Pith. sign in

REVIEW 3 major objections 5 minor 19 references

On Ehrhart theory for tropical vector bundles

T0 review · 3 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read For a tropical vector bundle on a toric variety, the equivariant Euler characteristic at a character u equals the value at u of an associated convex chain α_E, yielding a combinatorial Hirzebruch–Riemann–Roch formula.

desk verdict The main theorem is real — the stress-test counterexample to Proposition 4.10 drops a cone, so it does not land; the real soft spots are an asserted refinement step and a garbled involution proof in §5. read the letter →

arxiv 2603.05292 v2 pith:V7Q2NGRY submitted 2026-03-05 math.AG

Suhyon Chong , Kiumars Kaveh This is my paper · ORCID
classification math.AG MSC 14T0514M2552B2014C40
keywords tropicalvectorbundlestoricvarietiesEhrharttheoryconvexchainsEulercharacteristicHirzebruch–Riemann–RochmatroidsBergmanfan
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 a precise equality for tropical vector bundles on toric varieties: the equivariant Euler characteristic at a character u equals the value at u of a convex chain α_E built from the bundle's equivariant Chern roots. Because convex chains are governed by Khovanskii–Pukhlikov's Ehrhart theory, this equality yields a combinatorial Hirzebruch–Riemann–Roch formula, in which a Todd-class differential operator applied to the integral of α_E recovers the Euler characteristic. The paper also proves that the tautological bundle of any matroid has equal Euler characteristic and global-section rank, interpreted as vanishing of higher cohomology. The key supporting steps are a split-bundle resolution for tropical vector bundles and invariance of the Euler characteristic under fan refinements.

What carries the argument

Khovanskii–Pukhlikov theory of convex chains: a convex chain is a finite integer combination Σ n_i 1_{P_i} of indicator functions of polytopes; its lattice sum S(α) and integral I(α) generalize the Ehrhart polynomial and volume. The paper encodes a tropical vector bundle E, defined by a piecewise linear map Φ_E : |Σ| → the lifted Bergman fan of a matroid, by a multi-valued support function h_E(x) = Σ_i [⟨u_{σ,i}, x⟩] on each maximal cone σ. This support function determines the convex chain α_E. The Khovanskii–Pukhlikov theorem, an exact Euler–Maclaurin formula sometimes called a multidimensional Riemann–Roch for polytopes, connects S and I, giving the combinatorial HRR formula. The proof als

What would settle it

Compute the equivariant Euler characteristic directly from its alternating-sum definition on any fan, and compute α_E(u) for the same tropical vector bundle; a disagreement on a single example would refute Theorem 1.1. For instance, take the Fano-plane bundle of Example 4.15, where h_1 is non-convex, refine the fan to linearize h_1, and compare the two computations.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.1: for a tropical vector bundle E on X_Σ, the convex chain α_E — constructed from the equivariant Chern roots of E — satisfies χ(X_Σ,E)_u = α_E(u) for every character u. The chain is derived from a multi-valued support function h_E(x) = Σ_i [⟨u_{σ,i}, x⟩] on each maximal cone σ. The proof first shows the Euler characteristic is invariant under refinements of the fan (Theorem 4.11), then passes to a refinement where the sorted Chern-root functions h_i are linear on each cone, so each h_i is the support function of a virtual polytope; Brianchon–Gram then sums the contributions to α_E. Combined with the Khovanskii–Pukhlikov theorem, this yields the combinatorial H

Load-bearing premise

The proof that the Euler characteristic matches the convex chain rests on the claim that after refining the fan, the sorted Chern-root functions become linear on every cone; the paper asserts this refinement exists but does not prove it.

