A Sharp Curvature Threshold for GLMY Path Homology

Jingyan Li; Shing-Tung Yau; Shuliang Bai

A Sharp Curvature Threshold for GLMY Path Homology

T0 review · 0 major / 3 minor · reviewed 2026-08-28 · deepseek-v4-flash

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

Pith's one-line read A finite graph with all edge curvatures above one half has vanishing first GLMY path homology, and one half is the best possible threshold.

desk verdict Sharp 1/2 curvature threshold for first GLMY path homology is real, well-proved, and likely citable; minor caveats are external premises and an enumerated LP check. read the letter →

arxiv 2608.23187 v1 pith:KIV7QKY2 submitted 2026-08-24 math.AT math.CO

Shuliang Bai , Jingyan Li , Shing-Tung Yau This is my paper · ORCID

classification math.ATmath.CO

keywords curvatureglmyhomologypathedgeeveryfinitelin--lu--yau

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

Graph curvature is a discrete version of Ricci curvature: it measures, for each edge, how much the random-walk neighborhoods of the two endpoints pull toward each other. GLMY path homology is an algebraic way to detect holes in a graph, defined through chains of allowed directed paths. Previous work showed that one kind of curvature, the Bakry-Emery type, forces the first hole group to vanish when positive, but for the Lin-Lu-Yau curvature the five-cycle C5 has positive curvature and still has a one-dimensional hole. This paper finds the exact boundary: if every edge has curvature greater than 1/2, the first hole group over the reals is automatically zero. The number 1/2 is optimal, because C5 sits exactly at the boundary.

The proof works by assuming a nonzero hole class exists. It picks a circulation that is orthogonal to all triangle and square relations, normalizes it, and chooses an edge where the circulation is maximal. On the two endpoint neighborhoods this circulation can be integrated into a local height function, because triangle and square relations make the edge differences consistent. A compression step and the McShane extension produce a global test function, and the variational formula for Lin-Lu-Yau curvature then forces the curvature of that edge to be at most 1/2. This contradicts the hypothesis that all edges have curvature above 1/2.

The paper also shows that positive curvature implies the fundamental group of the complex obtained by filling all cycles of length at most five is finite, with an explicit bound. In higher degrees, strict positivity does not force vanishing: Cartesian powers of C5 have positive curvature and nonzero homology in every degree up to the number of factors.

Extended reading notes

Core claim

Theorem 1.1: If G is a finite simple graph with at least one edge and kappa_min^LLY(G) > 1/2, then H_1^GLMY(G;R) = 0. Equivalently, nonzero first GLMY path homology forces an edge of Lin-Lu-Yau curvature at most 1/2. The constant is optimal and is attained by C5. If the paper is correct, positivity of all edge curvatures beyond one half is a complete certificate for vanishing of first GLMY path homology over the reals.

Load-bearing premise

Lemma 2.3, the Kempton-Muench-Yau cycle-space presentation H_1^GLMY(G;F) being isomorphic to C(G;F)/C_{≤4}(G;F), is the load-bearing external premise: the proof of Theorem 1.1 begins by choosing a nonzero circulation orthogonal to the 3- and 4-cycle relations inside that quotient. If that identification failed, there would be no cycle-space object to extract the local potential from. The Muench-Wojciechowski Laplacian characterization (equation 6) is a second external premise: it converts curvature into a test-function infimum that the constructed function is plugged into. Both are prior published theorems rather than statements proved in this paper.

Share X Bluesky LinkedIn Reddit HN

Formalized claims in Lean

Claim #1: Theorem 1.1: If G is a finite simple graph with at least one edge and kappa_min^LLY(G) > 1/2, then H_1^GLMY(G;R) = 0. Equivalently, nonzero first GLMY path homology forces an edge of Lin-Lu-Yau curvature at most 1/2. The constant is optimal and is attained by C5. If the paper is correct, positivity of all edge curvatures beyond one half is a complete certificate for vanishing of first GLMY path ho

