Pith. sign in

REVIEW 2 major objections 4 minor 37 references

Volume Entropy Rigidity for Random Groups at Low Densities

T0 review · 2 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Pith's one-line read For a torsion-free hyperbolic group with a translation-apparent presentation, the volume entropy has a unique normalized minimizing weight, and such presentations are generic for low-density random groups.

desk verdict A genuinely new rigidity theorem for random groups at low densities, with a small but fixable factor-of-2 error in Proposition 4.8. read the letter →

arxiv 2505.23364 v1 pith:6PS67ITG submitted 2025-05-29 math.GR math.MG

classification math.GRmath.MG MSC 20F6720F6520P05
keywords volumeentropyweightedwordmetricshyperbolicgroupsrandomsmallcancellationrigiditylengthspectrumtranslation-apparentpresentation
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 that, for a large class of hyperbolic groups, assigning positive weights to the generators (normalized to sum to 1) produces exactly one weighted word metric that minimizes the volume entropy, the exponential growth rate of balls. The enabling object is a 'translation-apparent presentation': a small-cancellation presentation in which each generator appears evenly inside every relator at three separate scales. The paper shows that in such a presentation every power of a generator is a geodesic with respect to every weight, so the stable translation length of a generator equals its weight; distinct weights therefore yield non-comparable length spectra, and an external rigidity theorem converts this into strict convexity and uniqueness of the minimizer. It then proves that these presentations are generic in Gromov's density model of random groups at any density below an explicit threshold d_m, and that for such generic groups the unique minimizer is almost the uniform weight and the minimal entropy is almost the free group's entropy. The result matters because volume entropy is the natural normalizable invariant for comparing metrics on a group, and this is a setting where rigidity can be located precisely and the minimizer described explicitly.

What carries the argument

The load-bearing object is the λ-translation-apparent presentation: a symmetrized, cyclically reduced presentation satisfying the C'(λ) small-cancellation condition together with three even-distribution conditions — no relator contains a subword s^n with n≥λ|r|, no subword of length ⌈4λ|r|⌉ contains half of any generator's occurrences in the relator, and every subword of length ⌈λ|r|⌉ contains at least 1/(8m) of its letters equal to any given generator. The small-cancellation side, through the lemma of [14], shows that any word representing the identity must contain a long piece of a relator; the even-distribution side ensures that such a piece cannot have small weight, so any word avoiding long relator subwords (in particular any power of a generator) is the unique geodesic for every weight. This yields the identity ℓ_{d_w}(s)=w(s) for each generator, the direct link from weights to the length spectrum. The third even-distribution condition also feeds the weighted subword-avoidance generating functions of [28], which produce the free-group entropy bounds in Proposition 4.8 and hence Theorem C.

What would settle it

For m=2 at a density such as d=d_2/2, sample random presentations with relator length ℓ up to roughly $10^{4}$ and test whether the symmetrized presentation satisfies the three even-distribution conditions of Definition 4.1 with λ=1/16; the paper predicts an exponentially small failure probability in ℓ, so observing a failure rate that stays bounded away from zero as ℓ grows would refute the genericity claim of Theorem B.

Watch

Extended reading notes

Core claim

The central claim is Theorem A (Section 4.5): if a torsion-free non-elementary hyperbolic group (G,S) admits a λ-translation-apparent presentation, then the volume entropy is strictly convex on the simplex of normalized weights on S, and there is a unique normalized weight minimizing it. The proof shows that in such a presentation every power of a generator s is a geodesic with respect to every weight w, so the stable translation length of s equals w(s); hence distinct weights give distinct length spectra, and the length-spectrum rigidity theorem cited as [6] upgrades this to non-rough-isometry of the two weighted metrics, which the convexity-up-to-rough-isometry result rules out for two minimizers. The genericity claim is Theorem B (Section 5.11): for each m≥2 and each density 0≤d<d_m, where d_m=(m-1)^2/(6144 $m^{2}$(2m-1)^2 ln(2m-1)), a generic random group on m letters admits a 1/16-translation-apparent presentation by symmetrizing the random relators. Finally, Theorem C (Section 5.13) states that for such generic groups, every normalized weight w satisfies h(F_m,w)-ε≤h(G,w)≤h(F_m,w), and the unique minimizer is within ε of the uniform weight, with the minimum entropy within ε of the free group's minimum.