Editorial extensions

If this is right

  • The Euler characteristic of a tropical vector bundle can be computed as S(α_E), a lattice-point count, making it accessible to Ehrhart-type algorithms.
  • The combinatorial HRR formula gives a Todd-correction to the volume of α_E that reproduces the Euler characteristic, a concrete analogue of the classical Hirzebruch–Riemann–Roch theorem.
  • Theorem 1.2 shows the equivariant Euler characteristic is invariant under passing to a refinement of the fan, so it is a well-defined invariant of the tropical bundle independent of the chosen fan presentation.
  • Theorem 5.7 shows that for matroid tautological bundles, higher cohomology vanishes in the only sense currently available, matching the representable case.
  • The split-resolution theorem extends Klyachko's K-theoretic decomposition from toric vector bundles to the tropical setting, providing a new tool for studying tropical bundles.

Reading between the lines

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

  • The unproved linearization step — that the sorted Chern-root functions can be made simultaneously linear on a refinement — is a concrete geometric property; testing it could either complete the proof or reveal counterexamples for more general piecewise linear maps.
  • The convex-chain encoding suggests that tropical vector bundles may admit an Ehrhart theory with multiplicities beyond matroids, possibly connecting to valuated matroids and matroid polytope invariants.
  • If the equality χ = h^0 holds for tautological bundles without a constructed cohomology theory, it raises the question of whether a derived category or homological algebra for tropical bundles can be defined so that the equality becomes a genuine vanishing theorem.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper develops an Ehrhart-type theory for tropical vector bundles on complete toric varieties. It associates to a tropical vector bundle E a convex chain α_E and claims that its evaluation at a character u equals the equivariant Euler characteristic χ(X_Σ,E)_u (Theorem 1.1). From this it derives a Khovanskii–Pukhlikov-style Hirzebruch–Riemann–Roch formula (Corollary 1.3). The proof strategy is to establish invariance of the combinatorial Euler characteristic under fan refinement (Theorem 4.11), then refine the fan so that the order-statistic functions h_i are linear and apply Brianchon–Gram (Theorem 4.13). The paper also gives a split-resolution analogue for tropical vector bundles (Section 4.2) and proves that for the tautological matroid bundle E_M one has χ(X_m,E_M)_u = h^0(X_m,E_M)_u for all characters u (Theorem 5.7), interpreted as vanishing of higher cohomologies.

Significance. If its main theorem is correct, the paper gives a clean combinatorial formula for the equivariant Euler characteristic of tropical vector bundles and answers a question of Kaveh–Manon for tautological matroid bundles. The convex-chain viewpoint is natural, the split-resolution construction is potentially useful, and the Fano-plane and U_{2,3} examples are welcome. However, the central proof has a serious gap: Proposition 4.10, which underpins Theorem 4.11 and hence Theorem 4.13, is false as stated. Because the main theorem depends on this unproved/false step, the paper is not yet in publishable form, although the overall approach may be salvageable.

