 1

2027
[15]

Gedeon, N

K. Gedeon, N. Proudfoot, and B. Young. Kazhdan-Lusztig polynomials of matroids: a survey of results and conjectures. S´ eminaire Lotharingien de Combinatoire. 2017, 78B, Article 80. 1 22

2017
[16]

Joswig and B

M. Joswig and B. Schr¨ oter. Matroids from hypersimplex splits. Journal of Combina- torial Theory, Series A . 2017, 151: 254-284. 12

2017
[17]

Mayhew, M

D. Mayhew, M. Newman, D. Welsh, and G. Whittle. On the asymptotic proportion of connected matroids. European Journal of Combinatorics . 2011, 32(6): 882-890. 12

2011
[18]

J. Oxley. Matroid theory. Oxford University Press . New York, USA, 2011. 2

2011
[19]

Proudfoot

N. Proudfoot. The algebraic geometry of Kazhdan-Lusztig-Stanley polynomials. EMS Surveys in Mathematical Sciences . 2018, 5(1): 99-127. 1, 4

2018
[20]

Proudfoot, Y

N. Proudfoot, Y. Xu, and B. Young. The Z-polynomial of a matroid. The Electronic Journal of Combinatorics . 2018, 25(1): #P1.26. 1, 2, 3, 5, 10, 17

2018
[21]

G.-C. Rota. On the foundations of combinatorial theory. I. Theory of M¨ obius functions. Z. Wahrscheinlichkeitstheorie und Verw. Gebiete . 1964, 2:340–368. 4

1964
[22]

Zaslavsky

T. Zaslavsky. The M¨ obius function and the characteristic polynomial. InCombinatorial geometries, volume 29 of Encyclopedia Math. Appl. , pages 114–138. Cambridge Univ. Press, Cambridge, 1987. 6, 9 23

1987

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Ardila and M

F. Ardila and M. Sanchez. Valuations and the Hopf monoid of generalized permutahe- dra. International Mathematics Research Notices , 2023, 2023(5): 4149-4224. 8 21

work page 2023

[2] [2]

B´ erczi, T

K. B´ erczi, T. Kir´ aly, T. Schwarcz, Y. Yamaguchi, and Y. Yokoi. Hypergraph characteri- zation of split matroids. Journal of Combinatorial Theory, Series A . 2023, 194: 105697. 12

work page 2023

[3] [3]

Braden, J

T. Braden, J. Huh, J. Matherne, N. Proudfoot, and B. Wang. Singular Hodge theory for combinatorial geometries. arXiv:2010.06088, 2020. 1, 4, 5, 17

arXiv 2010

[4] [4]

Braden and A

T. Braden and A. Vysogorets. Kazhdan-Lusztig polynomials of matroids under deletion. The Electronic Journal of Combinatorics , 2020, #P1.17. 1

work page 2020

[5] [5]

Elias, N

B. Elias, N. Proudfoot, and M. Wakefield. The Kazhdan-Lusztig polynomial of a ma- troid. Advances in Mathematics , 2016, 299: 36-70. 1, 5, 11

work page 2016

[6] [6]

Ferroni and M

L. Ferroni and M. Larson. Kazhdan-Lusztig polynomials of braid matroids. Communi- cations of the American Mathematical Society . 2024, 4(02): 64-79. 1

work page 2024

[7] [7]

Ferroni L, J

L. Ferroni L, J. Matherne, M. Stevens, and L. Vecchi. Hilbert-Poincar´ e series of matroid Chow rings and intersection cohomology. Advances in Mathematics. 2024, 449: 109733. 1, 2

work page 2024

[8] [8]

Ferroni, G

L. Ferroni, G. Nasr, and L. Vecchi. Stressed hyperplanes and Kazhdan-Lusztig gamma- positivity for matroids. International Mathematics Research Notices . 2023, 2023(24): 20883-20942. 1, 12, 13

work page 2023