Pith. sign in

REVIEW 5 minor 43 references

Sharp Poincar\'e Interpolation Along Wasserstein Geodesics

T0 review · 0 major / 5 minor · reviewed 2026-07-14 · grok-4.5

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

Pith's one-line read Along a Wasserstein geodesic, the Poincaré scale is at most the linear interpolation of the reciprocal square roots of the endpoint curvatures.

desk verdict Sharp Poincaré interpolation along true Wasserstein geodesics, with a clean two-endpoint Bochner method and a resolved open question; the math holds up. read the letter →

arxiv 2607.10769 v1 pith:FBEDXNSQ submitted 2026-07-12 math.PR math.FAmath.MGmath.SP

Bang-Xian Han , Zhuo-Nan Zhu This is my paper · ORCID
classification math.PRmath.FAmath.MGmath.SP MSC 49Q2260E1535P1553C2152A40
keywords Poincaréinequalitystronglog-concavityoptimaltransportWassersteingeodesicGaussianentropyBrunn–Minkowskitwo-endpointBochnermethod
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 if two probability measures on Euclidean space each satisfy a strong log-concavity bound, then every intermediate measure on their quadratic optimal-transport geodesic has a Poincaré constant controlled by a simple weighted sum of the endpoint scales. The bound is sharp: pure Gaussians attain it, and equality at an interior time occurs only when both endpoints share a Gaussian factor of the right variance in a common direction. The argument never needs a curvature lower bound on the intermediate measures themselves. Instead it solves one transport-adapted Poisson equation that produces compatible vector fields at the two ends, applies the Bochner identity separately at those ends, and interpolates the resulting estimates along the Brenier map. As a special case the same estimate answers a question about odd test functions between even measures and yields finite-time distortion inequalities for translation-reduced Gaussian entropy together with centroid-corrected Gaussian Brunn–Minkowski inequalities.

What carries the argument

The two-endpoint Bochner method: a single Brenier-adapted Poisson equation produces dual Hessian symmetries at source and target, allowing independent sharp Bochner estimates that are then affinely interpolated along the optimal map, bypassing intermediate curvature.

What would settle it

Exhibit two strongly log-concave measures whose intermediate Poincaré constant on the optimal geodesic exceeds the claimed linear combination of reciprocal square roots, or find an interior equality case that does not split off matching Gaussian factors in a common direction.

Watch

Extended reading notes

Core claim

If μ0 and μ1 are κi-strongly log-concave (potentials allowed to be extended-valued) and (μt) is their quadratic Wasserstein geodesic, then the square root of the Poincaré constant of μt is at most (1−t)/√κ0 + t/√κ1 for every t and every test function. Equality at an interior time holds if and only if both endpoints factor off Gaussians of variances κi−1 in a common direction; that common subspace is maximal.

Load-bearing premise

The regularization that smooths the endpoint potentials must preserve their exact curvature lower bounds while producing W2-convergent Brenier maps to which the smooth Poisson–Bochner argument still applies.

Editorial extensions

If this is right

  • Odd functions between even 1-strongly log-concave measures satisfy the Poincaré inequality with constant 1 along the optimal geodesic, answering the open question of Aishwarya–Rotem.
  • Translation-reduced Gaussian relative entropy obeys a finite-time (2,n)-distortion inequality along the whole geodesic, extending earlier even-case results to arbitrary barycenters.
  • Convex sets with equal Gaussian centroids satisfy the dimensional Gaussian Brunn–Minkowski inequality; unequal centroids receive an explicit exponential correction involving the interpolated centroid.
  • Equality analysis identifies the maximal common Gaussian subspace of any two strongly log-concave measures and yields a lineality-space criterion for conditioned Gaussians.

