Pith. sign in

REVIEW 3 major objections 4 minor 28 references

Stationary measures for log-gamma polymer on a strip and in half-space

T0 review · 3 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper proves that, as the width of a log-gamma polymer strip goes to infinity, the stationary measure converges to the half-space stationary measure with drift parameter min{0,u,v}, and it establishes the complete phase diagram for the

desk verdict Strong new results on strip and half-space stationary measures, but a load-bearing sign error in the Beta identity undermines the coexistence-line proofs as written. read the letter →

arxiv 2607.17030 v1 pith:TUU4MKOD submitted 2026-07-19 math.PR

Wlodek Bryc , Jacek Wesolowski This is my paper · ORCID
classification math.PR MSC 60K3582D60
keywords stationarymeasuresdirectedpolymerslog-gammapolymerphasediagramcontourintegralformulasBetaIIrandomvariableshalf-spacestrip
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 is about the stationary measures of the log-gamma directed polymer, first on a finite diagonal strip with two boundary parameters and then in the half-space limit. It establishes a complete phase diagram: the rescaled free energy of the stationary strip polymer converges to one of two deterministic values, or, on the coexistence line, to a random uniform value. Its central theorem says the finite-dimensional stationary law of the strip converges, as the width tends to infinity, to the half-space stationary measure with drift parameter min{0,u,v}. To prove this, the authors develop a new representation of the Laplace transform of the stationary measure by independent Beta II random variables and obtain explicit analytic continuations of the existing contour-integral representation. If correct, this gives a tractable, explicit description of the half-space stationary law and a systematic way to pass from strip to half-space integrable polymer models.

What carries the argument

The main mechanism is a representation of the normalizing constant Z_N(α,u,v) of the stationary density as an expectation involving the perpetuity R_n = 1 + ζ_1 + ζ_1 ζ_2 + ... + ζ_1...ζ_n, with independent Beta II(α+u, α−u) or Beta II(α+v, α−v) variables ζ_i (Beta II(a,b) has density proportional to x^{a−1}/(1+x)^{a+b} on (0,∞)). This turns the 2N-dimensional Laplace-transform integral into a one-dimensional perpetuity expectation whose asymptotics are controlled by exponential functionals of random walks. The paper supplements this with an analytic continuation of the existing one-point and multipoint contour integrals, using Barnes-type beta integral identities, to cover all parameter reg

What would settle it

Take a fixed parameter set, say α=1, u=2, v=1, and numerically evaluate the one-point Laplace transform of the strip stationary measure for large N using (2.1) and (2.2), then compare with the contour integral (1.6) at v~=0; a persistent mismatch refutes Theorem 1.2 in the maximal-current region. For the coexistence line u=v=−1/2, compare the finite-N Laplace transform with the uniform-law Laplace transform; a mismatch refutes Theorem 1.1(iv).

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.2: for bulk parameter α>0 and boundary parameters u,v with u+α>0 and v+α>0, the finite-dimensional stationary law {(L_k^(N))}_{k≥0} of the log-gamma polymer on a strip of width N converges weakly, as N→∞, to the half-space stationary measure {log Z_k}_{k≥0} defined by Z_k = 1/(X_1...X_k) + Σ_{j=1}^k V/(Y_1...Y_j X_j...X_k), where X_j, Y_j are gamma variables and V is gamma(u−v~), with v~=min{0,u,v}. Alongside, Theorem 1.1 gives the phase diagram for the rescaled endpoint: limits −ψ(α), −ψ(α+v), −ψ(α−u), or a uniform law, depending on the signs and relative size of u and v. Theorem 1.3 provides an explicit k-dimensional contour integral for the multipoint Laplac

Load-bearing premise

The entire Laplace-transform machinery rests on the previously established representation of the unique stationary measure of the strip as the law (1.3)-(1.4) with parameters u+α>0, v+α>0; the later asymptotics also lean on the perpetuity identity for R^{-1}, stated in the paper as Beta II(α+u,−2u), which should read Beta(−2u,α+u) since R^{-1}∈(0,1).