Load-bearing premise

The proof assumes the length-spectrum rigidity theorem cited as [6], namely that two hyperbolic metrics on a non-elementary hyperbolic group with equal stable translation lengths for every element are necessarily roughly isometric, and the paper cites rather than reproves that external result.

Editorial extensions

If this is right

  • Any torsion-free hyperbolic group with a translation-apparent presentation has a unique entropy-minimizing normalized weight, and its volume entropy is strictly convex on the weight simplex.
  • For every m≥2 and every density d<d_m, generic random groups on m letters have unique minimizers, giving an explicit infinite family of groups with volume-entropy rigidity.
  • For these generic groups the minimizer is nearly uniform and the minimal entropy is nearly that of the free group, so the uniform weighting is essentially optimal relative to any perturbation.
  • Stability holds: whenever a marked group has a unique minimizing weight, any sequence of weights whose entropy converges to the minimum must converge to that weight.
  • The same uniqueness criterion applies to surface groups with standard generators, where the uniform weight is the unique minimizer even though the standard presentation is not translation-apparent.

Reading between the lines

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

  • Because the proof transfers along the existence of a translation-apparent presentation and a length-spectrum rigidity theorem, the same uniqueness conclusion should hold for any other class of groups (for instance relatively hyperbolic groups) in which both ingredients are available; the paper does not discuss this extension.
  • The explicit value of d_m is an artifact of the Chernoff-bound estimates and the arbitrary choice λ=1/16; the author notes that d_m could be improved, which suggests the true regime where uniqueness is generic may be substantially larger than the theorem states.
  • The near-uniformity of the minimizer is a quantitative statement about the natural generating set, indicating that for fixed high-length random groups the entropy landscape is very flat around the uniform weight; one testable consequence is that small random perturbations of the uniform weight should change the entropy by an amount of order 1/ℓ, matching the paper's error bounds.
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

2 major / 4 minor

Summary. The paper studies the rigidity of the volume entropy for weighted word metrics on hyperbolic groups. It introduces the notion of a λ-translation-apparent presentation, which combines the C′(λ) small cancellation condition with three 'even distribution' conditions on relators. The main results are: Theorem A, asserting that a torsion-free hyperbolic group admitting such a presentation has a unique normalized weight minimizing the volume entropy; Theorem B, asserting that for each m ≥ 2 there is an explicit small density d_m such that a generic random group on m letters at density d < d_m admits a 1/16-translation-apparent presentation; and Theorem C, asserting that such random groups have entropy arbitrarily close to that of the free group for every weight, that the unique minimizer is arbitrarily close to the uniform weight, and that the minimum entropy is arbitrarily close to that of the free group. The proofs combine small cancellation theory, a counting method of Myers for weighted subword avoidance, a Chernoff bound for Markov chains, and the recent convexity and length-spectrum results of Cantrell–Tanaka. The paper also proves a stability result for minimizers when uniqueness holds.

Significance. The paper makes a valuable contribution to the study of volume entropy rigidity for weighted word metrics. The notion of a translation-apparent presentation is a new and potentially useful tool, and the explicit genericity threshold in Theorem B gives a concrete family of hyperbolic groups where the minimizer is unique and almost uniform. The proof of Theorem C relies on a detailed counting argument with Myers' generating functions, and the paper provides both explicit constants and a self-contained appendix. If the technical gaps identified below are repaired, the results will constitute a solid advance. The main external dependency, the Cantrell–Tanaka length-spectrum rigidity theorem, is clearly stated and used in a standard way. The paper is well organized and the exposition is generally clear, though several proof details need correction.

