Pith. sign in

REVIEW 2 major objections 4 minor 22 references

Induced equators in flag spheres

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read The paper proves a new nonlinear h-vector inequality for flag polytopes and proposes that, for flag homology spheres, every induced codimension-one homology sphere (an equator) has gamma-polynomial no larger than the ambient sphere.

desk verdict Solid, mainly unconditional contributions to flag sphere face enumeration; the unproved Lemma 3.6 is real but not as load-bearing as the stress-test claims, because Theorem 3.7 independently proves the dimension-two structural statement. read the letter →

arxiv 1908.08727 v1 pith:6OTKNJDD submitted 2019-08-23 math.CO

classification math.CO MSC 05E4552B0505C70
keywords flaghomologyspheregamma-vectorequatorhalf-integralperfectmatchingpolytopeh-vectorcrosspolytopebalanced
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

This paper tries to establish that for a flag homology sphere, the gamma-polynomial of any induced codimension-one homology sphere inside it is coefficientwise no larger than the sphere's own gamma-polynomial. This Equator Conjecture is shown to be equivalent to the older Link Conjecture, and would imply the conjectured nonnegativity of gamma-vectors for flag homology spheres. On the unconditional side, the paper proves a new nonlinear inequality for the h-vectors of flag polytopes, obtained by shelling from pairs of non-adjacent vertices and using a half-integral perfect matching in the complement graph. It also proves the Link Conjecture for the family of flag polytopes obtained from crosspolytopes by stellar subdivisions of edges, and verifies the structural route in dimensions up to two.

What carries the argument

The central objects are the gamma-polynomial $\gamma_\Delta(t)=\sum_i \gamma_i t^i$, defined from the palindromic h-polynomial by $h_\Delta(t)=\sum_i \gamma_i t^i(1+t)^{d-2i}$, and the notion of an equator. The paper's main mechanism is a pair of inequalities. Lemma 4.1 says that for every non-edge $uv$ of a flag polytope boundary, $h_{\operatorname{lk}u\Delta}(t)+t h_{\operatorname{lk}v\Delta}(t)\le h_\Delta(t)$, obtained by line shellings of the polytope. Theorem 1.7 says the complement $G$ of the 1-skeleton of a flag homology sphere has a half-integral perfect matching, a function $f:E(G)\to\{0,\tfrac12,1\}$ such that for every vertex the sum of $f$ over incident edges is $1$; equivalently, its vertex set partitions into a matching and odd cycles. Summing Lemma 4.1 over the oriented edges of this decomposition gives Theorem 1.6. For the conditional conjecture, the key reduction is Proposition 3.1, which rewrites the equator algebra as $\gamma_\Delta=\gamma_{\Delta_1}+\gamma_{\Delta_2}-\gamma_E$, and Proposition 3.2, which reduces the conjecture to minimal flag spheres.

What would settle it

Enumerate all flag homology 3-spheres in the minimal family (not suspensions, every edge in some chordless four-cycle) up to a small vertex count; for each, compute the gamma-vector of every vertex link and compare it coefficientwise with the gamma-vector of the sphere. A single link with a larger coefficient would refute the Equator and Link Conjectures; separately, the same check on flag 2-spheres would confirm or kill the omitted lemma.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is Conjecture 1.3: if $E$ is an equator of a flag homology sphere $\Delta$ — an induced subcomplex that is itself a homology sphere of codimension one — then $\gamma(E)\le \gamma(\Delta)$ coefficientwise. The paper proves this conjecture is equivalent to the Link Conjecture for vertex links, and that both would imply the nonnegativity of the gamma-vector for flag homology spheres. Unconditionally, it proves Theorem 1.6: for the boundary complex $\Delta$ of any flag $d$-polytope, $$(1+t)\sum_{v\in\Delta_0} h_{\operatorname{lk}_v\$\Delta$}(t)\le f_0(\$\Delta$)\,h_\$\Delta$(t),$$ with equality only for crosspolytopes. Combined with the standard link-sum formula for h-vectors this yields the h-vector inequality $$h_1 h_i\ge (d-i+1)h_{i-1}+(i+1)h_{i+1}\qquad (0\le i\le d),$$ and the same inequality is shown to hold for balanced polytopes. The engine is a new graph-theoretic fact: the complement of the 1-skeleton of any flag homology sphere has a half-integral perfect matching.