Editorial extensions

If this is right

  • As the strip width grows without bound, every boundary-driven phase of the strip stationary measure survives in the half-space limit, with drift parameter equal to the minimum of 0, u, and v.
  • The phase diagram gives the leading-order free energy per site of the stationary strip polymer: deterministic in maximal-current, high-density, and low-density regimes, and uniformly random on the coexistence line.
  • The multipoint contour integral gives an explicit way to compute joint Laplace transforms of the half-space stationary process, such as E[∏_{r=1}^k Z_r^{2t_{r+1}−2t_r}].
  • The Beta II representations provide finite-N formulas for the strip Laplace transforms that are valid outside the maximal-current region and that yield the N→∞ limits used in the proof.

Reading between the lines

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

  • The perpetuity-based method is likely portable: any integrable polymer whose strip stationary measure is a product-density reweighting of log-gamma walks should admit the same strip-to-half-space passage, with the drift parameter read off from the phase diagram.
  • The explicit analytic continuation formulas imply finite-N corrections; one could test numerically whether the convergence rate near the coexistence line is O(1/N), as the expansion behind Lemma 3.1 suggests.
  • The paper's stated identity R^{-1}≃Beta II(α+u,−2u) has a parameter-order slip: R^{-1} lies in (0,1), so the standard law is Beta(−2u,α+u). Correcting it preserves the argument and yields a cleaner derivation of E[R^{−u−v}].
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

3 major / 4 minor

Summary. The paper studies stationary measures of the log-gamma polymer on a finite diagonal strip and in a half-space. Theorem 1.1 establishes the phase diagram for the rescaled endpoint free energy L_N^{(N)}/N, with maximal-current, high-density, low-density, and coexistence-line limits; Theorem 1.2 proves that, as the strip width tends to infinity, the finite-dimensional stationary law of the strip polymer converges weakly to the half-space stationary measure with drift parameter v~ = min{0,u,v}; Theorem 1.3 gives a new multipoint contour integral formula for the Laplace transform of the half-space stationary measure. The proofs combine Beta/Beta II stochastic representations, perpetuity asymptotics, Barnes/de Branges integral identities, and explicit analytic continuation of Barraquand's contour integral representation.

Significance. If the results hold, they give a complete phase diagram for the strip stationary measure, establish a rigorous strip-to-half-space convergence theorem, and provide new contour integral formulas for the half-space stationary measure. The paper is substantial: it contains original Beta II representations of the strip Laplace transform, a self-contained proof of Barraquand's multipoint contour formula, and two different routes to the phase diagram (probabilistic and analytic). These are valuable technical contributions. However, the manuscript contains a recurring misstatement of the law of the perpetuity R^{-1}, which affects the proof of Lemma 3.1 and the normalization formulas used in the proof of Theorem 1.2. The errors appear to be local and correctable, but they are load-bearing as written.