Reading between the lines

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

  • The same Poisson-lift idea may extend to multi-marginal Wasserstein barycenters if a single scalar equation can be made compatible with several Brenier Hessians simultaneously, which the paper leaves open.
  • Because the method never uses intermediate Hessians, it may apply to other spectral quantities (log-Sobolev, spectral gaps of higher-order operators) along displacement interpolations where only endpoint curvature is known.
  • The simultaneous-rigidity characterization suggests a quantitative stability version: small deficit in the Poincaré constant at one interior time should force both endpoints to be close to product Gaussians in a common direction.
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

0 major / 5 minor

Summary. The paper proves a sharp interpolation bound for the Poincaré constant along the quadratic Wasserstein geodesic between two strongly log-concave measures on R^n. If μ_i is κ_i-strongly log-concave (potentials allowed to be extended-valued) and (μ_t) is the Brenier displacement interpolation, then √C_P(μ_t) ≤ (1−t)/√κ_0 + t/√κ_1 for every t ∈ [0,1] and every C^∞_c test function (Theorem 1.1). The constant is optimal by explicit Gaussian examples. Equality at an interior time holds if and only if both endpoints split off Gaussian factors of variances κ_i^{−1} in a common direction (Theorem 1.3); the equality directions form the maximal common Gaussian subspace. As a special case the authors answer a question of Aishwarya–Rotem on odd functions between even endpoints. The argument is a two-endpoint Bochner method: a transport-adapted Poisson equation produces compatible fields at source and target; dual Hessian symmetries make the vector Bochner identity coercive separately at each endpoint; the resulting L^2 bounds are then affinely interpolated along the Brenier map. Regularization (Gaussian convolution + compensating dilation), Caffarelli bounds, and W_2 stability extend the smooth case to general extended-valued potentials. Applications include a finite-time (2,n)-distortion inequality for translation-reduced Gaussian entropy and a Gaussian Brunn–Minkowski inequality under a centroid condition.

Significance. The result is a clean, sharp spectral interpolation theorem that converts endpoint curvature into a Poincaré bound along the entire geodesic without any intermediate Hessian lower bound. The two-endpoint Bochner method is a genuine methodological contribution: it separates the source and target estimates and only combines them at the final interpolation step, which is what yields the optimal coefficient and the simultaneous rigidity analysis. The paper resolves an open question of Aishwarya–Rotem for the optimal coupling (rather than a common-Gaussian-source coupling), and the entropy and Brunn–Minkowski consequences extend previous even/symmetric results to the non-symmetric setting under a barycenter condition. Optimality is verified by explicit Gaussians, equality is fully characterized, and the approximation scheme is carried out with explicit estimates. These are strengths that make the manuscript suitable for a strong probability/analysis journal.

minor comments (5)
  1. Abstract and introduction: the phrase “one of the most important contribution at the methodological level” is slightly ungrammatical; “contributions” or a more restrained formulation would read better.
  2. Proposition 2.3 is labeled “Theorem 2.3” in the proof of Theorem 1.1 and in the strategy outline; the numbering should be made consistent throughout.
  3. Section 4, display (4.8): the chain of inequalities is clear, but a short sentence noting that the asymptotic sharpness of the triangle inequality forces both the endpoint norms and the inner product to saturate would help the reader track the passage to (4.9)–(4.10).
  4. Corollary 4.3: the notation “K_i = int K_i” appears to be a typographical slip for the closure; a one-line clarification that the boundary has measure zero would remove any ambiguity.
  5. A few references (e.g., the arXiv numbers of concurrent works) could be updated if final versions are available, but this is purely bibliographic.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the sharp Poincaré interpolation is derived from endpoint Bochner identities, Brenier structure, and a controlled approximation, not from self-definition or fitted inputs.

