Pith. sign in

REVIEW 2 major objections 7 minor 62 references

The KLS isoperimetric constant for isotropic log-concave measures is at most a constant times the fourth root of log n.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-31 22:12 UTC pith:H23T5F5W

load-bearing objection Solid new quadratic Poincaré with sharp constant 2; the log^{1/4} claim hangs on one uncited extraction from Klartag that needs a precise statement before the improvement is bankable. the 2 major comments →

arxiv 2607.24164 v1 pith:H23T5F5W submitted 2026-07-27 math.PR math.FAmath.MG

The KLS constant is O(log^(1/4) n)

Brayden Letwin This is my paper
classification math.PR math.FAmath.MG MSC 52A2360E1546B09
keywords KLS conjecturelog-concave measuresPoincaré inequalitymoment measuresStein kernelsisoperimetric constantquadratic forms
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper improves the best known bound on the Kannan–Lovász–Simonovits (KLS) constant ψ_n, which controls how well isotropic log-concave probability measures on R^n satisfy a Poincaré inequality. It proves that every quadratic form on such a measure already obeys a Poincaré inequality with absolute constant 2, independent of dimension. Feeding that fact into an existing spectral-gap comparison then yields ψ_n ≤ C (log n)^{1/4}. A sympathetic reader cares because the full KLS conjecture (that ψ_n stays bounded) is a central open problem in high-dimensional convex geometry, and every improvement tightens the known control on thin-shell concentration, sampling algorithms, and related isoperimetric questions.

Core claim

For every isotropic log-concave random vector X in R^n and every symmetric matrix M, the quadratic form ⟨MX, X⟩ satisfies Var(⟨MX, X⟩) ≤ 2 E|∇⟨MX, X⟩|². Applying the inequality to the third-moment matrices that define the parameter κ_n shows that κ_n is bounded by an absolute constant; combined with a Lichnerowicz-type comparison this produces the improved bound ψ_n ≤ C log^{1/4} n.

What carries the argument

Moment-map transport: the measure μ is realized as the push-forward of e^{-φ} dy under ∇φ, yielding a Stein kernel τ_μ = ∇²φ ∘ (∇φ)^{-1}. Differentiating the Monge–Ampère equation for φ produces a pointwise identity that, after Brascamp–Lieb and an H^{-1} estimate, controls the Hilbert–Schmidt norm of the transported Stein kernel and therefore the variance of every quadratic form.

Load-bearing premise

The final step from a bounded third-moment parameter to the fourth-root-log bound on the KLS constant relies on reading a specific spectral-gap comparison out of earlier inequalities that were not stated in exactly that form.

What would settle it

Exhibit a single isotropic log-concave measure in some dimension n for which the Poincaré constant of a quadratic form exceeds 2, or for which the third-moment Hilbert–Schmidt norms grow with n; either would break the claimed chain.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The KLS constant is now known to grow no faster than a constant times (log n)^{1/4}.
  • Every isotropic log-concave measure satisfies a dimension-free Poincaré inequality when restricted to quadratic forms, with sharp constant 2.
  • The third-moment parameter κ_n that appears in stochastic-localization arguments is bounded by an absolute constant.
  • Average Kolmogorov distance of one-dimensional marginals to the Gaussian improves to O(log n / n) for centrally symmetric isotropic log-concave laws.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the same moment-map identity can be pushed to higher-degree polynomials, the remaining logarithmic factors in related thin-shell and slicing bounds may continue to fall.
  • The sharp constant 2 for quadratics suggests that the obstruction to a fully dimension-free KLS bound, if any, must live in functions of higher complexity than degree two.
  • The reduction of κ_n to an absolute constant isolates the remaining logarithmic loss inside the spectral-gap comparison itself, offering a concrete target for further improvement.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 7 minor