major comments (2)
  1. [Section 4.2, Proposition 4.8] The lower bound on |u|_w uses an incorrect interpretation of the third even distribution condition. Definition 4.1(iii) states that for each pair a ∈ A, #_{a±}(u) > |u|/(8m). It does not imply #_s(u) > |u|/(8m) for each individual s ∈ S. Summing the pair contributions gives |u|_w = Σ_a #_{a±}(u) w(a) > (|u|/(8m)) Σ_{a∈A} w(a) = N|u|/(16m), not N|u|/(8m). Consequently, the estimate q(b) ≤ 2mj/b^{Nl/(8m)} should read q(b) ≤ 2mj/b^{Nl/(16m)}, and the subsequent derivation of p(b) < 0 is invalid. This error propagates to Proposition 4.9 and Theorem C. A repair is to replace the sufficient condition by 8mjl < M_0^{N(l/(32m)-2)} and to choose ℓ in Theorem 5.13 accordingly; however, the hypothesis as stated, 8mjl < (2m−1)^{l/(16m)-2}, does not imply the stronger corrected condition, so the statement of Proposition 4.8 must be amended.
  2. [Section 4.1, Proposition 4.3] The proof applies the second even distribution condition to the word u3, whose length is only known to be less than ⌈4λ|r|⌉, whereas condition (ii) is formulated for subwords of length exactly ⌈4λ|r|⌉. The argument should first extend u3 to a subword of r of that length, which is possible because u3 is a suffix of a cyclic permutation of r, and then apply condition (ii). As written, this step is unjustified. Since Proposition 4.3 underpins the proof of Theorem A via Corollary 4.4, this detail needs to be fixed.
minor comments (4)
  1. [Lemma 5.7] In the proof, the line 'the distribution of uj is the uniform distribution on reduced words of length ⌈ℓ/16⌉' should read '⌈ℓ/4⌉'.
  2. [Lemma 5.10] The cited formula for the number of cyclically reduced words, (2m−1)^ℓ + m + (−1)^ℓ(m−1) ≥ (2m−1)^ℓ, appears to be incorrect: for m=2, ℓ=2 it would give 15, exceeding the total number of reduced words of length 2. Only the exponential rate matters for the subsequent argument, so the proof can be repaired by using the asymptotic count c_m(2m−1)^ℓ, but the displayed inequality should be corrected.
  3. [Theorem 5.13] The bound j ≤ |R_ℓ^*| ≤ ℓ(2m−1)^{dℓ} omits the factor coming from symmetrization; each relator gives at most 2ℓ cyclic permutations and inverses, so the correct upper bound is |R_ℓ^*| ≤ 2ℓ(2m−1)^{dℓ}. The argument is asymptotic, so this does not change the main conclusion, but the displayed inequality should be fixed.
  4. [Throughout] There are numerous typos and spacing errors in the text, such as 'W e', 'a', and 'A T' in the title, which should be corrected in a revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the derivation is self-contained and the cited external results are not self-citations.

full rationale

The paper's main chain is: define translation-apparent presentations, prove that such presentations force the generator translation lengths to equal their weights, invoke the external length-spectrum rigidity theorem of Cantrell-Tanaka to conclude uniqueness of the entropy minimizer, and then prove genericity of translation-apparent presentations for random groups via a Chernoff bound for Markov chains. None of these steps reduces to its own input. The condition 'translation-apparent' is not defined in terms of the minimizer or the entropy; it is an explicit small-cancellation plus letter-frequency hypothesis, and the equality of translation length with weight is proved in Proposition 4.3 and Corollary 4.4 rather than assumed. The constants lambda = 1/16 and d_m are chosen to make the probabilistic estimates work; they are not fitted to force the desired minimizer, and the minimizer is not prescribed by the construction. The most load-bearing external result, Theorem 2.3 from Cantrell-Tanaka, is a genuine external theorem, not a self-citation, so it does not constitute circularity under the stated rules. The skeptical concern about a factor-of-2 in Proposition 4.8 is a correctness or repairability issue in the written estimate, not a circularity: an erroneous intermediate bound does not make the conclusion equivalent to the hypotheses. Overall, the central claims carry independent content and no self-referential reduction is present.