full rationale

The central claim (Theorem 1.1) is obtained by solving a transport-adapted Poisson equation at the source, producing dual Hessian symmetries that make the Euclidean vector Bochner identity coercive separately at each endpoint (Proposition 2.2), then affinely interpolating the resulting fields along the Brenier map (Theorem 2.3). The general case follows by a standard regularization (Gaussian convolution plus compensating dilation) that preserves the exact curvature lower bounds κ i (Lemma 3.2), together with Caffarelli bounds on the Brenier Hessian (Lemma 3.3) and W2-stability of the geodesic. Optimality is verified by explicit isotropic Gaussians (Remark 1.2), not by fitting. Equality rigidity (Theorem 1.3) likewise follows from near-extremizers forcing both endpoint deficits to vanish and a common affine direction, then Gaussian integration by parts. Self-citations (e.g., Han [24], Han–Liu [25]) appear only as background or open questions and are not load-bearing for the main inequality. No step reduces the claimed spectral bound to its own definition or to a parameter fitted from the target quantity. The derivation is therefore self-contained against external benchmarks.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

The paper rests on standard Euclidean analysis, Brenier’s theorem, the classical Brascamp–Lieb inequality and Caffarelli’s contraction theorem. No free parameters are fitted; the only ‘invented’ object is a methodological construction (the transport-adapted Poisson lift) that is fully defined and used inside the proofs.

assumptions (5)
  • standard math Brenier’s theorem: the quadratic optimal transport map from an absolutely continuous measure is the gradient of a convex potential.
    Used throughout to define the displacement interpolation and the Hessian H = D²Φ.
  • standard math Classical Brascamp–Lieb inequality: a κ-strongly log-concave measure satisfies CP ≤ 1/κ.
    Supplies the endpoint spectral gaps that are interpolated.
  • standard math Caffarelli’s contraction theorem (two-potential form) giving Lipschitz bounds on the Brenier map under two-sided Hessian bounds.
    Invoked in Lemma 3.3 to obtain global diffeomorphism and Hessian bounds for the regularized maps.
  • standard math Euclidean vector Bochner identity for the weighted divergence (Lemma 2.1).
    The identity is classical; the paper only records a short proof for completeness.
  • domain assumption Strong log-concavity of the endpoint measures (Vi − κi|x|²/2 convex, possibly extended-valued).
    The standing hypothesis of Theorems 1.1 and 1.3.
invented entities (1)
  • Transport-adapted Poisson lift (X,Y) generated by −divμ0(H⁻¹∇u)=g
    purpose: Produces compatible divergence fields at source and target that inherit dual Hessian symmetries, allowing separate sharp Bochner estimates.
    Defined and used only inside the paper; no claim of independent physical or geometric existence outside the proof.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sharp Poincar\'e Interpolation Along Wasserstein Geodesics." pith.science (2026). https://pith.science/paper/FBEDXNSQ

@misc{pith2026260710769,
  author       = {Pith},
  title        = {Pith review of: Sharp Poincar\'e Interpolation Along Wasserstein Geodesics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FBEDXNSQ}},
  note         = {Machine review of arXiv:2607.10769}
}
abstract

We prove a sharp interpolation inequality for the Poincar\'e constant along quadratic Wasserstein geodesics. Let $\mu_i$, $i=0,1$, be $\kappa_i$-strongly log-concave probability measures on $\mathbb R^n$, and let $(\mu_t)_{t\in[0,1]}$ be their optimal displacement interpolation. Then \[ \sqrt{C_P(\mu_t)} \leq \frac{1-t}{\sqrt{\kappa_0}} + \frac{t}{\sqrt{\kappa_1}}. \] This estimate is optimal for every $t,\kappa_0,\kappa_1$, holds for all test functions without symmetry assumptions, and remains valid for extended-valued potentials. We also characterize equality at an interior time: it holds if and only if the two endpoints split off curvature-saturating Gaussian factors in a common direction. The equality directions form the maximal subspace on which both endpoints have the corresponding Gaussian factors. As a special case, we resolve a question of Aishwarya and Rotem concerning odd functions along optimal interpolations between even strongly log-concave measures. The proof uses a two-endpoint Bochner method, which is also one of the most important contribution at the methodological level: it converts curvature information available only at the endpoints directly into a sharp spectral estimate along the connecting geodesic, bypassing the generally inaccessible curvature of the intermediate measures.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 15 canonical work pages

  1. [1]