major comments (3)
  1. [Section 2 (after Lemma 2.4) and Section 3.4, Eq. (3.6)] The paper states R^{-1} ~ Beta II(α+u, -2u) and later R^{-1} ~ Beta II(α-w, 2w). Since R >= 1, R^{-1} lies in (0,1], whereas the Beta II distribution has unbounded support (0,∞). The correct law is the standard Beta(-2u, α+u), equivalently Beta(2w, α-w) in the notation of Section 3.4. This is not a purely cosmetic issue: equation (3.6) defines Δ = Γ(α+w)/(Γ(2w)Γ(α-w)) (ψ(α-w)-ψ(α+w)), which is negative for w>0 because ψ is increasing. But E[f(R)] is nonnegative for f(x)=x^{2w}-(x-1)^{2w} ≥ 0. With the correct law, Δ = Γ(α+w)/(Γ(2w)Γ(α-w)) (ψ(α+w)-ψ(α-w)) > 0. As written, Lemma 3.1 and the proof of Theorem 1.1(iv) assert that the nonnegative sequence M_k/k converges to a negative constant. This must be corrected; the surrounding argument appears repairable once the sign is fixed.
  2. [Sections 7.2.2 and 7.2.4] The same Beta II / Beta confusion recurs in the proof of Theorem 1.2. In Section 7.2.2 the text says R^{-1} ~ Beta II(α+v, -2v) and then uses E[R^{-u-v}] = Γ(u-v)Γ(α-v)/(Γ(α+u)Γ(-2v)). The explicit moment formula is consistent with the standard Beta(-2v, α+v) law, not with the printed Beta II statement; the distributional identity must be corrected. The same slip appears in Section 7.2.4 with R^{-1} ~ Beta II(α-w, 2w). These identities are used to normalize the Laplace transforms in the high-density and coexistence cases of Theorem 1.2, so the correction is load-bearing, not merely notational.
  3. [Section 5.2 and Section 6.3] The analytic proof of Theorem 1.1 explicitly omits some boundary cases: Section 6.3 says 'We omit the proof for the cases u−v∈Z≤0 or u,v∈Z≤0, as we did not provide the corresponding versions of Propositions 5.4 and 5.6 in this paper.' If Section 6 is intended as a complete second proof of Theorem 1.1, this is a gap. The probabilistic proof in Section 3 appears to cover these cases once the sign error in Lemma 3.1 is fixed, so the theorem itself is not at risk, but the claim that the paper gives two complete proofs should be adjusted or the missing cases supplied.
minor comments (4)
  1. [Section 2, after Lemma 2.5] The phrase 'Here is the proof of the second part of the lemma.' appears after the proof of Lemma 2.5 and is followed by a repeated argument. This looks like a leftover editing artifact and should be removed or integrated.
  2. [Abstract and Section 1] The abstract contains braces around 'the Laplace transform of'. This is a formatting artifact and should be cleaned up.
  3. [Throughout] The occurrences of 'BetaII' without a space (e.g., in Sections 7.2.2 and 7.2.6) should be made consistent with the 'Beta II' notation defined in Section 2.
  4. [Section 5.2] Several formulas in Section 5.2 are introduced with 'We omit the details' followed by a block of computations. The reader would benefit from a clearer statement of which cases are fully proved and which are only sketched.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the main inputs are prior published theorems and independently proved integral identities; self-citations are not load-bearing.

full rationale

The paper's central inputs are explicitly external: the strip stationary law (1.3)-(1.4) is taken from [BCY24, Theorem 1.6], and the half-space stationary family (1.5) is taken from [BC23, Theorem 1.8]. No parameter is fitted to the target results; the Laplace-transform machinery starts from these published representations and derives new Beta-II, perpetuities, and contour-integral identities. The phase diagram (Theorem 1.1) is proved twice: probabilistically in Section 3 from Corollary 2.2 and Lemma 2.4-2.5, and analytically in Section 6 from Proposition 4.6, whose proof is self-contained. Theorem 1.2 is obtained by computing N→∞ limits of Laplace transforms and identifying them with the known half-space family for v~=min{0,u,v}; the necessary perpetuity asymptotics and exponential-functional asymptotics are cited from external sources [CL91, Xu23, Hir97], not from the conclusions being proved. Theorem 1.3 is proven by induction from de Branges-Barnes integral identities, so it is not a renaming of the strip formulas. Self-citations: [WWoY24] is used only as an analogy for the coexistence-line structure, and [BW19, Theorem A.1] is a standard external theorem used to pass from Laplace transforms to weak convergence; neither reduces the argument to its own conclusion. The reader's flagged Beta-II distributional identity in Section 2 may be a mathematical error, but it is not a circular step: it is an imported external formula, and a wrong imported formula is a correctness issue, not an instance of the paper defining or fitting its target in terms of itself. Therefore the derivation chain is self-contained and non-circular.