Summary. The paper proves a Poincaré inequality for quadratic forms over isotropic log-concave measures (Theorem 1.2): Var(⟨MX,X⟩) ≤ 2·E|∇⟨MX,X⟩|² for every symmetric M, with sharp constant 2 (attained by isotropic exponential coordinates with M=Id). The proof works on the moment-measure side ν = e^{−φ}dy (Cordero-Erausquin–Klartag): differentiating the Monge–Ampère equation twice yields Lemma 2.4; an L(∇²φ) identity combined with a PSD comparison (2.12) and Brascamp–Lieb gives the key estimate Theorem 2.5, E Tr(B∇²φB∇²φ) ≤ 2 Tr(B²); Fathi's Stein kernel τ_μ = ∇²φ∘(∇φ)^{−1} and the Barthe–Klartag H^{−1} inequality, applied after absorbing |M| into the test vector (2.21), convert this into Theorem 1.2. Applied to M = E[⟨X,θ⟩X⊗X] it gives κ_n ≤ 2√2, and quoting Klartag's improved Lichnerowicz inequality as C_P(μ) ≤ C·κ_n√log n yields ψ_n ≤ C log^{1/4} n (Theorem 1.1), improving Klartag's C√log n.

Significance. If correct, this improves the best known KLS bound from C√log n (Klartag 2023) to C log^{1/4} n, and Theorem 1.2 is a strong standalone result: it settles KLS for quadratic forms with the sharp, parameter-free constant 2 (the sharpness example is verifiable by hand), gives the second-order correlation condition of Bobkov–Chistyakov–Götze with constant 8, and hence O(log n/n) average Kolmogorov bounds for marginals of centrally symmetric isotropic log-concave laws. Strengths to credit: the core derivation is self-contained and algebraic (Monge–Ampère differentiation, PSD comparisons, Brascamp–Lieb, Stein identities), the regularity reductions are written out in full in Appendix A, the sharpness constant is falsifiable and checked, and the AI-assistance disclosure is explicit and specific. The one unverified load-bearing link is the extraction of the κ_n-dependence from [42], detailed below; this appears repairable within the manuscript's scope.

major comments (2)
  1. [Proof of Theorem 1.1, p. 3 (final paragraph)] The entire improvement over Klartag's ψ_n ≤ C√log n rests on the sentence 'Klartag's improved Lichnerowicz inequality [42, Theorem 1.3, Corollary 3.2, and the discussion following Equation 3.13] implies C_P(μ) ≤ C·κ_n√log n', followed by the admission that 'this bound is not the final bound that Klartag uses, but tracing his inequalities shows that such a bound appears.' Since κ_n ≤ 2√2 is dimension-free, the exponent on κ_n is immaterial, but the exponent on log n multiplying the κ-dependent term is decisive: if the traced bound instead reads, e.g., C_P(μ) ≤ C₁κ_n log n + C₂√log n, substituting κ_n ≤ 2√2 gives C_P ≲ log n and hence ψ_n ≲ √log n — exactly Klartag 2023, no improvement. The manuscript must state the extracted inequality as a lemma with a verifiable derivation from numbered displays in [42], tracking all terms (including κ-independent log n contributions) so the reader canf
  2. [§1, proof of Theorem 1.1] Related to the previous comment but a distinct presentational gap: Theorem 1.1 is described as following 'immediately' from Theorem 1.2, yet the reduction chain (κ_n bound + traced Lichnerowicz bound + Cheeger/Buser comparison ψ²_n ≤ C·sup C_P) is compressed into a few lines with the middle link uncited (see above). Given that this three-line argument is the paper's headline claim and its only externally dependent step, I recommend expanding it into a self-contained section: (i) state the precise intermediate bound imported from [42] as a displayed lemma, (ii) prove or carefully reference it, and (iii) only then apply κ_n ≤ 2√2. This would also insulate the result against the reasonable reader objection that the log^{1/4} exponent is inherited rather than derived.
minor comments (7)
  1. [§2.1, Eq. (2.12)] The invariance claim used to normalize ∇²φ(y)=Id and diagonalize B needs one sentence of justification: the two terms are contractions of the third-derivative tensor with (∇²φ)^{−1} and B, and their difference transforms covariantly under the required change of variables; as written, 'we see that (2.11) is invariant' is asserted rather than shown.
  2. [§2.1, paragraph preceding Lemma 2.3] 'Wonderfully though, we have the following as a suitable replacement...' and earlier 'we may lose in the fact that ν is isotropic' — the latter is a grammatical error (ν is simply not isotropic in general), and the informal tone ('Wonderfully') should be removed for journal style.
  3. [Notation conflicts] The cutoff in Lemma A.4 is called η, conflicting with η = (|M|^{1/2})#μ introduced in the proof of Theorem 1.2 (§2.2); similarly W is reused for the generic random vector in Definition 2.6 and the Lyapunov function in Lemma A.4. Rename for clarity.
  4. [§2.2, Eq. (2.22)] The identity '8 Tr(M²) = 2·E|∇⟨MX,X⟩|²' uses isotropy via E|2MX|² = 4 Tr(M²Cov(X)) = 4 Tr(M²); this one-line computation is worth displaying, as it is where isotropy enters the final step.
  5. [Definition (2.19) and Proposition 2.8] In (2.19) it would help to note that since f is centered, the test functions g may equivalently be taken centered, matching the convention of Barthe–Klartag [7, Proposition 10] verbatim; as stated the reader must check that the imported Proposition 2.8 uses the same H^{−1} normalization.
  6. [Footnote 1, p. 8] Footnote 1 (motivation and AI provenance for Theorem 2.5) is unusually detailed for a footnote; consider moving it to an acknowledgments section. The transparency itself is welcome and should be retained.
  7. [References] The self-citation [56] ('2026') lacks an arXiv identifier; also the arXiv rendering of the title ('ISO(log 1/4 n)') is garbled in the metadata.