Load-bearing premise

The argument's weakest load-bearing premise is the omitted Lemma 3.6: every flag two-sphere except the octahedron's boundary has an edge that is not a side of any chordless four-cycle, and the paper's dimension-two evidence for the main conjecture depends entirely on that lemma.

Editorial extensions

If this is right

  • If the Equator Conjecture holds, every flag homology sphere has coefficientwise nonnegative gamma-vector, settling the flag version of the gamma-nonnegativity conjecture.
  • The h-vector inequality $h_1 h_i\ge (d-i+1)h_{i-1}+(i+1)h_{i+1}$ holds for all flag polytopes and all balanced polytopes, giving concrete numerical upper bounds on their face numbers.
  • The structural conjecture is verified for flag spheres of dimension at most two, so the Equator and Link Conjectures hold in dimensions zero, one, and two.
  • Every flag polytope obtained from a crosspolytope by successive stellar subdivisions of edges satisfies the Link Conjecture, and this family is closed under suspensions and links.
  • The half-integral matching theorem applies to all flag homology spheres, not only to polytopal boundaries, so the matching mechanism is available beyond the polytope case.

Reading between the lines

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

  • If the Equator Conjecture is true, the gamma-polynomial behaves like a monotone size statistic under induced codimension-one inclusions, which may point toward a hidden variational or geometric meaning for gamma-vectors that the paper does not identify.
  • The half-integral matching result could plausibly be extended to flag triangulations of homology manifolds with boundary or to flag complexes with Cohen-Macaulay links, where a similar shelling-plus-matching argument might yield analogous h-inequalities.
  • A direct computational census of minimal flag 3-spheres, comparing each vertex link's gamma-vector with the sphere's, would be the most immediate test of the conjecture now that the dimension-two case is known.
  • The omitted proof of the key lemma in dimension two can probably be reconstructed from the cited vertex-splitting lemma; if that lemma were false, only the dimension-two evidence would collapse, leaving the polytopal h-inequality and matching theorem intact.
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 flag homology spheres and proposes the Equator Conjecture: for an induced flag homology sphere E of codimension 1 in a flag homology sphere Δ, the γ-vector inequality γ(E) ≤ γ(Δ) holds coefficientwise. It proves that this conjecture is equivalent to the Link Conjecture for vertex links (Proposition 3.1). The main unconditional result is Theorem 1.6: for the boundary complex Δ of a flag d-polytope, (1+t)∑_v h_{lk_v(Δ)}(t) ≤ f0(Δ) h_Δ(t), which is tight only for crosspolytopes. This is obtained via a half-integral perfect matching theorem for the complement of the 1-skeleton of a flag homology sphere (Theorem 1.7), and yields the nonlinear h-vector inequality h1 h_i ≥ (d−i+1)h_{i−1} + (i+1)h_{i+1} (Corollary 1.8). The paper also proves the analog for balanced d-polytopes (Section 5) and gives conditional evidence for the Equator Conjecture: a reduction to minimal flag spheres (Proposition 3.2, Corollary 3.3), verification of the Link Conjecture for boundary complexes obtained from crosspolytopes by edge subdivisions (Proposition 1.4), and a structural conjecture implying the Equator Conjecture (Proposition 3.5), verified in dimension two via an unproved lemma (Lemma 3.6) and a detailed local analysis (Theorem 3.7).