Assumptions & free parameters 2 free parameters · 8 assumptions · 0 invented entities

The paper does not introduce new physical or mathematical entities; it introduces a new property of group presentations (translation-apparent), which is a condition, not an entity. The free parameters are hand-chosen proof constants, not fitted to data. The axioms are standard results from small cancellation theory, random group theory, and Markov chain analysis, plus external rigidity theorems cited from the literature.

free parameters (2)
  • lambda (small-cancellation parameter) = 1/16
    Chosen by hand (Remark 5.14); defines translation-apparent presentations, the density threshold d_m, and all derived constants. The authors state this choice is arbitrary and unoptimized.
  • delta (Chernoff slack in Lemma 5.4) = 1/(4m)
    Hand-picked slack in the Markov chain Chernoff bound; enters the constant C_m and therefore the density threshold d_m. Any sufficiently small delta would work, so this is a proof constant rather than a fitted physical parameter.
assumptions (8)
  • standard math Greendlinger's Lemma (Theorem 2.7)
    Used in Proposition 4.3 to force a large common subword with a relator; requires C'(lambda), lambda <= 1/6, satisfied by lambda = 1/16.
  • domain assumption Cantrell-Tanaka convexity and length-spectrum rigidity (Theorems 2.2, 2.3)
    Main external input: Manhattan curve strict convexity up to rough similarity, and rough isometry iff translation lengths agree. Used for existence of rough-unique minimizer and Theorem A.
  • standard math Gromov's lemma on high powers generating free subgroups (Lemma 3.1)
    Used to prove existence of minimizers (Prop 3.3) by bounding entropy below by free group entropy.
  • standard math Balacheff-Merlin entropy formula for weighted free groups (Lemma 3.2)
    Gives exact entropy for F_k with arbitrary weights; used to compute M0 and free-group comparisons.
  • standard math Myers' generating-function system for weighted subword avoidance (Theorem A.1)
    Framework for counting lambda-reduced words; the paper provides a corrected proof in Appendix A, so it is self-contained.
  • standard math Lezaud's Chernoff bound for reversible Markov chains (Theorem 5.1)
    Used with spectral gap estimate to show even distribution is generic; requires irreducibility and reversibility, satisfied by the free-group walk.
  • standard math Sinclair-Jerrum Cheeger bound for the spectral gap (Theorem 5.3)
    Used to lower-bound the spectral gap epsilon(M) by (m-1)^2/(2(2m-1)^2), yielding the exponential rate C_m.
  • domain assumption Known genericity of hyperbolicity and C'(2d+epsilon) for random groups at density d<1/2
    Used in Theorem 5.11 to get C'(1/16); since d < d_m < 1/32, can choose epsilon with 2d + epsilon < 1/16.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Volume Entropy Rigidity for Random Groups at Low Densities." pith.science (2026). https://pith.science/paper/6PS67ITG

@misc{pith2026250523364,
  author       = {Pith},
  title        = {Pith review of: Volume Entropy Rigidity for Random Groups at Low Densities},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6PS67ITG}},
  note         = {Machine review of arXiv:2505.23364}
}
read the original abstract

We study the rigidity of the volume entropy for weighted word metrics on hyperbolic groups, building on a recent convexity result due to Cantrell-Tanaka. Using ideas from small cancellation theory, we give conditions under which a hyperbolic group admits a unique normalized weight minimizing the entropy. Moreover, we show that these conditions are generic for random groups at small densities, and that the unique minimizer of such a generic group is arbitrarily close to the uniform weight.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 33 canonical work pages

  1. [1]

    Ivan Babenko and St´ ephane Sabourau, Minimal volume entropy and fiber growth , Journal de l’´Ecole polytechnique — Math´ ematiques 12 (2025), 481–521 (en)