Circularity Check

0 steps flagged

No circularity: quadratic Poincaré is derived from Monge–Ampère/Brascamp–Lieb/Stein identities; the ψ_n bound only multiplies that by an external Klartag estimate.

full rationale

Theorem 1.2 is obtained by differentiating the Cordero–Erausquin–Klartag Monge–Ampère equation for the moment map, integrating the resulting L-identity after nonnegativity checks, comparing the two cubic terms via the elementary (b_i−b_j)² identity after simultaneous diagonalization, applying Brascamp–Lieb entrywise to B^{1/2}∇²φ B^{1/2}, and transporting the resulting HS bound through Fathi’s Stein kernel and the Barthe–Klartag H^{-1} inequality (with the |M|-absorption trick so that the controlled matrix is exactly |M|). None of these steps defines the target variance in terms of itself or fits a parameter later called a prediction. The passage to Theorem 1.1 only inserts the newly proved dimension-free bound κ_n≤2√2 into Klartag’s external Lichnerowicz-type inequality from [42]; that citation is independent prior work by a different author, not a self-citation chain or uniqueness theorem of the present author. Concerns about whether “tracing his inequalities” yields precisely the √log n factor are correctness/auditability issues, not circularity. No equation reduces the claimed Poincaré constant or the log^{1/4} exponent to an input by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 8 axioms · 0 invented entities

Pure analytic proof. No fitted numerical parameters. Load-bearing inputs are standard facts about log-concave measures and a short list of named theorems from the moment-measure / Stein-kernel literature, plus one informally traced consequence of Klartag (2023).

axioms (8)
  • standard math Cordero-Erausquin–Klartag moment-measure existence and uniqueness (Lemma 2.1)
    Used as the starting transport from μ to ν = e^{-φ} dy; cited as [21, Thm 2].
  • standard math Klartag’s regularity and Monge–Ampère properties of the moment map for isotropic log-concave μ (Lemma 2.2)
    Supplies smoothness, positive-definiteness of ∇²φ, and the pointwise Monge–Ampère equation differentiated throughout §2.
  • standard math Fathi’s theorem that τ_μ = ∇²φ ∘ (∇φ)^{-1} is a symmetric Stein kernel (Lemma 2.7)
    Bridge from Theorem 2.5 to the H^{-1} estimate in the proof of Theorem 1.2; cited as [29, Thm 2.3].
  • standard math Barthe–Klartag H^{-1} Poincaré inequality for centered locally Lipschitz functions on log-concave measures (Prop. 2.8)
    Converts Stein-kernel control of linear functionals into a variance bound for the quadratic form.
  • standard math Brascamp–Lieb inequality on the moment-measure space ν (Lemma A.6)
    Closes the variance comparison for the entries of B^{1/2}∇²φ B^{1/2} in the proof of Theorem 2.5.
  • domain assumption Klartag’s improved Lichnerowicz inequality yields C_P(μ) ≤ C·κ_n √log n for isotropic log-concave μ
    Invoked in the proof of Theorem 1.1 without a matching cited theorem statement; author says the form appears by tracing inequalities in [42].
  • domain assumption Isotropy and log-concavity of μ; symmetry of M
    Standing hypotheses of Theorems 1.1–1.2; used for E X⊗X = Id, centering, and convexity of V.
  • ad hoc to paper Approximation by smooth compactly supported isotropic log-concave densities preserves moments through degree 4 (Lemma A.1)
    Justifies working under the regularity assumptions of §2; standard in spirit but the concrete conditioning-and-renormalization argument is paper-specific and AI-written.

