Pith. sign in

REVIEW 2 major objections 5 minor 9 references

Quantitative Obata's theorem in discrete setting

T0 review · 2 major / 5 minor · reviewed 2026-08-05 · 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 when a graph's d-th nonzero Laplacian eigenvalue is close to its curvature bound, the graph must be combinatorially and quantitatively a d-dimensional hypercube, with distance functions approximated by the first d eige

desk verdict Quantitative Obata for hypercubes is a genuine advance; the main proof has one repairable gap in the compactness step. read the letter →

arxiv 2508.20815 v1 pith:F73PCZRW submitted 2025-08-28 math.DG math.COmath.SP

Shiping Liu , Chiyu Zhou This is my paper · ORCID
classification math.DGmath.COmath.SP MSC 05C5053C2135P15
keywords Bakry–ÉmerycurvaturehypercubegraphquantitativeObatatheoremLaplacianeigenvaluesspectralrigidityFrobeniusdistanceRicciongraphsdiscretealmost
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 proves a discrete counterpart of two classical Riemannian results: almost rigidity and quantitative Obata. It shows that if a connected weighted graph satisfies a Bakry–Émery curvature bound K>0, has maximum combinatorial degree d, and its d-th nonzero Laplacian eigenvalue λ_d is within ε of K, then the graph is combinatorially a d-dimensional hypercube. Moreover, its edge weights and vertex measures are close to those of the constant-weight hypercube H_d(K/2), with errors of order √ε. The same closeness holds for eigenfunctions: near any chosen vertex, some linear combination of the first d eigenfunctions reproduces the combinatorial distance function up to O(√ε). This gives a quantitative, purely discrete analogue of Obata's theorem and explains why the hypercube plays the role of the sphere in this setting.

What carries the argument

The argument runs on the Bakry–Émery curvature-dimension inequality CD(K,∞), the d-th nonzero Laplacian eigenvalue λ_d, and the Frobenius distance between weighted graphs. The main technical engine is a family of restriction maps L_x from the low-energy eigenspace (spanned by the first d eigenfunctions plus constants) to functions on the one-ball around a vertex; when λ_d is close to K these maps are bijective with bounded inverse. The eigenvalue gap is converted into pointwise estimates on Γφ and two-step harmonicity, showing that low eigenfunctions are almost distance-like. A compactness argument, relying on the finite diameter bound from CD(K,∞) and on the rigidity theorem for hypercubes,

What would settle it

Take a sequence of graphs in G(D,d,δ) satisfying CD(K,∞) with λ_d → K and examine the limiting weighted graph; if some edge weight tends to zero, then the limiting maximum degree is d′ < d and the rigidity theorem cannot directly apply. Finding such a sequence, or proving that no edge weight can vanish, would settle whether the hypercube conclusion in Theorem 10 is actually forced by the spectral condition.

Watch

Extended reading notes

Core claim

For graphs in a bounded class—controlled weighted degree, controlled vertex measures, and maximal combinatorial degree d—the combination of CD(K,∞) curvature and λ_d ≤ K+ε forces the graph's combinatorial structure to be exactly the d-dimensional hypercube. The weighted graph is then close, in Frobenius distance, to H_d(K/2), the hypercube with constant edge degree K/2, with a bound of order √(λ_d−K). Separately, for any vertex x0, there is a function u in the span of the first d eigenfunctions such that the combinatorial distance function dist_x0 satisfies ||dist_x0 − d/2 − u||_2 ≤ C√(λ_d−K). These are the paper's Theorems 3 and 4, stated as discrete analogues of the almost-rigidity theorem

Load-bearing premise

The compactness proof assumes that in the limit no edge weight vanishes, so the limiting graph still has maximum combinatorial degree d; the paper's one-line contradiction presumes the limit is already d-regular, and only then does the hypercube rigidity theorem apply.

Editorial extensions

If this is right

  • Any graph in the class satisfying CD(K,∞) with λ_d ≤ K+ε is combinatorially a d-dimensional hypercube; no edge can be added or removed without breaking the assumptions.
  • The weighted graph is Frobenius-close to H_d(K/2): each oriented edge degree satisfies |q(x,y)−K/2| ≤ C√ε and each vertex measure satisfies |m(x)−1| ≤ C√ε.
  • For any vertex x0, there is a combination u of the first d eigenfunctions with ||dist_x0 − d/2 − u||_2 ≤ C√(λ_d−K), so distance functions become spectral objects when the spectral gap is small.
  • After a self-improvement step, the constants in the Frobenius and eigenfunction estimates depend only on the dimension d and the curvature scale K, not on the auxiliary bounds D and δ.
  • The threshold ε0 depends on D, K, d, and δ, so the rigidity is uniform over the whole graph class.