/-- @claim 1 Theorem 1.1: If G is a finite simple graph with at least one edge and kappa_min^LLY(G) > 1/2, then H_1^GLMY(G;R) = 0. Equivalently, nonzero first GLMY path homology forces an edge of Lin-Lu-Yau curvature at most 1/2. The constant is optimal and is attained by C5. If the paper is correct, positivity of all edge curvatures beyond one half is a complete certificate for vanishing of first GLMY path ho -/
def central_claim : Prop :=

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

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

The paper you want to know about: Bai, Li, and Yau prove that a finite simple graph with Lin–Lu–Yau curvature strictly greater than 1/2 on every edge has vanishing first GLMY path homology over R (and char 0), and the constant is sharp, attained by C5. That is a clean, genuinely new theorem. The proof is the real content: it chooses a nonzero circulation orthogonal to the 3- and 4-cycle relations, builds a local potential on the two endpoint neighborhoods, compresses it with a piecewise-linear map, extends by McShane, and feeds it into the Münch–Wojciechowski Laplacian formula. The edge-difference identity and the halving from compression are the mechanism; the estimate comes out at 1/2 naturally. I checked the case analysis in Section 3 and it holds; no missing adjacency cases.

The paper also does two useful side things. The fundamental-group consequence—positive curvature forces π_1(X_{≤5}(G)) finite, with an explicit bound via the universal-cover diameter argument—is a nice application of the Hehl–Münch local-cover mechanism, though it is a fairly straightforward specialization. And the C5 products T_r = C5^{□r} give curvature 1/(2r) with H_p = F^{binom(r,p)}, which correctly shows positivity does not force vanishing in higher degrees and sets up an interesting extremal problem.

Where are the soft spots? They are minor and mostly external. The proof leans on two prior theorems: the Kempton–Münch–Yau cycle-space presentation of H1 and the Münch–Wojciechowski limit-free Laplacian formula. Both are published, but not reproved here; if either had hidden conditions, the threshold theorem would inherit them. I do not think that is a real problem—the stress test checked the 4-cycle edge case in Lemma 2.3 and it is fine. The G7 example in Section 5.2 gives explicit maximizing functions, but the optimality claim rests on a total-unimodularity plus finite enumeration argument that is described rather than fully itemized. That is acceptable, though a referee might ask for the enumeration script or a more formal certificate. The characteristic-zero caveat is stated honestly.

Overall: the central result is new, the proof is coherent, the examples are consistent, and the citation pattern is not circular—the prior papers with overlapping authors do not contain the 1/2 threshold. This is a paper for people working on graph curvature and GLMY path homology; it deserves a serious referee. I would send it to review, and I would likely accept after minor revision asking for a bit more detail on the G7 enumeration.

Referee Report

0 major / 3 minor

Summary. This paper establishes a sharp curvature threshold for vanishing of first GLMY path homology of finite simple graphs. Theorem 1.1 states that if the minimum Lin–Lu–Yau edge curvature exceeds 1/2 then H_1^GLMY(G;R)=0, equivalently every nontrivial first homology class forces an edge of curvature at most 1/2; the bound is optimal, with C5 providing equality and nonzero homology. The proof combines the Kempton–Münch–Yau presentation of H_1 as the cycle space modulo simple 3- and 4-cycle relations with the Münch–Wojciechowski variational formula for Lin–Lu–Yau curvature: it selects a circulation orthogonal to the short-cycle relations, constructs a local potential on the union of endpoint neighborhoods, compresses and McShane-extends it, and bounds the Laplacian difference by 1/2. The paper also proves a fundamental-group corollary: for connected G with positive minimum curvature, filling all simple cycles of length at most five yields a finite π_1 with an explicit order bound, because the curvature is preserved by 5-cycle-preserving covers and Bonnet–Myers bounds the diameter of the universal cover's one-skeleton. It closes with examples: a positively curved seven-vertex graph with infinite nonabelian GLMY fundamental group, and Cartesian powers of C5 showing that positive curvature does not force vanishing of higher-dimensional GLMY path homology.