pith-pipeline@v1.2.0-grok45-kimik3 · 25195 in / 3679 out tokens · 86527 ms · 2026-07-31T22:12:13.281004+00:00 · methodology

0 comments
read the original abstract

We confirm the Kannan--Lov\'asz--Simonovits conjecture for quadratic forms: if $X \sim \mu$ is an isotropic log-concave random vector in $\mathbb{R}^n$, then for any symmetric matrix $M$ one has $$ \operatorname{Var}_{X \sim \mu}(\langle MX,X\rangle) \leq 2\,\mathbb{E}_{X \sim \mu}|\nabla\langle MX,X\rangle|^2. $$ As an application, we apply the above to $M=\mathbb{E}_{X \sim \mu}(\langle X,\theta\rangle X\otimes X)$ for $\theta\in S^{n-1}$ and show that the Kannan--Lov\'asz--Simonovits constant $\psi_n$ satisfies $$ \psi_n\leq C\log^{1/4}n $$ for some absolute constant $C>0$.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

62 extracted references · 2 linked inside Pith

  1. [1]

    2131, Springer, Cham, 2015

    David Alonso-Gutiérrez and Jesús Bastero,Approaching the Kannan–Lovász–Simonovits and variance conjec- tures, Lecture Notes in Mathematics, vol. 2131, Springer, Cham, 2015. 1

    2015

  2. [2]

    Milla Anttila, Keith Ball, and Irini Perissinaki,The central limit problem for convex bodies, Transactions of the American Mathematical Society355(2003), 4723–4735. 2

    2003

  3. [3]

    Milman,Asymptotic geometric analysis, part I, Mathematical Surveys and Monographs, vol

    Shiri Artstein-Avidan, Apostolos Giannopoulos, and Vitali D. Milman,Asymptotic geometric analysis, part I, Mathematical Surveys and Monographs, vol. 202, American Mathematical Society, Providence, RI, 2015. 1

    2015

  4. [4]

    Milman,Asymptotic geometric analysis, part II, Mathematical Surveys and Monographs, vol

    Shiri Artstein-Avidan, Apostolos Giannopoulos, and Vitali D. Milman,Asymptotic geometric analysis, part II, Mathematical Surveys and Monographs, vol. 261, American Mathematical Society, Providence, RI, 2021. 1

    2021

  5. [5]

    348, Springer, Cham, 2014

    Dominique Bakry, Ivan Gentil, and Michel Ledoux,Analysis and geometry of Markov diffusion operators, Grundlehren der mathematischen Wissenschaften, vol. 348, Springer, Cham, 2014. 6

    2014

  6. [6]

    Keith Ball,Logarithmically concave functions and sections of convex sets inRn, Studia Mathematica88(1988), 69–84. 2

    1988

  7. [7]

    Franck Barthe and Bo’az Klartag,Spectral gaps, symmetries and log-concave perturbations, Bulletin of the Hellenic Mathematical Society64(2020), 1–31. 14, 18

    2020

  8. [8]

    Pierre Bizeul,The slicing conjecture via small ball estimates, To appear in The Annals of Probability; arXiv:2501.06854, 2025. 2

    Pith/arXiv arXiv 2025

  9. [9]

    Bobkov,On isoperimetric constants for log-concave probability distributions, Geometric Aspects of Functional Analysis (Vitali D

    Sergey G. Bobkov,On isoperimetric constants for log-concave probability distributions, Geometric Aspects of Functional Analysis (Vitali D. Milman and Gideon Schechtman, eds.), Lecture Notes in Mathematics, vol. 1910, Springer, Berlin, Heidelberg, 2007, pp. 81–88. 2

    1910

  10. [10]

    Bobkov, Gennadiy P

    Sergey G. Bobkov, Gennadiy P. Chistyakov, and Friedrich Götze,Normal approximation for weighted sums under a second-order correlation condition, The Annals of Probability48(2020), 1202–1219. 3

    2020

  11. [11]

    Bobkov and Alexander Koldobsky,On the central limit property of convex bodies, Geometric Aspects of Functional Analysis (Vitali D