Reading between the lines

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

  • The proof suggests that the spectral condition λ_d ≈ K acts as a dimensional detector: it selects the hypercube among all graphs with degree d, whereas pinching fewer than d eigenvalues can produce products H_l × G with high multiplicity, as the paper's Example 1 shows.
  • The √ε rate emerges from quadratic-form and discriminant estimates; whether this rate is optimal, or whether a sharper exponent holds, is not addressed and could be probed numerically on weighted hypercubes with a single perturbed edge weight.
  • A testable extension would be to apply the same pinching to the second spectral gap λ_{d+1}−λ_d; the rigidity theorem implies the unperturbed hypercube has λ_{d+1}=2K, so a quantitative statement about the next eigenvalue may follow from similar methods, though the paper does not pursue it.
  • Unlike the continuous quantitative Obata theorem, the discrete eigenfunction statement holds for every reference vertex x0, suggesting a stronger structural rigidity that may transfer to product graphs or to finite Markov chains with hypercube-like geometry.
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 / 5 minor

Summary. The paper establishes quantitative discrete analogues of almost-rigidity and Obata-type theorems for weighted graphs satisfying the Bakry-Émery condition CD(K,∞). For the class G(D,d,δ) of connected weighted graphs with maximal combinatorial degree d, bounded weighted degree, and bounded vertex measure, the authors prove that if λ_d is sufficiently close to K, then (i) the graph is combinatorially the d-dimensional hypercube, (ii) it is close in Frobenius distance to the weighted hypercube H_d(K/2), and (iii) distance functions from vertices are approximated by spans of the first d eigenfunctions. The proof combines spectral perturbation estimates (Theorems 6 and 7), an L∞ approximation of distance functions by eigenfunctions (Lemma 4), quantitative estimates of degree and edge weights (Theorem 9), and a compactness argument using the rigidity theorem of Liu–Münch–Peyerimhoff (Theorem 2). The paper also includes examples showing that weaker spectral assumptions fail.

Significance. If the proof is completed, this is a meaningful contribution: it gives the first quantitative version of the discrete Obata rigidity result of [LMP24], with explicit polynomial dependence on the spectral deficit and fully explicit constants (depending on D,K,d,δ). The self-improving argument in Theorem 3, which removes the dependence on the artificial lower bound η on edge weights, is elegant and gives a genuine quantitative statement. The paper also correctly identifies and illustrates the obstruction to replacing λ_d by λ_ℓ for ℓ<d. The central estimates in Theorem 6, Lemma 4, and Theorem 9 appear sound, and the claimed results are falsifiable and do not depend on fitted parameters. However, the compactness proof of Theorem 10 contains a genuinely circular step that must be repaired before the main theorems are fully established.

major comments (2)
  1. [Theorem 10, proof of part (i), Section 4] The proof applies the Rigidity Theorem to the limit graph G0 after establishing only λ_d(G0)=K. The Rigidity Theorem (Theorem 2) requires λ_{degmax(G0)}=K, not merely λ_d(G0)=K. The sentence 'w0(x,y) will not vanish, otherwise deg(x) ≤ d−1. This contradicts with G0 being a hypercube' presupposes that G0 is already known to be a hypercube, which is exactly what is to be proved. This is a circularity. A repair is needed: if degmax(G0)=d'<d, then since Lichnerowicz gives λ_1(G0)≥K and λ_d(G0)=K, one has λ_1=...=λ_d=K, hence λ_{d'}=K. The rigidity theorem applied to (G0,d') forces G0=H_{d'}(K/2), whose (d'+1)-st eigenvalue is 2K. Since d>d', this contradicts λ_d(G0)=K. Thus degmax(G0)=d and all edges have positive weights in the limit, after which rigidity applies. This missing argument is load-bearing: it underlies part (i), the uniform lower bound η in part (ii), and ultimately Theorems 3
  2. [Theorem 9 and Theorem 3, Section 3.3 and Section 4] Theorem 9 assumes as a hypothesis that min_{w(x,y)>0} w(x,y) ≥ η. In the proof of Theorem 3, this η is supplied by Theorem 10(ii), which in turn relies on the unproved compactness argument of Theorem 10(i). Consequently, the current manuscript has a dependency cycle: the quantitative estimates of Theorem 9 are used to prove the main theorem, but their key hypothesis is only available after the compactness step that is not rigorously established. This does not invalidate the overall strategy, but the missing degree-drop argument must be inserted in Theorem 10 before the later theorems can be considered proved.
minor comments (5)
  1. [Abstract] Typo: 'Rimennian manifolds' should be 'Riemannian manifolds'.
  2. [Section 2.2, Definition 4] Typo: 'over all all permutations' should be 'over all permutations'.
  3. [Section 3.2 (introductory paragraph)] The reference to 'Bertrand’s result' is not explained; the authors likely mean Bertrand’s quantitative Obata theorem, but the connection is not made explicit. Please clarify.
  4. [Theorem 8, proof] The sentence 'Theorem 8 follows by Lemma 4’s argument' is terse. Since Theorem 8 is not a formal corollary of Lemma 4 (it uses a restriction map on a subspace of possibly smaller dimension), a few more details would help the reader.
  5. [Throughout] The notation q(y,x) is used in (3.20) but q is defined as w(x,y)/m(x) for oriented edge (x,y); when writing q(y,x)=w(x,y)/m(y), it would be clearer to state that q(y,x) denotes the edge degree of the reverse orientation.