major comments (3)
  1. [§4.3, Proposition 4.10, Eq. (3)] The claimed refinement identity is false. Take N=R^2, σ=cone(e1,e2), and refine by the ray δ=cone(e1+e2). Then φ^{-1}(σ)={σ1,σ2} with σ1=cone(e1,δ) and σ2=cone(δ,e2), all of codimension 0. Equation (3) becomes 1_{σ^∨}=1_{σ1^∨}+1_{σ2^∨}. At u=(1,1), both σ1^∨ and σ2^∨ contain u, so the right-hand side equals 2 while the left-hand side equals 1. The 'Möbius inversion' step is invalid because the functions 1_{τ^∨} are not ordered with the appropriate triangularity: for τ⊂σ one has τ^∨⊃σ^∨. Since Theorem 4.11 is proved directly from this identity, and Theorem 4.13 explicitly invokes Theorem 4.11 to replace Σ by a refinement, the proof of Theorem 1.1 collapses at this point.
  2. [§4.3, proof of Theorem 4.13] The proof asserts that 'after replacing Σ with a refinement of Σ, we can assume that the h_i are linear on each cone of Σ'. This simultaneous linearization of the order-statistic functions is not proved; it is plausible via a common refinement of hyperplane arrangements, but as written it is an unproved assumption. More importantly, the replacement of Σ uses Theorem 4.11, whose proof is invalid because of the false Proposition 4.10. Thus the reduction to a sum of line-bundle contributions and the application of Brianchon–Gram are not justified. The authors need either a correct proof of refinement invariance or a direct argument for Theorem 4.13.
  3. [§5.2, proof of Theorem 5.7] The involution argument proving cancellation in Corollary 5.5 is not rigorous as written. The condition 'if |S_{k-1} \ S_k|>1' is impossible because S_{k-1}⊂S_k; presumably |S_k \ S_{k-1}| was intended. More seriously, the text does not verify that the proposed map preserves the defining property of Π_u(S), namely that the distinguished flat is the first one at which ⟨u,e_{S_t}⟩=1, nor that the map is a sign-reversing involution. The coefficient computation for rank(S) is therefore asserted rather than demonstrated. Since this is the proof of Theorem 1.4, the vanishing statement is not established as written.
minor comments (5)
  1. [Global] There are numerous typos and formatting issues: the title is given as 'EHRHART THEOR Y', 'Gröbner' is misspelled, and 'fGF(M)' is used without a consistent definition in the notation list.
  2. [Remark 1.5] The paper explicitly notes that no higher cohomology functors are defined for tropical vector bundles. The phrase 'vanishing of higher cohomologies' in the abstract and Theorem 1.4 should therefore be understood only as the equality χ=h^0; this caveat should be stated in the main theorem statement itself, not only in a remark.
  3. [§4.2, Proposition 4.2] The compatibility proof for the split bundles F^k is hard to follow: it does not explicitly define the direct-sum basis for the total space or verify the compatibility condition on overlaps of cones. This is not central to Theorem 1.1, but a clearer proof would improve the paper.
  4. [§4.3, Theorem 4.3] In the proof, the statement that the exponential sum is determined by its values on the rays of τ is justified by smoothness, but it only requires that τ is a full-dimensional cone generated by its rays; smoothness is not needed. The reasoning should be rephrased.
  5. [References] The reference [CHK] is incomplete ('arXiv:' with no number). Also, the manuscript lacks page numbers in the citation to [KM25], which should be supplied.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Theorem 1.1 is an independent computed identity, and the paper's self-citations are imported tools rather than load-bearing conclusions.

full rationale

The central claim χ(XΣ,E)_u = α_E(u) (Theorem 1.1/4.13) is not circular. χ is defined directly from affine-chart section ranks as an alternating sum (Definition 2.20), while α_E is defined separately from the multi-valued support function of the bundle's character data (Definition 4.1) via the Khovanskii–Pukhlikov support-function isomorphism (Theorem 3.7). The proof connects the two by Lemma 4.7 (rewriting the rank on U_σ as a sum of dual-cone indicators), a fan refinement to linearize the sorted characters, and Brianchon–Gram for virtual polytopes. Neither side is fitted to the other, and Definition 2.20 does not mention α_E. The tautological-bundle theorem (Theorem 5.7) is likewise a combinatorial proof of an equality between the alternating sum in Corollary 5.5 and the explicitly computed h^0, not a renaming of h^0 as χ; Remark 1.5 appropriately concedes that 'vanishing of higher cohomologies' is only an interpretation. The paper imports definitions and theorems from [KM], [KhM], [KM25], [Klyachko89], and [KhP92a, KhP92b], but always as tools or prior classification results, never as the grounds for Theorems 1.1 or 1.4; the self-citation [CHK] in Remark 4.12 merely records that a toric-vector-bundle version of refinement invariance was known. The unproved refinement assertion in the proof of Theorem 4.13 is a mathematical robustness gap (if the claimed simultaneous linearization or the refinement-invariance formula fails), not a circular reduction, since it does not assume χ=α_E. No fitted parameter is relabeled as a prediction, and no uniqueness theorem is invoked to force the authors' choices.