    Sergey G. Bobkov and Alexander Koldobsky,On the central limit property of convex bodies, Geometric Aspects of Functional Analysis (Vitali D. Milman and Gideon Schechtman, eds.), Lecture Notes in Mathematics, vol. 1807, Springer, Berlin, Heidelberg, 2003, pp. 44–52. 2

    2003

  12. [12]

    Christer Borell,Convex set functions ind-space, Periodica Mathematica Hungarica6(1975), 111–136. 1

    1975

  13. [13]

    Jean Bourgain,On high-dimensional maximal functions associated to convex bodies, American Journal of Math- ematics108(1986), 1467–1476. 2

    1986

  14. [14]

    Herm Jan Brascamp and Elliott H. Lieb,On extensions of the Brunn–Minkowski and Prékopa–Leindler theorems, including inequalities for log-concave functions, and with an application to the diffusion equation, Journal of Functional Analysis22(1976), 366–389. 11, 18

    1976

  15. [15]

    196, American Mathematical Society, Prov- idence, RI, 2014

    Silouanos Brazitikos, Apostolos Giannopoulos, Petros Valettas, and Beatrice-Helen Vritsiou,Geometry of isotropic convex bodies, Mathematical Surveys and Monographs, vol. 196, American Mathematical Society, Prov- idence, RI, 2014. 1

    2014

  16. [16]

    Peter Buser,A note on the isoperimetric constant, Annales scientifiques de l’École Normale Supérieure15(1982), 213–230. 2

    1982

  17. [17]

    Gunning, ed.), Princeton Mathematical Series, vol

    Jeff Cheeger,A lower bound for the smallest eigenvalue of the Laplacian, Problems in Analysis: A Symposium in Honor of Salomon Bochner (Robert C. Gunning, ed.), Princeton Mathematical Series, vol. 31, Princeton University Press, Princeton, NJ, 1970, pp. 195–199. 2

    1970

  18. [18]

    Yuansi Chen,An almost constant lower bound of the isoperimetric coefficient in the KLS conjecture, Geometric and Functional Analysis31(2021), 34–61. 2

    2021

  19. [19]

    Yuansi Chen and Ronen Eldan,Localization schemes: A framework for proving mixing bounds for Markov chains, Duke Mathematical Journal174(2025), 1431–1510. 1

    2025

  20. [20]

    Sinho Chewi,Log-concave sampling, Book manuscript, version of June 12, 2026, 2026. 1

    2026

  21. [21]

    Darío Cordero-Erausquin and Bo’az Klartag,Moment measures, Journal of Functional Analysis268(2015), 3834–3866. 5

    2015

  22. [22]

    Thomas A. Courtade, Max Fathi, and Ashwin Pananjady,Existence of Stein kernels under a spectral gap, and discrepancy bounds, Annales de l’Institut Henri Poincaré, Probabilités et Statistiques55(2019), 777–790. 13

    2019

  23. [23]

    Nicolò De Ponti and Andrea Mondino,Sharp Cheeger–Buser type inequalities inRCD(K,∞)-spaces, The Journal of Geometric Analysis31(2021), 2416–2438. 2

    2021

  24. [24]

    Ronen Eldan,Thin shell implies spectral gap up to polylog via a stochastic localization scheme, Geometric and Functional Analysis23(2013), 532–569. 2, 3

    2013

  25. [25]

    Ronen Eldan and Bo’az Klartag,Pointwise estimates for marginals of convex bodies, Journal of Functional Analysis254(2008), 2275–2293. 2

    2008

  26. [26]

    545, American Mathematical Society, Providence, RI, 2011, pp

    Ronen Eldan and Bo’az Klartag,Approximately Gaussian marginals and the hyperplane conjecture, Concentra- tion, Functional Inequalities and Isoperimetry, Contemporary Mathematics, vol. 545, American Mathematical Society, Providence, RI, 2011, pp. 55–68. 2 19

    2011

  27. [27]

    Ronen Eldan, Dan Mikulincer, and Alex Zhai,The CLT in high dimensions: Quantitative bounds via martingale embedding, The Annals of Probability48(2020), 2494–2524. 2

    2020

  28. [28]

    Xiao Fang and Yuta Koike,Sharp high-dimensional central limit theorems for log-concave distributions, Annales de l’Institut Henri Poincaré, Probabilités et Statistiques60(2024), 2129–2156. 2