Circularity Check

1 steps flagged · score 4.0 of 10

Theorem 10's compactness step is circular: the limit graph is declared a hypercube before proving degmax(G0)=d, and vanishing edges are excluded by appealing to the very hypercube conclusion at issue. Otherwise the derivation uses prior rigidity/diameter theorems as legitimate external inputs and involves no fitted parameters; the gap is repairable.

  1. other [Theorem 10(1), Section 4 (proof of almost-rigidity)]
    "By continuity of spectrum, λd(G0) = K. By the Rigidity Theorem (see Theorem 2) , G0 is a hypercube. Note that for any {x, y} ∈E, w0(x, y) will not vanish, otherwise deg(x) ≤ d − 1. This contradicts with G0 being a hypercube."

    Rigidity Theorem 2 is applicable only when λ_{degmax(G0)} = K. The limit construction establishes only λ_d(G0) = K; if some edge weight vanishes, then degmax(G0) < d and the theorem does not apply. The next sentence attempts to rule out vanishing by saying it contradicts 'G0 being a hypercube' — but that G0 is a hypercube is exactly the conclusion the rigidity application was supposed to establish. Thus the proof assumes the conclusion to eliminate the case (degree drop) in which its rigidity hypothesis is unmet. A non-circular proof would first exclude degmax(G0)=d'<d: since λ_1,...,λ_d all equal K, λ_{d'}=K, rigidity forces G0=H_{d'}(K/2), whose (d'+1)-st eigenvalue is 2K, contradicting λ_d=K when d>d'. Only after this exclusion are all weights positive, degmax(G0)=d, and the rigidity th

full rationale

The paper's central quantitative claims do not reduce by construction to fitted inputs: no parameters are calibrated to data, and the statements in Theorems 3 and 4 are genuine spectral-stability results rather than renamings of assumptions. The cited rigidity theorem [LMP24] and diameter bound [LMP18] are prior mathematical results, some co-authored by the current authors, but they are external benchmarks and not the targets of the present derivation. The one true circularity is localized in Theorem 10's compactness argument: the proof applies the rigidity theorem to the limit graph G0 without first proving degmax(G0)=d, and then rules out vanishing edge weights by invoking the hypercube conclusion that the rigidity application was meant to prove. This is a genuine logical gap in the written proof and is load-bearing for the existence of epsilon0 and eta, hence for Theorems 3 and 4. Because the gap is repairable by a standard degree-drop/multiplicity argument and does not make the whole derivation equivalent to its inputs, the score is 4 rather than higher. No other circular steps of the enumerated kinds were found.

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

The paper introduces no new mathematical entities and no free parameters. Its constants C(D,K,d,delta) are inequality constants, not fitted values. The main external load is the rigidity theorem of [LMP24] and the diameter bound of [LMP18], both prior results in the same research program led in part by the current first author.

assumptions (5)
  • domain assumption Rigidity theorem: G satisfies CD(K, infinity) and lambda_degmax = K iff G is the degmax-dimensional weighted hypercube with constant edge degree K/2 (Theorem 2, from [LMP24]).
    Used as a black box in the contradiction proof of Theorem 10 to identify the limiting graph as a hypercube.
  • domain assumption Bonnet-Myers type bound: a connected CD(K, infinity) graph with K > 0 satisfies diam(G) <= 2 Degmax/K (Theorem 5, from [LMP18]).
    Used throughout to get finiteness, norm equivalence, and induction bounds in Lemmas 1, 4 and Theorems 6-9.
  • standard math Continuity of graph Laplacian eigenvalues under limits of weights and vertex measures on a fixed finite vertex set.
    Used in the compactness argument of Theorem 10 to pass from lambda_d(G_n) <= K + 1/n to lambda_d(G0) = K.
  • domain assumption Cartesian product curvature formula: if G1 and G2 satisfy CD(K1, infinity) and CD(K2, infinity), then their product satisfies CD(min{K1,K2}, infinity) ([LP18]).
    Used only in Examples 1 and 2 to construct counterexamples for smaller spectral indices.
  • standard math Finite-dimensional norm equivalence and Cauchy-Schwarz / Holder inequalities.
    Used repeatedly in eigenfunction estimates and in Lemma 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Quantitative Obata's theorem in discrete setting." pith.science (2026). https://pith.science/paper/F73PCZRW