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

The paper introduces no new particles, forces, or free parameters. It works with the standard α,u,v model parameters and relies on established representation theorems and classical beta-integrals. The ledger count is therefore low, which is appropriate for a rigorous integrable probability paper.

assumptions (5)
  • domain assumption The unique stationary measure of the strip is the law of (1.4) with joint density (1.3) [BCY24, Theorem 1.6].
    All Laplace-transform formulas in Sections 2–7 start from this representation. The paper does not re-prove existence or uniqueness.
  • domain assumption Perpetuity identities for Beta-II random walks, in particular that for R = 1 + ζ_1 + ζ_1ζ_2 + ... with ζ ~ Beta II(a,b), R^{-1} has a standard Beta distribution with parameters (b-a,a); the paper writes this with a typo as Beta II.
    These identities are imported from [CL91, Example 9] and used to obtain the asymptotics of the normalizing constant and the high-density limit.
  • standard math De Branges–Wilson and Barnes integral identities (B.1)–(B.4) hold with the required complex parameters.
    These are classical beta and Barnes integrals, cited to [dB72, Wil80, KLS10, DLMF25] and used in the induction proof of Theorem 4.8 and in Theorem 1.3.
  • domain assumption Exponential functional asymptotics for centered random walks [Hir97, Xu23], used in the boundary cases u=0, v>0 and u>0, v=0 of Theorem 1.2.
    The paper states the needed results in Appendix C and uses them to prove Lemma 7.4.
  • standard math Convergence of Laplace transforms on an open set implies weak convergence of probability measures.
    Invoked via [MRS06, Theorem 2] and [BW19, Theorem A.1] to convert Laplace-transform limits into weak convergence of stationary measures.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Stationary measures for log-gamma polymer on a strip and in half-space." pith.science (2026). https://pith.science/paper/TUU4MKOD

@misc{pith2026260717030,
  author       = {Pith},
  title        = {Pith review of: Stationary measures for log-gamma polymer on a strip and in half-space},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TUU4MKOD}},
  note         = {Machine review of arXiv:2607.17030}
}
abstract

We study stationary measures of the log-gamma polymer on a finite diagonal strip and in a half-space. We establish the phase diagram for the stationary measure on the strip. To this end we develop a representation of {the Laplace transform of} this stationary measure by independent $\mathrm{Beta}_{II}$ random variables. We also present an analytic approach that extends Barraquand's contour integral representation of the Laplace transform. We prove that, as the strip width tends to infinity, the stationary measure of the log-gamma polymer on the strip converges to the stationary measure of the half-space log-gamma polymer. Finally, we derive a contour integral formula for the Laplace transform of the stationary measure of the half-space log-gamma polymer.

Figures

Figures reproduced from arXiv: 2607.17030 by the authors.

Figure 1
Figure 1. The subset SN of the Z 2 lattice for N = 6. According to [BCY24, Theorem 1.6], the log-gamma polymer on a strip has a unique stationary measure, which is the law of the first vector component, which is (1.4) below, of [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Phase diagram for the stationary measure of the log-gamma polymer. The limit 1 N LN → ρ = ρ(α, u, v) depends on the boundary param￾eters u, v > −α. The line between the points (−α, −α) and (0, 0) separates the regions with the low ρ = −ψ(α − u) and high ρ = −ψ(α + v) density, and for the parameters (u, v) on that line, the limit of 1 N LN is random. As in Definition 1.1, a stationary measure is the law of a random s… view at source ↗
Figure 3
Figure 3. Contour C divides C into two open sets C + C and C − C . Definition 5.1. We call a contour C (of the type defined above) a good contour if it satisfies the following properties: • if z ∈ C + C then the whole half-line {z + t : t ∈ [0, ∞)} ⊂ C + C ; • if z ∈ C − C then the whole half-line {z + t : t ∈ (−∞, 0]} ⊂ C − C . For example, the contour C in [PITH_FULL_IMAGE:figures/full_fig_p031_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: A good contour that separates s = (s1, s2) and t = (t1, t2), with the locations of the poles of H(s, t; z) marked. With s1 − s2 ̸∈ Z and t1 − t2 ̸∈ Z, all poles are simple poles. When the curve C is replaced back by iR, the poles marked in blue contribute positive resi…
Figure 5
Figure 5. Figure 5: In the proof of Proposition 5.1, we can take A = max{−c, C}. □ This proposition leads to the following result [PITH_FULL_IMAGE:figures/full_fig_p033_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 1 linked inside Pith