    Agueh and G

    M. Agueh and G. Carlier,Barycenters in the Wasserstein space, SIAM J. Math. Anal.43 (2011), no. 2, 904–924. 7 27

    2011

  2. [2]

    Aishwarya and D

    G. Aishwarya and D. Li,Entropic and functional forms of the dimensional Brunn–Minkowski inequality in Gauss space, Math. Ann.393(2025), 3025–3042, doi:10.1007/s00208-025-03294-4. 3, 4, 7

    work page doi:10.1007/s00208-025-03294-4 2025

  3. [3]

    Aishwarya and L

    G. Aishwarya and L. Rotem,New Brunn–Minkowski and functional inequalities via convexity of entropy, Adv. Math.490(2026), 110841, doi:10.1016/j.aim.2026.110841. 3, 4, 7, 20, 24

    work page doi:10.1016/j.aim.2026.110841 2026

  4. [4]

    Bolley, I

    F. Bolley, I. Gentil, and A. Guillin,Dimensional improvements of the logarithmic Sobolev, Tala- grand and Brascamp–Lieb inequalities, Ann. Probab.46(2018), no. 1, 261–301, doi:10.1214/17- AOP1184. 7

    doi:10.1214/17- 2018

  5. [5]

    S. G. Bobkov, N. Gozlan, C. Roberto, and P.-M. Samson,Bounds on the deficit in the logarithmic Sobolev inequality, J. Funct. Anal.267(2014), no. 11, 4110–4138. 25

    2014

  6. [6]

    H. J. Brascamp and E. H. Lieb,On extensions of the Brunn–Minkowski and Pr´ ekopa–Leindler theorems, including inequalities for log concave functions, and with an application to the diffusion equation, J. Funct. Anal.22(1976), 366–389. 2, 26

    1976

  7. [7]

    Brenier,Polar factorization and monotone rearrangement of vector-valued functions, Comm

    Y. Brenier,Polar factorization and monotone rearrangement of vector-valued functions, Comm. Pure Appl. Math.44(1991), 375–417. 3, 12, 14

    1991

  8. [8]

    L. A. Caffarelli,The regularity of mappings with a convex potential, J. Amer. Math. Soc.5 (1992), no. 1, 99–104. 13

    1992

Show all 43 references
  1. [9]

    L. A. Caffarelli,Monotonicity properties of optimal transportation and the FKG and related inequalities, Comm. Math. Phys.214(2000), no. 3, 547–563; erratum, Comm. Math. Phys. 225(2002), no. 2, 449–450. 12

    2000

  2. [10]

    Cheng and D

    X. Cheng and D. Zhou,Eigenvalues of the drifted Laplacian on complete metric measure spaces, Commun. Contemp. Math.19(2017), no. 1, 1650001, 17 pp. 7

    2017

  3. [11]

    Cordero-Erausquin and A

    D. Cordero-Erausquin and A. Eskenazis,Concavity principles for weighted marginals, arXiv:2506.16941 (2025). 7

    2025 arXiv

  4. [12]

    Cordero-Erausquin, R

    D. Cordero-Erausquin, R. J. McCann, and M. Schmuckenschl¨ ager,A Riemannian interpo- lation inequality ` a la Borell, Brascamp and Lieb, Invent. Math.146(2001), no. 2, 219–257, doi:10.1007/s002220100160. 2

    2001 doi

  5. [13]

    T. A. Courtade and M. Fathi,Stability of the Bakry– ´Emery theorem on Rn, J. Funct. Anal. 279(2020), no. 2, 108523, 28 pp. 7

    2020

  6. [14]

    T. A. Courtade, M. Fathi, and A. Pananjady,Existence of Stein kernels under a spectral gap, and discrepancy bounds, Ann. Inst. Henri Poincar´ e Probab. Stat.55(2019), no. 2, 777–790, doi:10.1214/18-AIHP898. 5

    2019 doi

  7. [15]