Significance. The main theorem is a clean, parameter-free result: it gives a complete certificate for vanishing of first GLMY path homology over R and identifies the exact threshold. The proof is written in full, with a self-contained local-potential construction; I verified the case split in equation (13) and found no missing adjacency configuration. The sharpness example and the higher-degree product examples are explicit and make the scope of the theorem precise. The fundamental-group corollary is a substantive application of the curvature-preserving cover mechanism. The paper does not ship machine-checked proofs, but the finite curvature computation for G7 is supported by explicit certificate vectors and a total-unimodularity argument. Two external premises—Lemma 2.3 and equation (6)—are prior published theorems, and the new argument is not a restatement of them. The work is likely to be of interest to researchers in graph curvature and path homology.

minor comments (3)

[§5.2, Proposition 5.1] The exhaustive finite enumeration that is used to certify optimality of the table of av_x g - av_y g values is only described, not documented; please provide the enumerating code, a complete list of integral vertex labelings, or an explicit LP-duality certificate so the exact-rational claim is independently checkable.
[§5.3, Theorem 5.2] The computation P_{C5}(t;F)=1+t is cited to [10] without a theorem or page number; a precise pointer would help readers verify the base case of the Künneth iteration.
[§4.1, Lemma 4.3] In Lemma 4.3 the notation W=AabaB for an immediate reversal is informal; a short formal definition of the segments A and B would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No load-bearing circularity: the one-half threshold proof is a genuine consequence of prior structural theorems, not an input restatement.

full rationale

The derivation chain for Theorem 1.1 is not circular. The proof assumes nonzero H1, uses Lemma 2.3 (Kempton–Münch–Yau) only to obtain the cycle-space presentation C/C≤4, chooses an orthogonal representative ω, and constructs the local potential F via equations (9)–(13). Those relations are exactly the 3- and 4-cycle orthogonality conditions supplied by the cited lemma; they do not encode the target curvature bound. The curvature inequality (20) is then obtained by evaluating the independent Münch–Wojciechowski limit-free Laplacian formula (6) on a compressed McShane extension of F. No parameter is fitted and no quantity defined in terms of the conclusion is used to establish the conclusion. The cited results [2], [3], [9], [10] share authors with this paper, but each is a prior published structural theorem whose assumptions do not include the 1/2 threshold; under the stated criteria they count as independent evidence. Sharpness and higher-degree examples are separately computed (Section 5), so the paper is self-contained against its claimed benchmarks. No significant circularity found.

Assumptions & free parameters 0 free parameters · 9 assumptions · 0 invented entities

No free parameters are fitted: the constants 1/2, 5, and 2/kappa_0 are forced by the proof, not chosen from data. The paper relies on prior domain theorems (some sharing authors) for the cycle-space presentation, the Laplacian characterization, curvature product formula, Kueneth theorem, and abelianization; these are external established results, not consequences of the target theorem. No new entities are postulated.

assumptions (9)

domain assumption Kempton-Muench-Yau cycle-space presentation: H_1^GLMY(G;F) is isomorphic to C(G;F)/C_{≤4}(G;F) for finite undirected graphs (Lemma 2.3).
Invoked at the start of the proof of Theorem 1.1; without it the chosen orthogonal circulation has no connection to first GLMY path homology.
domain assumption Muench-Wojciechowski limit-free Laplacian characterization of LLY curvature (equation 6).
Turns curvature into an infimum over 1-Lipschitz test functions; the constructed test function is inserted here to obtain the bound kappa(x,y) ≤ 1/2.
domain assumption Lin-Lu-Yau Bonnet-Myers estimate: positive edge curvature lower bound kappa_0 forces diameter at most 2/kappa_0.
Used in Lemma 2.5 and Proposition 4.6 to bound distances in the universal cover one-skeleton.
domain assumption Lin-Lu-Yau product curvature formula for regular graph Cartesian products.
Used in Theorem 5.2 to get curvature 1/(2r) on every edge of T_r.
domain assumption Kueneth theorem for GLMY path homology under Cartesian products of digraphs.
Used in Theorem 5.2 to obtain the Poincare polynomial (1+t)^r for T_r.
domain assumption GLMY abelianization theorem: the abelianization of pi_1^GLMY(G,o) is H_1^GLMY(G;Z).
Used in Section 5.2 to identify H_1(G7;R) with the abelianization of the free group F2.
standard math Standard covering space existence and the covering-space index theorem for finite CW complexes.
Used in Proposition 4.6 to pass from a finite universal cover of X_{≤5}(G) to the fundamental group size bound.
standard math McShane extension theorem: a 1-Lipschitz function on a metric subspace extends to a 1-Lipschitz function on the whole space.
Used in the proof of Theorem 1.1 to extend the local function f on S to a global test function preserving values on N[x] and N[y].
standard math Total unimodularity of signed incidence matrices and integrality of the normalized Lipschitz polytope vertices.
Used in Section 5.2 to justify exact rational verification of the G7 curvature maxima by integer enumeration.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A Sharp Curvature Threshold for GLMY Path Homology." pith.science (2026). https://pith.science/paper/KIV7QKY2