  1. [1]

    Generalized hypergeometric series

    Wilfrid Norman Bailey. Generalized hypergeometric series . Camrbridge University Press, 1935

    1935

  2. [2]

    Integral formulas for two-layer S chur and W hittaker processes

    Guillaume Barraquand. Integral formulas for two-layer S chur and W hittaker processes. Bulletin de la Société Mathématique de France , 2024. arXiv preprint https://arxiv.org/abs/2409.08927

    arXiv 2024

  3. [3]

    Stationary measures for the log-gamma polymer and KPZ equation in half-space

    Guillaume Barraquand and Ivan Corwin. Stationary measures for the log-gamma polymer and KPZ equation in half-space. Ann. Probab. , 51(5):1830--1869, 2023

    2023

  4. [4]

    Stationary measures for integrable polymers on a strip

    Guillaume Barraquand, Ivan Corwin, and Zongrui Yang. Stationary measures for integrable polymers on a strip. Invent. Math. , 237(3):1567--1641, 2024

    2024

  5. [5]

    Stochastic Models with Power-Law Tails: The Equation X=AX+B

    Dariusz Buraczewski, Ewa Damek, and Thomas Mikosch. Stochastic Models with Power-Law Tails: The Equation X=AX+B . Springer, 2016

    2016

  6. [6]

    Limit fluctuations for density of asymmetric simple exclusion processes with open boundaries

    W odzimierz Bryc and Yizao Wang. Limit fluctuations for density of asymmetric simple exclusion processes with open boundaries. Ann. Inst. Henri Poincar\'e Probab. Stat. , 55(4):2169--2194, 2019

    2019

  7. [7]

    Explicit stationary distributions for compositions of random functions and products of random matrices

    Jean-Fran c ois Chamayou and G \'e rard Letac. Explicit stationary distributions for compositions of random functions and products of random matrices. Journal of Theoretical Probability , 4(1):3--36, 1991

    1991

  8. [8]

    John B. Conway. Functions of one complex variable , volume 11 of Graduate Texts in Mathematics . Springer-Verlag, New York-Berlin, second edition, 1978

    1978

Show all 28 references
  1. [9]

    Gauss spaces of entire functions

    Louis de Branges. Gauss spaces of entire functions. Journal of Mathematical Analysis and Applications , 37(1):1--41, 1972

    1972

  2. [10]

    Evans, Vincent Hakim, and Vincent Pasquier

    Bernard Derrida, Martin R. Evans, Vincent Hakim, and Vincent Pasquier. Exact solution of a 1 D asymmetric exclusion model using a matrix formulation. J. Phys. A , 26(7):1493--1517, 1993

    1993

  3. [11]

    https://dlmf.nist.gov/, Release 1.2.4 of 2025-03-15, 2025

    NIST Digital Library of Mathematical Functions . https://dlmf.nist.gov/, Release 1.2.4 of 2025-03-15, 2025. F. W. J. Olver, A. B. Olde Daalhuis , D. W. Lozier, B. I. Schneider, R. F. Boisvert, C. W. Clark, B. R. Miller, B. V. Saunders, H. S. Cohl, and M. A. McClain, eds

    2025

  4. [12]

    Perpetuities with thin tails

    Charles M Goldie and Rudolf Gr \"u bel. Perpetuities with thin tails. Advances in Applied Probability , 28(2):463--480, 1996