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

No free parameters are fitted anywhere; the paper proves parameter-free combinatorial identities. The main imported ingredients are prior theorems and definitions. The only ad hoc assumption introduced by this paper is the unproved refinement linearization. No new postulated entities are introduced.

assumptions (5)
  • standard math Klyachko classification of toric vector bundles by compatible filtrations (Theorem 2.2, [Klyachko89])
    Background theorem motivating the definition of tropical vector bundles and used for the split-resolution construction.
  • domain assumption Equivalence of Klyachko-data and piecewise-linear-map definitions of tropical vector bundles (Proposition 2.14, [KM])
    The paper adopts the [KM] definition; all subsequent constructions presuppose it.
  • standard math Khovanskii–Pukhlikov theorems: convex-chain algebra isomorphism and Todd/lattice-sum relation (Theorems 3.7 and 3.11)
    Imported as black boxes; the HRR corollary and the interpretation of χ as a lattice sum depend on them.
  • ad hoc to paper Existence of a fan refinement making each i-th-minimum function h_i linear on every cone (proof of Theorem 4.13)
    Asserted without proof; the reduction to split line bundles relies on it.
  • domain assumption Construction of the tautological bundle E_M via the projection |ĝΣ_m| → ĝBerg(M) (Section 5.1, from [KM])
    Section 5's main theorem is about this bundle and its diagram.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On Ehrhart theory for tropical vector bundles." pith.science (2026). https://pith.science/paper/V7Q2NGRY

@misc{pith2026260305292,
  author       = {Pith},
  title        = {Pith review of: On Ehrhart theory for tropical vector bundles},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V7Q2NGRY}},
  note         = {Machine review of arXiv:2603.05292}
}
read the original abstract

The notion of a tropical vector bundle on a toric variety was recently introduced by Khan-Maclagan and Kaveh-Manon. In this paper, we study the Euler characteristic and rank of global sections for tropical vector bundles. We associate a convex chain (a finite integer linear combination of indicator functions of convex polytopes) to a tropical vector bundle encoding its Euler characteristic. We then see that the Khovanskii-Pukhlikov theory of convex chains gives a combinatorial Hirzebruch-Riemann-Roch theorem for tropical vector bundles. This, in particular, applies to toric vector bundles. Also, we extend Klyachko's resolution of a toric vector bundle by split toric vector bundles to tropical vector bundles. As shown by Kaveh-Manon, every matroid comes with a tautological tropical vector bundle. We answer positively a question posed by Kaveh-Manon about equality of Euler characteristic with rank of space of global sections (in other words, vanishing of higher cohomologies) for the tautological bundle of a matroid.

Figures

Figures reproduced from arXiv: 2603.05292 by the authors.

Figure 1
Figure 1. The Fano plane. Let ρ1, ρ2, ρ3 be the rays of the fan of P 2 with ray generators {v1, v2, v3} where v1, v2 are the standard basis of N and v3 = −v1 − v2. To describe a tropical vector bundle over P 2 , it 18 [PITH_FULL_IMAGE:figures/full_fig_p018_1.png] view at source ↗
Figure 2
Figure 2. The parliament of polytopes for E. We compute X u∈M h 0 (u) = 27 by the definition h 0 (u) = rank{e ∈ G : u ∈ Pe}. To compute αE , let {e1, e2} be the standard basis of M. For the full dimensional cones, σ1 = cone(v2, v3), σ2 = cone(v1, v3), σ3 = cone(v1, v2), the multi-sets of characters are u(σ1) = {−2e1+2e2, −2e1, −2e1+e2}, u(σ2) = {2e1−2e2, −2e2, e1−2e2}, and u(σ3) = {2e1, 2e2, e1+e2}. For x = (x1, x2) ∈ NR, hE … view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 2 linked inside Pith

  1. [1]

    Chern classes of tropical vector bundles , Ark

    Allermann, L. Chern classes of tropical vector bundles , Ark. Mat. 50 (2012), no. 2, 237--258

    2012

  2. [2]

    Tautological classes of matroids

    Berget, A.; Eur, C.; Spink, H.; Tseng, D. Tautological classes of matroids . Invent. Math.233(2023), no.2, 951--1039