@misc{pith2026250820815,
  author       = {Pith},
  title        = {Pith review of: Quantitative Obata's theorem in discrete setting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F73PCZRW}},
  note         = {Machine review of arXiv:2508.20815}
}
abstract

Under mild assumptions, we show that a connected weighted graph $G$ with lower Ricci curvature bound $K>0$ in the sense of Bakry-\'Emery and the $d$-th non-zero Laplacian eigenvalue $\lambda_d$ close to $K$, with $d$ being the maximal combinatorial vertex degree of $G$, has an underlying combinatorial structure of the $d$-dimensional hypercube graph. Moreover, such a graph $G$ is close in terms of Frobenius distance to a properly weighted hypercube graph. Furthermore, we establish their closeness in terms of eigenfunctions. Our results can be viewed as discrete analogies of the almost rigidity theorem and quantitative Obata's theorem on Rimennian manifolds.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

9 extracted references · 9 canonical work pages

  1. [1]

    Pincement sur le spectre et le volume en courbure de Ricci positive

    [Aub05] Erwann Aubry. “Pincement sur le spectre et le volume en courbure de Ricci positive”. In: Ann. Sci. ´Ecole Norm. Sup. (4) 38.3 (2005), pp. 387–405. [Bau+17] F. Bauer et al. “Curvature aspects of graphs”. In: Proc. Amer. Math. Soc. 145.5 (2017), pp. 2033–2042. [Ber07] J´ erˆ ome Bertrand. “Pincement spectral en courbure de Ricci positive”. In: Comme...

    work page 2005

  2. [9]

    Curvature and higher order Buser inequalities for the graph connection Laplacian

    [LMP19] Shiping Liu, Florentin M¨ unch, and Norbert Peyerimhoff. “Curvature and higher order Buser inequalities for the graph connection Laplacian”. In: SIAM J. Discrete Math. 33.1 (2019), pp. 257–305. [LMP24] Shiping Liu, Florentin M¨ unch, and Norbert Peyerimhoff. “Rigidity properties of the hypercube via Bakry- ´Emery curvature”. In: Math. Ann. 388.2 (...

    work page 2019

  3. [16]

    Volume doubling, Poincar´ e inequality and Gaussian heat kernel esti- mate for non-negatively curved graphs

    [Hor+19] Paul Horn et al. “Volume doubling, Poincar´ e inequality and Gaussian heat kernel esti- mate for non-negatively curved graphs”. In: J. Reine Angew. Math. 757 (2019), pp. 89–

    work page 2019

  4. [26]

    Curvature and transport inequalities for Markov chains in discrete spaces

    Progr. Probab. Birkh¨ auser Boston, Boston, MA, 1991, pp. 96–110. [FS18] Max Fathi and Yan Shu. “Curvature and transport inequalities for Markov chains in discrete spaces”. In: Bernoulli 24.1 (2018), pp. 672–698. [Gro99] Misha Gromov. Metric structures for Riemannian and non-Riemannian spaces. Vol

    work page 1991

  5. [34]

    Math. Sci. Res. Inst. Publ. Cambridge Univ. Press, Cambridge, 1999, pp. 189–197. 18

    work page 1999

  6. [117]

    Leibniz Int

    LIPIcs. Leibniz Int. Proc. Inform. Schloss Dagstuhl. Leibniz-Zent. Inform., Wadern, 2018, Art. No. 20,

    work page 2018

  7. [130]

    Bakry- ´Emery curvature and diameter bounds on graphs

    [LMP18] Shiping Liu, Florentin M¨ unch, and Norbert Peyerimhoff. “Bakry- ´Emery curvature and diameter bounds on graphs”. In: Calc. Var. Partial Differential Equations 57.2 (2018), Paper No. 67,

    work page 2018

  8. [152]

    Graph similarity and ap- proximate isomorphism

    Progress in Mathematics. Based on the 1981 French original, With appendices by M. Katz, P. Pansu and S. Semmes, Translated from the French by Sean Michael Bates. Birkh¨ auser Boston, Inc., Boston, MA, 1999, pp. xx+585. [GR W18] Martin Grohe, Gaurav Rattan, and Gerhard J. Woeginger. “Graph similarity and ap- proximate isomorphism”. In: 43rd International S...

    work page 1981

Show all 9 references
  1. [348]

    Bakry- ´Emery curvature func- tions on graphs

    Grundlehren der mathematischen Wissenschaften. Springer, Cham, 2014, pp. xx+552. [CLP20] David Cushing, Shiping Liu, and Norbert Peyerimhoff. “Bakry- ´Emery curvature func- tions on graphs”. In: Canad. J. Math. 72.1 (2020), pp. 89–143. [CMS23] Fabio Cavalletti, Andrea Mondino,...

    2014

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.