    2024

  29. [29]

    Max Fathi,Stein kernels and moment maps, The Annals of Probability47(2019), 2172–2185. 8, 13

    2019

  30. [30]

    Max Fathi,Higher-order Stein kernels for Gaussian approximation, Studia Mathematica256(2021), 241–258. 13

    2021

  31. [31]

    Max Fathi and Dan Mikulincer,Stability estimates for invariant measures of diffusion processes, with applications to stability of moment measures and Stein kernels, Annali della Scuola Normale Superiore di Pisa, Classe di Scienze23(2022), 1417–1445. 5, 13

    2022

  32. [32]

    Fleury,Concentration in a thin Euclidean shell for log-concave measures, Journal of Functional Analysis259 (2010), 832–841

    B. Fleury,Concentration in a thin Euclidean shell for log-concave measures, Journal of Functional Analysis259 (2010), 832–841. 2

    2010

  33. [33]

    Olivier Guédon and Emanuel Milman,Interpolating thin-shell and sharp large-deviation estimates for isotropic log-concave measures, Geometric and Functional Analysis21(2011), 1043–1068. 2

    2011

  34. [34]

    Vempala,A slightly improved bound for the KLS constant, 2022, arXiv:2208.11644v2

    Arun Jambulapati, Yin Tat Lee, and Santosh S. Vempala,A slightly improved bound for the KLS constant, 2022, arXiv:2208.11644v2. 2

    Pith/arXiv arXiv 2022

  35. [35]

    Haotian Jiang, Yin Tat Lee, and Santosh S. Vempala,A generalized central limit conjecture for convex bod- ies, Geometric Aspects of Functional Analysis (Bo’az Klartag and Emanuel Milman, eds.), Lecture Notes in Mathematics, vol. 2266, Springer, Cham, 2020, pp. 1–41. 2

    2020

  36. [36]

    Ravi Kannan, László Lovász, and Miklós Simonovits,Isoperimetric problems for convex bodies and a localization lemma, Discrete & Computational Geometry13(1995), 541–559. 1

    1995

  37. [37]

    Ravi Kannan, László Lovász, and Miklós Simonovits,Random walks and anO∗(n5)volume algorithm for convex bodies, Random Structures & Algorithms11(1997), 1–50. 1

    1997

  38. [38]

    Bo’az Klartag,A central limit theorem for convex sets, Inventiones Mathematicae168(2007), 91–131. 2

    2007

  39. [39]

    Bo’az Klartag,Power-law estimates for the central limit theorem for convex sets, Journal of Functional Analysis 245(2007), 284–310. 2

    2007

  40. [40]

    Math- ématiques22(2013), 1–41

    Bo’az Klartag,Poincaré inequalities and moment maps, Annales de la Faculté des Sciences de Toulouse. Math- ématiques22(2013), 1–41. 5

    2013

  41. [41]

    2116, Springer, Cham, 2014, pp

    Bo’az Klartag,Logarithmically-concave moment measures I, Geometric Aspects of Functional Analysis (Bo’az Klartag and Emanuel Milman, eds.), Lecture Notes in Mathematics, vol. 2116, Springer, Cham, 2014, pp. 231–

    2014

  42. [42]

    4, 17 pp

    Bo’az Klartag,Logarithmic bounds for isoperimetry and slices of convex sets, Ars Inveniendi Analytica (2023), Paper No. 4, 17 pp. 2, 3

    2023

  43. [43]

    Kolesnikov,Remarks on curvature in the transportation metric, Analysis Math- ematica43(2017), 67–88

    Bo’az Klartag and Alexander V. Kolesnikov,Remarks on curvature in the transportation metric, Analysis Math- ematica43(2017), 67–88. 5

    2017

  44. [44]

    Bo’az Klartag and Joseph Lehec,Bourgain’s slicing problem and KLS isoperimetry up to polylog, Geometric and Functional Analysis32(2022), 1134–1159. 2, 3

    2022

  45. [45]

    Bo’az Klartag and Joseph Lehec,Affirmative resolution of Bourgain’s slicing problem using Guan’s bound, Geometric and Functional Analysis35(2025), 1147–1168. 2

    2025

  46. [46]

    Bo’az Klartag and Joseph Lehec,Isoperimetric inequalities in high-dimensional convex sets, Bulletin of the American Mathematical Society62(2025), 575–642. 1