@misc{pith2026260823187,
  author       = {Pith},
  title        = {Pith review of: A Sharp Curvature Threshold for GLMY Path Homology},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KIV7QKY2}},
  note         = {Machine review of arXiv:2608.23187}
}

abstract

Let $G$ be a finite simple graph with at least one edge. We prove the sharp vanishing theorem \[ \kappa_{\min}^{\mathrm{LLY}}(G)>\frac12 \quad\Longrightarrow\quad \PathH_1(G;\R)=0. \] Equivalently, nonzero first GLMY path homology forces an edge of Lin--Lu--Yau curvature at most $1/2$. The threshold $1/2$ is sharp and is attained by $C_5$. The proof combines the cycle-space description of first GLMY path homology with the limit-free Laplacian characterization of Lin--Lu--Yau curvature. As a secondary consequence of the curvature-preserving universal-cover method, we prove that if $G$ is connected and $\kappa_{\min}^{\mathrm{LLY}}(G)>0$, then $\pi_1(\Xshort{5}(G),o)$ is finite, where $\Xshort{5}(G)$ is obtained by filling every simple cycle of length at most five. Equivalently, the normal subgroup generated by based simple $5$-cycle loops has finite index in $\pi_1^{\mathrm{GLMY}}(G,o)$. In higher degrees the situation is different: for each integer $r\geq1$, the Cartesian product $T_r=C_5^{\square r}$ has curvature $1/(2r)$ on every edge and, for every field $\F$, \[ \PathH_p(T_r;\F)\cong\F^{\binom rp}\qquad(0\leq p\leq r), \] so strict positivity of Lin--Lu--Yau curvature does not force higher-dimensional GLMY path homology to vanish.

Figures

Figures reproduced from arXiv: 2608.23187 by the authors.

**Figure 1.** The graph G7. Its unique simple 3-cycle class vanishes in first GLMY path homology, while two independent simple 5-cycle classes remain. We write F2 := ⟨a, b | ⟩ for the free group of rank two; this should not be confused with the twoelement field F2. Proposition 5.1. The graph G7 satisfies κ LLY min (G7) = 1 6 , HGLMY 1 (G7; R) ∼= R 2 , πGLMY 1 (G7, 0) ∼= F2. Moreover, N5(G7, 0) = π GLMY 1 (G7, 0), πGLMY 1 (G7, 0)… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 15 canonical work pages

[1]

J.68(2018), no

Alexander Grigor’yan, Rolando Jimenez, and Yuri Muranov,Fundamental groupoids of digraphs and graphs, Czechoslovak Math. J.68(2018), no. 1, 35–65

work page 2018
[2]

Alexander Grigor’yan, Yong Lin, Yuri Muranov, and Shing-Tung Yau,Homotopy theory for digraphs, Pure Appl. Math. Q.10(2014), no. 4, 619–674

work page 2014
[3]

Alexander Grigor’yan, Yuri Muranov, and Shing-Tung Yau,Homologies of digraphs and Künneth formulas, Comm. Anal. Geom.25(2017), no. 5, 969–1018

work page 2017
[4]

Grigor’yan, Yong Lin, Yuri V

Alexander A. Grigor’yan, Yong Lin, Yuri V. Muranov, and Shing-Tung Yau,Path complexes and their homologies, J. Math. Sci.248(2020), no. 5, 564–599

work page 2020
[5]