    De Philippis and A

    G. De Philippis and A. Figalli,The Monge–Amp` ere equation and its link to optimal transporta- tion, Bull. Amer. Math. Soc. (N.S.)51(2014), no. 4, 527–580. 13

    2014

  8. [16]

    De Philippis and A

    G. De Philippis and A. Figalli,Rigidity and stability of Caffarelli’s log-concave perturbation theorem, Nonlinear Anal.154(2017), 59–70. 7

    2017

  9. [17]

    M. D. Donsker and S. R. S. Varadhan,Asymptotic evaluation of certain Markov process expectations for large time. IV, Comm. Pure Appl. Math.36(1983), no. 2, 183–212. 22

    1983

  10. [18]

    Erbar, K

    M. Erbar, K. Kuwada, and K.-T. Sturm,On the equivalence of the entropic curvature-dimension condition and Bochner’s inequality on metric measure spaces, Invent. Math.201(2015), no. 3, 993–1071. 20

    2015

  11. [19]

    Eskenazis, A

    A. Eskenazis, A. Giannopoulos, and N. Tziotziou,Functional perimeter and the dimensional Brunn–Minkowski inequality for log-concave measures, arXiv:2605.02747 (2026). 7

    2026 arXiv

  12. [20]

    Eskenazis and G

    A. Eskenazis and G. Moschidis,The dimensional Brunn–Minkowski inequality in Gauss space, J. Funct. Anal.280(2021), 108914. 4

    2021

  13. [21]

    Eskenazis and Y

    A. Eskenazis and Y. Shenfeld,Intrinsic dimensional functional inequalities on model spaces, J. Funct. Anal.286(2024), no. 7, Paper No. 110338, 56 pp. 7

    2024

  14. [22]

    Fathi,Stein kernels and moment maps, Ann

    M. Fathi,Stein kernels and moment maps, Ann. Probab.47(2019), no. 4, 2172–2185, doi:10.1214/18-AOP1305. 5 28

    2019 doi

  15. [23]

    R. J. Gardner and A. Zvavitch,Gaussian Brunn–Minkowski inequalities, Trans. Amer. Math. Soc.362(2010), no. 10, 5333–5353, doi:10.1090/S0002-9947-2010-04891-3. 4

    2010 doi

  16. [24]

    Han,Measure rigidity of synthetic lower Ricci curvature bound on Riemannian manifolds, Adv

    B.-X. Han,Measure rigidity of synthetic lower Ricci curvature bound on Riemannian manifolds, Adv. Math.373(2020), 107327, 31 pp., doi:10.1016/j.aim.2020.107327. 2

    2020 doi

  17. [25]

    Han and D.-Y

    B.-X. Han and D.-Y. Liu,Wasserstein barycenter convexity detects Hilbertian geometry, arXiv:2606.28213 (2026). 8

    2026 arXiv

  18. [26]

    Jiang and P

    R. Jiang and P. Koskela,Isoperimetric inequality from the Poisson equation via curvature, Comm. Pure Appl. Math.65(2012), no. 8, 1145–1168, doi:10.1002/cpa.21405. 5

    2012 doi

  19. [27]

    Jiang, P

    R. Jiang, P. Koskela, and D. Yang,Isoperimetric inequality via Lipschitz regularity of Cheeger-harmonic functions, J. Math. Pures Appl. (9)101(2014), no. 5, 583–598, doi:10.1016/j.matpur.2013.07.002. 5

    2014 doi

  20. [28]

    Klartag,Poincar´ e inequalities and moment maps, Ann

    B. Klartag,Poincar´ e inequalities and moment maps, Ann. Fac. Sci. Toulouse Math. (6)22 (2013), no. 1, 1–41. 7

    2013

  21. [29]

    Klartag and E

    B. Klartag and E. Putterman,Spectral monotonicity under Gaussian convolution, Ann. Fac. Sci. Toulouse Math. (6)32(2023), no. 5, 939–967, doi:10.5802/afst.1759. 7

    2023 doi

  22. [30]