    1996

  5. [13]

    On the K ontorovich- L ebedev transform

    Dorian Goldfeld, Alex Kontorovich, and Eric Stade. On the K ontorovich- L ebedev transform. Comment. Math. Univ. St. Pauli , 60(1-2):37--46, 2011

    2011

  6. [14]

    Goldie and Ross A

    Charles M. Goldie and Ross A. Maller. Stability of perpetuities. Ann. Probab. , 28(3):1195--1218, 2000

    2000

  7. [15]

    A. K. Grincevi c ius. On the continuity of the distribution of a sum of dependent variables connected with independent walks. Theory of Probability and Its Applications , 19(1):163--168, 1975

    1975

  8. [16]

    Pinning and roughening of domain walls in I sing systems due to random impurities

    David A Huse and Christopher L Henley. Pinning and roughening of domain walls in I sing systems due to random impurities. Phys. Rev. Lett. , 54(25):2708, 1985

    1985

  9. [17]

    An asymptotic behavior of the mean of some exponential functionals of a random walk

    Katsuhiro Hirano. An asymptotic behavior of the mean of some exponential functionals of a random walk. Osaka J. Math. , 34(4):953--968, 1997

    1997

  10. [18]

    J. P. Imhof. Oscillations of continuous symmetric random walk. Ann. Probability , 4(4):662--666, 1976

    1976

  11. [19]

    Random difference equations and renewal theory for products of random matrices

    Harry Kesten. Random difference equations and renewal theory for products of random matrices. Acta Math. , 131:207--248, 1973

    1973

  12. [20]

    Hypergeometric orthogonal polynomials

    Roelof Koekoek, Peter A Lesky, and Ren \'e F Swarttouw. Hypergeometric orthogonal polynomials . Springer, 2010

    2010

  13. [21]

    A local limit theorem on the semi-direct product of R ^ *+ and R ^d

    \'Emile Le Page and Marc Peign\'e. A local limit theorem on the semi-direct product of R ^ *+ and R ^d . Ann. Inst. H. Poincar\'e Probab. Statist. , 33(2):223--252, 1997

    1997

  14. [22]

    Mukherjea, M

    A. Mukherjea, M. Rao, and S. Suen. A note on moment generating functions. Statist. Probab. Lett. , 76(11):1185--1189, 2006

    2006

  15. [23]

    Geometric RSK correspondence, W hittaker functions and symmetrized random polymers

    Neil O'Connell, Timo Sepp\"al\"ainen, and Nikos Zygouras. Geometric RSK correspondence, W hittaker functions and symmetrized random polymers. Invent. Math. , 197(2):361--416, 2014

    2014

  16. [24]

    Scaling for a one-dimensional directed polymer with boundary conditions

    Timo Sepp\"al\"ainen. Scaling for a one-dimensional directed polymer with boundary conditions. Ann. Probab. , 40(1):19--73, 2012

    2012

  17. [25]

    Methods of the theory of functions of many complex variables

    Vasili i Sergeevi c Vladimirov. Methods of the theory of functions of many complex variables . The M.I.T. Press, Cambridge, Mass.-London, 1966. Translated from the Russian by Scripta Technica, Inc

    1966

  18. [26]

    Some hypergeometric orthogonal polynomials

    James A Wilson. Some hypergeometric orthogonal polynomials. SIAM Journal on Mathematical Analysis , 11(4):690--701, 1980

    1980

  19. [27]

    Askey- W ilson signed measures and open ASEP in the shock region

    Yizao Wang, Jacek Weso owski, and Zongrui Yang. Askey- W ilson signed measures and open ASEP in the shock region. Int. Math. Res. Not. IMRN , (15):11104--11134, 2024

    2024

  20. [28]

    Asymptotics for exponential functionals of random walks

    Wei Xu. Asymptotics for exponential functionals of random walks. Stochastic Process. Appl. , 165:1--42, 2023

    2023

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.