  2. [2]

    06, 2429–2434

    Florent Balacheff and Louis Merlin, A curvature-free Log(2k − 1) theorem, Proceedings of the American Mathematical Society 151 (2023), no. 06, 2429–2434

  3. [3]

    5, 731–799

    G´ erard Besson, Gilles Courtois, and Sylvestre Gallot, Entropies et rigidit´ e des espaces lo- calement sym´ etriques de courbure strictement., Geometric and functional analysis 5 (1995), no. 5, 731–799

  4. [4]

    2, 123–148

    James W Cannon, The combinatorial structure of cocompact discrete hyperbo lic groups , Ge- ometriae Dedicata 16 (1984), no. 2, 123–148

  5. [5]

    Stephen Cantrell and Eduardo Reyes, Marked length spectrum rigidity from rigidity on sub- sets, arXiv:2304.13209 [math.GT], 2024

  6. [6]

    Stephen Cantrell and Ryokichi Tanaka, The manhattan curve, ergodic theory of topological flows and rigidity , Geometry and Topology (2024), In Press

  7. [7]

    4, 1615–1681

    Chris Connell, Xianzhe Dai, Jes´ us N´ u˜ nez-Zimbr´ on, Raquel Perales, Pablo Su´ arez-Serrato, and Guofang W ei, Maximal volume entropy rigidity for spaces , Journal of the London Mathemat- ical Society 104 (2021), no. 4, 1615–1681

  8. [8]

    Chris Connell, Xianzhe Dai, Jes´ us N´ u˜ nez-Zimbr´ on, Raquel Perales, Pablo Su´ arez-Serrato, and Guofang W ei, Volume entropy and rigidity for rcd-spaces , arXiv:2411.04327 [math.DG], 2024

Show all 37 references
  1. [9]

    de la Harpe, Topics in geometric group theory, Chicago Lectures in Mathematics, University of Chicago Press, 2000

    P. de la Harpe, Topics in geometric group theory, Chicago Lectures in Mathematics, University of Chicago Press, 2000

  2. [10]

    Drut ¸u and M

    C. Drut ¸u and M. Kapovich, Geometric group theory , American Mathematical Society Collo- quium Publications, vol. 63, American Mathematical Societ y, Providence, RI, 2018, With an appendix by Bogdan Nica

  3. [11]

    Flajolet and R

    P. Flajolet and R. Sedgewick, Analytic combinatorics, Cambridge University Press, 2009

  4. [12]

    3, 1427–1470

    Koji Fujiwara and Zlil Sela, The rates of growth in a hyperbolic group , Inventiones mathe- maticae 233 (2023), no. 3, 1427–1470

  5. [13]

    149–166, Springer Berlin Heidelberg, Berlin, Heidelbe rg, 2002

    Alex Furman, Coarse-geometric perspective on negatively curved manifo lds and groups , pp. 149–166, Springer Berlin Heidelberg, Berlin, Heidelbe rg, 2002

  6. [14]

    1, 67–83

    Martin Greendlinger, Dehn ’s algorithm for the word problem , Communications on Pure and Applied Mathematics 13 (1960), no. 1, 67–83

  7. [15]

    Grigorchuk and Pierre de la Harpe, On problems related to growth, entropy, and spectrum in group theory , Journal of Dynamical and Control Systems 3 (1997), 51–89

    Rostislav I. Grigorchuk and Pierre de la Harpe, On problems related to growth, entropy, and spectrum in group theory , Journal of Dynamical and Control Systems 3 (1997), 51–89

  8. [16]

    Gromov, Hyperbolic groups, Springer New York, New York, NY, 1987

    M. Gromov, Hyperbolic groups, Springer New York, New York, NY, 1987

  9. [17]

    , Asymptotic invariants of infinite groups , Institut des Hautes Etudes Scientifiques [IHES], 1992. 32 D. (M.) HUA

  10. [18]