    2023

  3. [3]

    Formulas of Brion, Lawrence, and Varchenko on rational generating functions for cones

    Beck, M.; Haase, C.; Sottile, F. Formulas of Brion, Lawrence, and Varchenko on rational generating functions for cones . Math. Intelligencer 31 (2009), no. 1, 9–17

    2009

  4. [4]

    Note on the Euler characteristic of a toric vector bundle

    Chong, S.; Huang, S.; Kaveh, K. Note on the Euler characteristic of a toric vector bundle . arXiv:

  5. [5]

    A.; Little, J

    Cox, D. A.; Little, J. B.; Schenck, H. K. Toric varieties . Vol. 124. American Mathematical Society (2024)

    2024

  6. [6]

    Stellahedral geometry of matroids

    Eur, C.; Huh, J.; Larson, M. Stellahedral geometry of matroids . Forum Math. Pi 11 (2023), Paper No. e24, 48

    2023

  7. [7]

    Cohomologies of tautological bundles of matroids

    Eur, C. Cohomologies of tautological bundles of matroids . Selecta Math. (N.S.) 30 (2024), no.5, Paper No. 85, 19

    2024

  8. [8]

    Introduction to Toric Varieties

    Fulton, W. Introduction to Toric Varieties . Vol. 131, Princeton University Press, (1993)

    1993

Show all 19 references
  1. [9]

    Principal bundles on metric graphs: the _n case , Adv

    Gross, A.; Ulirsch, M.; Zakharov, D. Principal bundles on metric graphs: the _n case , Adv. Math. 411 (2022), Paper No. 108775, 45

    2022

  2. [10]

    Vector bundles on tropical schemes

    Jun, J.; Mincheva, K.; Jeffrey Tolliver, J. Vector bundles on tropical schemes . Journal of Algebra 637 (2024), Pages 1--46

    2024

  3. [11]

    On equivariant vector bundles on an almost homogeneous variety

    Kaneyama, T. On equivariant vector bundles on an almost homogeneous variety . Nagoya Math. J. 57 (1975), 65--86

    1975

  4. [12]

    Toric vector bundles, valuations and tropical geometry

    Kaveh, K.; Manon, C. Toric vector bundles, valuations and tropical geometry. Algebras and Representation Theory (2025): 1-29

    2025

  5. [13]

    Tropical vector bundles and matroids

    Kaveh, K.; Manon, C. Tropical vector bundles and matroids . arXiv:2405.03576

    arXiv

  6. [14]

    Tropical vector bundles

    Khan, B.; Maclagan, D. Tropical vector bundles . arXiv:2405.03505

    arXiv

  7. [15]

    Algebra i Analiz 4 (1992), no

    Khovanskii, A.G.; Pukhlikov, A.V.; Finitely additive measures of virtual polytopes . Algebra i Analiz 4 (1992), no. 2, 161--185; English transl. in St. Petersburg Math. J. 4 (1993)

    1992

  8. [16]

    A Riemann-Roch theorem for integrals and sums of quasipolynomials over virtual polytopes

    Khovanskii, A.G.; Pukhlikov, A.V. A Riemann-Roch theorem for integrals and sums of quasipolynomials over virtual polytopes . Algebra i Analiz 4 (1992), no. 4, 188–216. English transl. in St. Petersburg Math. J. 4 (1993)

    1992

  9. [17]

    Klyachko, A. A. Equivariant bundles on toral varieties . (Russian) Izv. Akad. Nauk SSSR Ser. Mat. 53 (1989), no. 5, 1001--1039, 1135; translation in Math. USSR-Izv. 35 (1990), no. 2, 337--375

    1989

  10. [18]

    The polytope algebra

    McMullen, P. The polytope algebra . Adv. Math. 78(1) (1989), 76--130

    1989

  11. [19]

    Local tropical linear spaces

    Rinc\' o n, F. Local tropical linear spaces. Discrete Comput. Geom.50(2013), no.3, 700--713

    2013

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.