Significance. If the conjectures hold, the paper's framework gives a considerable strengthening of Gal's conjecture. The unconditional inequality of Theorem 1.6/Corollary 1.8 is new and seems nontrivial, and the proof method—first establishing a half-integral perfect matching in the complement graph and then summing shelling inequalities over the cycles of that matching—is elegant. The paper is parameter-free, with no fitted constants, and the main result is a solid contribution to the face-enumeration literature. The equivalence between the Equator and Link Conjectures is a clean conceptual step. The conditional evidence is thought-provoking, but one load-bearing lemma (Lemma 3.6) is asserted without proof, and another step (in Proposition 3.2) needs an explicit justification; these gaps affect the strength of the supporting evidence, not the main unconditional theorem.

major comments (2)
  1. [Section 3, Lemma 3.6] Lemma 3.6 asserts that every flag homology 2-sphere other than the octahedron's boundary contains an edge that is in no induced 4-cycle, but the proof is omitted (the text says 'we omit its simple proof'). This lemma is load-bearing: it is used to conclude that the structural conjecture of Problem 1.5 holds for all flag 2-spheres, and via Proposition 3.5 this is one of the two pillars of the dimension-two evidence for the Equator Conjecture. No derivation from Whiteley's result [19, Lem.6] is supplied, and no independent argument is given. The reader cannot verify this pivotal claim from the manuscript. Please provide a full proof or a precise reference, or state the lemma as an assumption.
  2. [Section 3, Proposition 3.2] In cases (i) and (iii) (the subcase f0(Δ)=f0(Δ1)), the proof concludes γ(Δ') ≤ γ(Δ) by citing Lemma 2.2(ii). That lemma gives γΔ(t) = γΔ'(t) + tγ_{lk_e Δ}(t), so the inequality γ(Δ') ≤ γ(Δ) requires γ(lk_e Δ) ≥ 0 coefficientwise. The induction hypothesis of Proposition 3.2 only assumes Conjecture 1.3 for smaller spheres, not Gal's conjecture. Although this nonnegativity can in principle be derived from the induction hypothesis by induction on dimension (because a vertex link is an equator, and the inequality γ(vertex link) ≤ γ(S) forces the higher γ-coefficients of S to be nonnegative), the manuscript does not state or prove this derivation. Please add this justification explicitly, or state the additional assumption used.
minor comments (4)
  1. [Section 4, Proof of Theorem 1.7] The word 'immdiatly' should be 'immediately'.
  2. [Section 3, Lemma 3.4] The proof of Lemma 3.4 refers to [3, Sec.3] for the case analysis. The cases are sketched, but a fuller self-contained explanation would significantly improve readability and verifiability.
  3. [Section 3, Theorem 3.7] The final paragraph of the proof ('This implies a specific structure on Δ...') is very terse. Expanding this part would help the reader follow the argument that both outcomes (i) and (ii) hold.
  4. [Introduction] The sentence 'It is known and easy that Gal's conjecture reduces to proving it for all Δ ∈ R (see Lemma 2.2)' would benefit from a more explicit indication of how Lemma 2.2(ii) is iterated to perform that reduction.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the main inequalities follow from independent shelling and matching arguments, and the conditional equivalences are genuine logical implications.

full rationale

The paper's strongest proved results, Theorem 1.6 and Corollary 1.8, are derived by combining an independent line-shelling inequality (Lemma 4.1) with a graph-theoretic half-integral perfect matching theorem (Theorem 1.7), whose proof uses Lemma 4.2 and a standard external criterion from Scheinerman–Ullman (Lemma 4.3). No parameter is fitted and no prediction is fed back into the derivation; the h-vector inequality (2) follows by substituting McMullen's formula, cited from Swartz [17], which is an established external identity. The conditional part of the paper is also non-circular: Proposition 3.1 proves equivalence between the Link and Equator conjectures via the explicit f-polynomial identity (3), and Proposition 3.5 derives the Equator Conjecture from the structural Problem 1.5 by a written induction that invokes the equivalence, not by assuming the target. Self-citations such as [9, Lem.3.4] and [10, Conj.6.1] are used for standard or previously stated facts, and the load-bearing reductions are proved in the text rather than resting solely on those citations. The one genuine gap is Lemma 3.6, whose proof is explicitly omitted ('we omit its simple proof') and which supports the dimension-two structural evidence in Theorem 3.7; however, an omitted proof is a verifiability or rigor concern, not circularity, because the lemma's statement is not defined in terms of the Equator Conjecture, nor is any fitted quantity renamed as a prediction. Accordingly, no circular step is exhibited, and the appropriate circularity score is 0.