    Hille and R.S

    E. Hille and R.S. Phillips, Functional analysis and semi-groups , American Mathematical Society Colloquium publications, no. v. 31; v. 1957, Americ an Mathematical Society, 1957

  11. [19]

    1, 139–165

    Fran¸ cois Ledrappier and Seonhee Lim, Volume entropy of hyperbolic buildings , Journal of Modern Dynamics 4 (2010), no. 1, 139–165

  12. [20]

    3, 849 – 867

    Pascal Lezaud, Chernoff-type bound for finite Markov chains , The Annals of Applied Proba- bility 8 (1998), no. 3, 849 – 867

  13. [21]

    10, 5089–5100

    Seonhee Lim, Minimal volume entropy for graphs , Transactions of the American Mathematical Society 360 (2008), no. 10, 5089–5100

  14. [22]

    Lyndon and Paul E

    Roger C. Lyndon and Paul E. Schupp, Combinatorial group theory , Classics in Mathematics, no. 1, Springer Berlin, Heidelberg, 2001

  15. [23]

    Avinoam Mann, How groups grow , London Mathematical Society Lecture Note Series, Cam- bridge University Press, 2011

  16. [24]

    3, 567–573

    Anthony Manning, Topological entropy for geodesic flows , Annals of Mathematics 110 (1979), no. 3, 567–573. 25. , Relating exponential growth in a manifold and its fundament al group , Proceedings of the American Mathematical Society 133 (2005), no. 4, 995–997

  17. [26]

    McMullen, Entropy and the clique polynomial , Journal of Topology 8 (2014), no

    Curtis T. McMullen, Entropy and the clique polynomial , Journal of Topology 8 (2014), no. 1, 184–212

  18. [27]

    Milnor, A note on curvature and fundamental group , Journal of Differential Geom- etry 2 (1968), 1–7

    John W. Milnor, A note on curvature and fundamental group , Journal of Differential Geom- etry 2 (1968), 1–7

  19. [28]

    Myers, Forbidden substrings on weighted alphabets , Australas

    Amy N. Myers, Forbidden substrings on weighted alphabets , Australas. J Comb. 45 (2009), 59–66

  20. [29]

    Yann Ollivier, Sharp phase transition theorems for hyperbolicity of rando m groups, Geometric & Functional Analysis GAF A 14 (2003), 595–679

  21. [30]

    , A january 2005 invitation to random groups , Ensaios Matem´ aticos (2005)

  22. [31]

    , Growth exponent of generic groups , Commentarii Mathematici Helvetici 81 (2006), 569–593

  23. [32]

    3, 914–942

    Eduardo Oreg´ on-Reyes, The space of metric structures on hyperbolic groups , Journal of the London Mathematical Society 107 (2023), no. 3, 914–942

  24. [33]

    4, 345–363

    Andrea Sambusetti, Growth tightness of surface groups , Expositiones Mathematicae 20 (2002), no. 4, 345–363

  25. [34]

    A. S. Schwarz, A volume invariant of coverings , Doklady Akad. Nauk SSSR 105 (1955), 32–34, In Russian. Zbl:0066.15903

  26. [35]

    4, 510–515

    AG Shukhov, On the dependence of the growth rate on the length of the defini ng relator , Mathematical Notes 65 (1999), no. 4, 510–515

  27. [36]

    1, 93–133

    Alistair Sinclair and Mark Jerrum, Approximate counting, uniform generation and rapidly mixing markov chains , Information and Computation 82 (1989), no. 1, 93–133

  28. [37]

    Antoine Song, Entropy and stability of hyperbolic manifolds , arXiv:2302.07422 [math.DG], 2023

  29. [38]

    Wilson, On exponential growth and uniformly exponential growth for groups, Inven- tiones mathematicae 155 (2004), 287–303

    John S. Wilson, On exponential growth and uniformly exponential growth for groups, Inven- tiones mathematicae 155 (2004), 287–303. Email address : dongming@ucsb.edu

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.