Allen Hatcher,Algebraic topology, Cambridge University Press, Cambridge, 2002

2002
[6]

Moritz Hehl,Graphs with Lin–Lu–Yau curvature at least one and regular bone-idle graphs, Calc. Var. Partial Differential Equations64(2025), Article 196

work page 2025
[7]

Ann.394(2026), Article 24

,Regular graphs with positive Ollivier–Ricci curvature, Math. Ann.394(2026), Article 24

work page 2026
[8]

Moritz Hehl and Florentin Münch,Betti number estimates for non-negatively curved graphs, 2025, arXiv:2503.11222

work page Pith review arXiv 2025

Show all 16 references

[9]

Mark Kempton, Florentin Münch, and Shing-Tung Yau,A homology vanishing theorem for graphs with positive curvature, Comm. Anal. Geom.29(2021), no. 6, 1449–1473

2021
[10]

Jingyan Li, Yuri Muranov, and Shing-Tung Yau,Methods of algebraic topology in graph theory, International Press, 2026

2026
[11]

Huiqiu Lin and Zhe You,Graphs with positive Lin–Lu–Yau curvature without quadri- laterals, Calc. Var. Partial Differential Equations65(2026), Article 221

2026
[12]

Yong Lin, Linyuan Lu, and Shing-Tung Yau,Ricci curvature of graphs, Tohoku Math. J. (2)63(2011), no. 4, 605–627

2011
[13]

E. J. McShane,Extension of range of functions, Bull. Amer. Math. Soc.40(1934), no. 12, 837–842

1934
[14]

Wojciechowski,Ollivier Ricci curvature for general graph laplacians: Heat equation, laplacian comparison, non-explosion and diameter bounds, Adv

Florentin Münch and Radosław K. Wojciechowski,Ollivier Ricci curvature for general graph laplacians: Heat equation, laplacian comparison, non-explosion and diameter bounds, Adv. Math.356(2019), 106759

2019
[15]

Yann Ollivier,Ricci curvature of Markov chains on metric spaces, J. Funct. Anal.256 (2009), no. 3, 810–864. A SHARP CUR V ATURE THRESHOLD FOR GLMY PATH HOMOLOGY 27

2009
[16]

58, American Mathematical Society, Providence, RI, 2003

Cédric Villani,Topics in optimal transportation, Graduate Studies in Mathematics, vol. 58, American Mathematical Society, Providence, RI, 2003

2003

Pith tools

Reviewed August 28, 2026 · model on record in the stance chip above.

[1] [1]

J.68(2018), no

Alexander Grigor’yan, Rolando Jimenez, and Yuri Muranov,Fundamental groupoids of digraphs and graphs, Czechoslovak Math. J.68(2018), no. 1, 35–65

work page 2018

[2] [2]

Alexander Grigor’yan, Yong Lin, Yuri Muranov, and Shing-Tung Yau,Homotopy theory for digraphs, Pure Appl. Math. Q.10(2014), no. 4, 619–674

work page 2014

[3] [3]

Alexander Grigor’yan, Yuri Muranov, and Shing-Tung Yau,Homologies of digraphs and Künneth formulas, Comm. Anal. Geom.25(2017), no. 5, 969–1018

work page 2017

[4] [4]

Grigor’yan, Yong Lin, Yuri V

Alexander A. Grigor’yan, Yong Lin, Yuri V. Muranov, and Shing-Tung Yau,Path complexes and their homologies, J. Math. Sci.248(2020), no. 5, 564–599

work page 2020

[5] [5]

Allen Hatcher,Algebraic topology, Cambridge University Press, Cambridge, 2002

2002

[6] [6]

Moritz Hehl,Graphs with Lin–Lu–Yau curvature at least one and regular bone-idle graphs, Calc. Var. Partial Differential Equations64(2025), Article 196

work page 2025

[7] [7]

Ann.394(2026), Article 24

,Regular graphs with positive Ollivier–Ricci curvature, Math. Ann.394(2026), Article 24

work page 2026

[8] [8]

Moritz Hehl and Florentin Münch,Betti number estimates for non-negatively curved graphs, 2025, arXiv:2503.11222

work page Pith review arXiv 2025