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

No free parameters are introduced. The results are proved from standard facts in face enumeration, shelling theory, and graph theory. The only unproved input is Lemma 3.6, whose proof is explicitly omitted, plus cited standard results listed above.

assumptions (6)
  • standard math Dehn-Sommerville relations hold for flag homology spheres, allowing h-vectors to be expressed in the gamma-basis.
    Invoked in Section 2.2 to define gamma-vectors for flag homology spheres.
  • standard math Gamma-vectors behave according to Lemma 2.2: suspension preserves gamma, and edge contraction gives gamma(Delta) = gamma(Delta') + t gamma(link_e(Delta)).
    Cited from Gal [6] and used in Propositions 3.1, 3.2, and 1.4.
  • standard math McMullen's formula: sum over vertices v of h_{i-1}(link_v(Delta)) = i h_i(Delta) + (d-i+1) h_{i-1}(Delta) for polytopes.
    Used to derive Corollary 1.8 from Theorem 1.6; cited from Swartz [17, Prop.2.3].
  • standard math A graph has a half-integral perfect matching if and only if for every X, G-X has at most |X| isolated vertices.
    Used in the proof of Theorem 1.7; cited from Scheinerman and Ullman [14, Thm 2.2.4].
  • domain assumption For every nonedge uv of a simplicial d-polytope, a line shelling can be chosen that shells facets containing v first and facets containing u last.
    Foundation of Lemma 4.1 and the proof of Theorem 1.6; standard shelling geometry for polytopes.
  • ad hoc to paper Lemma 3.6: every flag homology 2-sphere other than the octahedron's boundary has an edge contained in no induced 4-cycle.
    Asserted in Section 3 with proof omitted; supports Theorem 3.7 and the dimension-two evidence for the Equator Conjecture.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Induced equators in flag spheres." pith.science (2026). https://pith.science/paper/6OTKNJDD

@misc{pith2026190808727,
  author       = {Pith},
  title        = {Pith review of: Induced equators in flag spheres},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6OTKNJDD}},
  note         = {Machine review of arXiv:1908.08727}
}
abstract

We propose a combinatorial approach to the following strengthening of Gal's conjecture: $\gamma(\Delta)\ge \gamma(E)$ coefficientwise, where $\Delta$ is a flag homology sphere and $E\subseteq \Delta$ an induced homology sphere of codimension $1$. We provide partial evidence in favor of this approach, and prove a nontrivial nonlinear inequality that follows from the above conjecture, for boundary complexes of flag $d$-polytopes: $h_1(\Delta) h_i(\Delta) \ge (d-i+1)h_{i-1}(\Delta) + (i+1) h_{i+1}(\Delta)$ for all $0\le i\le d$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 20 canonical work pages

  1. [1]

    Upper bound theorem for odd-dimensional flag triangulations of manifolds

    Micha/suppress l Adamaszek and Jan Hladk´ y. Upper bound theorem for odd-dimensional flag triangulations of manifolds. Mathematika, 62(3):909–928, 2016

  2. [2]

    Adiprasito

    Karim A. Adiprasito. Combinatorial lefschetz theorems beyond p ositivity. math arXiv:1812.10454, v1,v2, Dec.2018, Jan.2019

  3. [3]

    gamma-vectors of edge subdivisions of the boundary of the cross polytope

    Natalie Aisbett. gamma-vectors of edge subdivisions of the boun dary of the cross polytope. arXiv:1209.1789, 2012

  4. [4]

    Athanasiadis

    Christos A. Athanasiadis. Flag subdivisions and γ-vectors. Pacific J. Math. , 259(2):257–278, 2012

  5. [5]