    2025

  47. [47]

    Bo’az Klartag and Joseph Lehec,Thin-shell bounds via parallel coupling, 2025, arXiv:2507.15495v2. 2

    arXiv 2025

  48. [48]

    Kolesnikov,Hessian metrics,CD(K, N)-spaces, and optimal transportation of log-concave mea- sures, Discrete and Continuous Dynamical Systems34(2014), 1511–1532

    Alexander V. Kolesnikov,Hessian metrics,CD(K, N)-spaces, and optimal transportation of log-concave mea- sures, Discrete and Continuous Dynamical Systems34(2014), 1511–1532. 5

    2014

  49. [49]

    Alexander V. Kolesnikov and Emanuel Milman,Remarks on the KLS conjecture and Hardy-type inequalities, Geometric Aspects of Functional Analysis: Israel Seminar (GAFA) 2011–2013, Lecture Notes in Mathematics, vol. 2116, Springer, Cham, 2014, pp. 273–292. 2

    2011

  50. [50]

    Alexander V. Kolesnikov and Emanuel Milman,Riemannian metrics on convex sets with applications to Poincaré and log-Sobolev inequalities, Calculus of Variations and Partial Differential Equations55(2016), 77. 5

    2016

  51. [51]

    Kolesnikov and Emanuel Milman,The KLS isoperimetric conjecture for generalized Orlicz balls, The Annals of Probability46(2018), 3578–3615

    Alexander V. Kolesnikov and Emanuel Milman,The KLS isoperimetric conjecture for generalized Orlicz balls, The Annals of Probability46(2018), 3578–3615. 2

    2018

  52. [52]

    Michel Ledoux,Spectral gap, logarithmic Sobolev constant, and geometric bounds, Surveys in Differential Geom- etry9(2004), 219–240. 2

    2004

  53. [53]

    Michel Ledoux, Ivan Nourdin, and Giovanni Peccati,Stein’s method, logarithmic Sobolev and transport inequal- ities, Geometric and Functional Analysis25(2015), 256–306. 13

    2015

  54. [54]

    Vempala,The Kannan–Lovász–Simonovits conjecture, Current Developments in Mathematics 2017, International Press, Somerville, MA, 2019, pp

    Yin Tat Lee and Santosh S. Vempala,The Kannan–Lovász–Simonovits conjecture, Current Developments in Mathematics 2017, International Press, Somerville, MA, 2019, pp. 1–36. 1 20

    2017

  55. [55]

    Vempala,Eldan’s stochastic localization and the KLS conjecture: isoperimetry, concentration and mixing, Annals of Mathematics199(2024), 1043–1092

    Yin Tat Lee and Santosh S. Vempala,Eldan’s stochastic localization and the KLS conjecture: isoperimetry, concentration and mixing, Annals of Mathematics199(2024), 1043–1092. 2

    2024

  56. [56]

    Brayden Letwin and Dan Mikulincer,Dimension-free Gaussian tail estimates for linear functionals on convex bodies, 2026. 2

    2026

  57. [57]

    László Lovász and Miklós Simonovits,Random walks in a convex body and an improved volume algorithm, Random Structures & Algorithms4(1993), 359–412. 1

    1993

  58. [58]

    Dan Mikulincer,A CLT in Stein’s distance for generalized Wishart matrices and higher-order tensors, Interna- tional Mathematics Research Notices2022(2022), 7839–7872. 13

    2022

  59. [59]

    Dan Mikulincer and Yair Shenfeld,The Brownian transport map, Probability Theory and Related Fields190 (2024), 379–444. 1, 13

    2024

  60. [60]

    Emanuel Milman,On the role of convexity in isoperimetry, spectral gap and concentration, Inventiones Mathe- maticae177(2009), 1–43. 2

    2009

  61. [61]

    Grigoris Paouris,Concentration of mass on convex bodies, Geometric and Functional Analysis16(2006), 1021–

    2006

  62. [62]

    5 Department of Mathematics, University of W ashington, Seattle, W ashington 98195 Email address:letwin@uw.edu 21

    Filippo Santambrogio,Dealing with moment measures via entropy and optimal transport, Journal of Functional Analysis271(2016), 418–436. 5 Department of Mathematics, University of W ashington, Seattle, W ashington 98195 Email address:letwin@uw.edu 21

    2016