    B. B. Klartag and A. V. Kolesnikov,Eigenvalue distribution of optimal transportation, Anal. PDE8(2015), no. 1, 33–55, doi:10.2140/apde.2015.8.33. 7

    2015 doi

  23. [31]

    A. V. Kolesnikov,Mass transportation and contractions, MIPT Proc.2(2010), no. 4, 90–99; in Russian, with English translation available as arXiv:1103.1479. 12

    2010 arXiv

  24. [32]

    A. V. Kolesnikov,Hessian metrics, CD(K, N)-spaces, and optimal transportation of log-concave measures, Discrete Contin. Dyn. Syst.34(2014), no. 4, 1511–1532. 7, 9

    2014

  25. [33]

    A. V. Kolesnikov and G. V. Livshyts,On the Gardner–Zvavitch conjecture: symmetry in inequalities of Brunn–Minkowski type, Adv. Math.384(2021), 107689. 4

    2021

  26. [34]

    A. V. Kolesnikov and E. Milman,Riemannian metrics on convex sets with applications to Poincar´ e and log-Sobolev inequalities, Calc. Var. Partial Differential Equations55(2016), no. 4, Art. 77, doi:10.1007/s00526-016-1018-3. 7

    2016 doi

  27. [35]

    A. V. Kolesnikov and E. Milman,Poincar´ e and Brunn–Minkowski inequalities on the boundary of weighted Riemannian manifolds, Amer. J. Math.140(2018), no. 5, 1147–1185. 7

    2018

  28. [36]

    Malliaris, J

    A. Malliaris, J. Melbourne, C. Roberto, and M. Roysdon,Functional liftings of restricted geometric inequalities, arXiv:2508.15247v3 (2025). 7

    2025

  29. [37]

    Mijoule, M

    G. Mijoule, M. Raiˇ c, G. Reinert, and Y. Swan,Stein’s density method for multivariate continuous distributions, Electron. J. Probab.28(2023), Paper No. 59, 40 pp., doi:10.1214/22-EJP883. 5

    2023 doi

  30. [38]

    Nayar and T

    P. Nayar and T. Tkocz,A note on a Brunn–Minkowski inequality for the Gaussian measure, Proc. Amer. Math. Soc.141(2013), no. 11, 4027–4030, doi:10.1090/S0002-9939-2013-11609-6. 4

    2013 doi

  31. [39]

    Otto and C

    F. Otto and C. Villani,Generalization of an inequality by Talagrand and links with the logarith- mic Sobolev inequality, J. Funct. Anal.173(2000), no. 2, 361–400, doi:10.1006/jfan.1999.3557. 2, 23

    2000 doi

  32. [40]

    Serres,Behavior of the Poincar´ e constant along the Polchinski renormalization flow, Commun

    J. Serres,Behavior of the Poincar´ e constant along the Polchinski renormalization flow, Commun. Contemp. Math.26(2024), no. 7, 2350035, doi:10.1142/S0219199723500359. 7

    2024 doi

  33. [41]

    Santambrogio and X.-J

    F. Santambrogio and X.-J. Wang,Convexity of the support of the displacement interpolation: counterexamples, Appl. Math. Lett.58(2016), 152–158, doi:10.1016/j.aml.2016.02.016. 2

    2016 doi

  34. [42]

    Villani,Optimal Transport: Old and New, Grundlehren der mathematischen Wissenschaften, vol

    C. Villani,Optimal Transport: Old and New, Grundlehren der mathematischen Wissenschaften, vol. 338, Springer-Verlag, Berlin, 2009, doi:10.1007/978-3-540-71050-9. 13

    2009 doi

  35. [43]

    von Renesse and K.-T

    M.-K. von Renesse and K.-T. Sturm,Transport inequalities, gradient estimates, entropy and Ricci curvature, Comm. Pure Appl. Math.58(2005), no. 7, 923–940, doi:10.1002/cpa.20060. 2 DeclarationThe authors declare no competing interests. No datasets were generated or analyzed in ...

    2005 doi

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.