    The Euler characteristic of a no npositively curved, piecewise Eu- clidean manifold

    Ruth Charney and Michael Davis. The Euler characteristic of a no npositively curved, piecewise Eu- clidean manifold. Pacific J. Math. , 171(1):117–137, 1995

  6. [6]

    ´Swiatos/suppress law R. Gal. Real root conjecture fails for five- and higher-dimensional spheres. Discrete Com- put. Geom., 34(2):269–284, 2005

  7. [7]

    Springer-Verlag, New York, second edition, 2003

    Branko Gr¨ unbaum.Convex polytopes, volume 221 of Graduate Texts in Mathematics . Springer-Verlag, New York, second edition, 2003. Prepared and with a preface by Volker Kaibel, Victor Klee and G¨ unter M. Ziegler

  8. [8]

    Face enumeration on simplicial comp lexes

    Steven Klee and Isabella Novik. Face enumeration on simplicial comp lexes. In Recent trends in com- binatorics, volume 159 of IMA Vol. Math. Appl. , pages 653–686. Springer, [Cham], 2016

Show all 22 references
  1. [9]

    Bounds for entries ofγ-vectors of flag homology spheres

    Jean-Philippe Labb´ e and Eran Nevo. Bounds for entries ofγ-vectors of flag homology spheres. SIAM J. Discrete Math. , 31(3):2064–2078, 2017

  2. [10]

    Lutz and Eran Nevo

    Frank H. Lutz and Eran Nevo. Stellar theory for flag complexes . Math. Scand., 118(1):70–82, 2016

  3. [11]

    McMullen

    P. McMullen. The numbers of faces of simplicial polytopes. Israel J. Math. , 9:559–570, 1971

  4. [12]

    Domination numbers and homology

    Roy Meshulam. Domination numbers and homology. J. Combin. Theory Ser. A , 102(2):321–330, 2003

  5. [13]

    Kyle Petersen

    Eran Nevo and T. Kyle Petersen. On γ-vectors satisfying the Kruskal-Katona inequalities. Discrete Comput. Geom., 45(3):503–521, 2011

  6. [14]

    Scheinerman and Daniel H

    Edward R. Scheinerman and Daniel H. Ullman. Fractional graph theory . Dover Publications, Inc., Mineola, NY, 2011. A rational approach to the theory of graphs, W ith a foreword by Claude Berge, Reprint of the 1997 original

  7. [15]

    Richard P. Stanley. The number of faces of a simplicial convex polytope. Adv. in Math., 35(3):236–238, 1980. 12 MARIA CHUDNOVSKY AND ERAN NEVO

  8. [16]

    Richard P. Stanley. Combinatorics and Commutative Algebra . Birkh¨ auser, 2nd edition, 1996

  9. [17]

    g-elements, finite buildings and higher Cohen-Macaulay connectivity

    Ed Swartz. g-elements, finite buildings and higher Cohen-Macaulay connectivity. J. Combin. Theory Ser. A , 113(7):1305–1320, 2006

  10. [18]

    V. D. Volodin. Geometric realization of γ-vectors of 2-truncated cubes. Uspekhi Mat. Nauk , 67(3(405)):181–182, 2012

  11. [19]

    Vertex splitting in isostatic frameworks

    Walter Whiteley. Vertex splitting in isostatic frameworks. Struc. Top., 16:23–30, 1989

  12. [20]

    The flag upper bound theorem for 3- and 5-manifo lds

    Hailun Zheng. The flag upper bound theorem for 3- and 5-manifo lds. Israel J. Math. , 222(1):401–419, 2017

  13. [21]

    Face enumeration on flag complexes and flag spher es

    Hailun Zheng. Face enumeration on flag complexes and flag spher es. arXiv:1809.06835, 2018

  14. [22]

    Ziegler.Lectures on polytopes , volume 152 of Graduate Texts in Mathematics

    G¨ unter M. Ziegler.Lectures on polytopes , volume 152 of Graduate Texts in Mathematics . Springer- Verlag, New York, 1995

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.