Pith. sign in

REVIEW 4 major objections 4 minor 28 references

KPZ equation from open ASEP with general boundary asymmetry

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Open ASEP with arbitrary local boundary speeds converges to the open KPZ equation, removing Liggett's condition.

desk verdict Important if true, but a verifiable sign error breaks the central cancellation; the main theorem is not proven as written. read the letter →

arxiv 2507.11537 v2 pith:ELXXL6U7 submitted 2025-07-15 math.PR

Kevin Yang This is my paper · ORCID
classification math.PR MSC 60K3582C2260H15
keywords openKPZequationASEPboundaryspeedchangesGartnertransformCole-Hopfnon-invariantmeasureKipnis-Varadhanestimatestationary
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

This paper proves that the open ASEP particle system on a finite interval, with reservoir dynamics whose speeds are general local functions of the configuration, has a height function whose exponential (Gärtner transform) converges in law to the solution of the open stochastic heat equation; equivalently the height function itself converges to the open KPZ equation. The boundary parameters of the limiting SPDE are explicit expectations of the boundary speed functions under a product measure that is not invariant for the dynamics. This removes the longstanding Liggett condition (constant equal boundary creation and removal speeds) required in earlier derivations, and also removes the assumption of product invariant measures. The same result is proved for the half-space, and for the interval the stationary measure of the height increment process is shown to converge to that of the open KPZ increment process.

What carries the argument

The central object is the Gärtner transform $Z^N_{t,x} = \exp\{-\lambda h^N_{t,x} + (\tfrac{\lambda^2}{2}N + \tfrac{\lambda^4}{24})t\}$, the discrete analogue of the Cole-Hopf map that turns the height equation into a multiplicative stochastic heat equation. The paper derives its infinitesimal evolution: the bulk ASEP generator plus boundary contributions containing fluctuating terms $f_{\mathrm{left}}$ and $f_{\mathrm{right}}$ that are mean-zero with respect to the non-invariant product measure $P_0$, together with bounded remainder terms. The main work is homogenizing these boundary fluctuations without an invariant measure; the engine is a Kipnis-Varadhan-type estimate (Proposition 5.6) for time-averaged mean-zero local functions on a fattened boundary interval, built on entropy-production bounds (Lemma 5.2), log-Sobolev and one-block estimates (Lemma 5.4), and a semigroup comparison to the symmetric generator that preserves $P_0$ (Lemma 5.7).

What would settle it

Simulate the boundary-localized process on a fattened interval of length $N^\kappa$ for parameters outside Liggett's condition, and measure the second moment of the time-average of $f_{\mathrm{left}}$ over $\tau = N^{-2+\rho}$: Proposition 5.6 predicts decay $N^{-\rho} + N^{-1/3}$. Alternatively, for a fixed non-Liggett choice of $\alpha,\beta,\gamma,\delta$, compute the stationary height increment measure of open ASEP numerically on large intervals and compare it with the open KPZ stationary measure with parameters (2.8)-(2.9); a mismatch at leading order would refute Corollary 2.2.

Watch

Extended reading notes

Core claim

The central claim is that, under the stated moment and Lipschitz bounds on the initial data, the linearly interpolated Gärtner transform $Z^N_{t,NX}$ converges in law in $D([0,1],C([0,1]))$ to the solution of the open stochastic heat equation with boundary parameters $A$ and $B$ given by the explicit expectations displayed in (2.8)-(2.9); equivalently, the height function converges to the open KPZ equation. This holds for general local functions $\alpha,\beta,\gamma,\delta$ governing the reservoir speeds, without Liggett's condition $\alpha=\gamma$, $\delta=\beta$ and without any product invariant measure for open ASEP. The half-space version with only a left reservoir converges to the half-space open KPZ equation with the same parameter $A$. The paper further proves that the stationary measure of the height increment process converges weakly to the unique stationary measure of the open KPZ increment process when $\lambda=-1$.

Load-bearing premise

The proof depends on the assumption that near the boundary the dynamics mixes fast enough relative to the non-invariant product measure $P_0$: specifically, that the time-integrated discrepancy from $P_0$ is $O(N^{-1/2})$ and that time-averaged mean-zero local boundary functions decay at the stated rates; if either decay is slower, the boundary error terms survive and the limiting boundary parameters would change.

Editorial extensions

If this is right

  • Earlier open-ASEP-to-open-KPZ derivations required Liggett's condition (constant $\alpha=\gamma$ and $\delta=\beta$); this result removes it, answering the question posed in [3] and [14].
  • The limiting boundary parameters are computed from expectations under the non-invariant product measure $P_0$, so the open KPZ fixed point organizes a whole family of boundary dynamics, not only those with simple invariant measures.
  • The half-space open KPZ equation is also derived from half-space open ASEP with general local left-boundary speeds, without Liggett's condition.
  • The stationary measure of the height increment process converges weakly to the unique stationary measure of the open KPZ increment process (for $\lambda=-1$), so equilibrium fluctuations of the particle system match those of the continuum SPDE.
  • The proof provides a template for boundary homogenization when invariant measures are unavailable, extending beyond product-measure and constant-boundary settings.

Reading between the lines

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

  • If the theorem is right, the same homogenization mechanism should apply when the bulk asymmetry is of order $N$ rather than $N^{3/2}$, or when the boundary functions depend on longer but still mesoscopic windows, suggesting that the open KPZ universality class is stable under an open set of boundary-speed perturbations.
  • A testable extension is to compute finite-dimensional stationary correlations of the increment process and compare them with the known open KPZ stationary correlations for parameters outside Liggett's condition; Corollary 2.2 predicts they match at leading order.
  • The appearance of $P_0$, not the true invariant measure, in the boundary parameters suggests that for boundary-driven systems the continuum limit can be read off from the local equilibrium distribution even when global equilibrium is non-product; this principle may extend to other boundary-driven exclusion processes with speed changes.
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

4 major / 4 minor

Summary. The paper studies open ASEP on {1,...,N} and on the half-line, with reservoir rates whose N^{3/2}-order coefficients are arbitrary local functions α, β, γ, δ of the configuration near the boundary. The main result (Theorem 2.1) asserts that the Gärtner-transformed height function converges in law to the solution of the open stochastic heat equation on [0,1] with Robin boundary parameters A, B given by (2.8)-(2.9), equivalently that the height function converges to the open KPZ equation. The proof proceeds by deriving an evolution equation for Z^N (Section 3), proving tightness (Lemma 4.1), reducing the limit identification to vanishing of boundary error terms (Proposition 4.4), and estimating those terms via entropy/Dirichlet-form estimates and a Kipnis-Varadhan inequality with respect to the non-invariant product measure P0 (Sections 5-6). Corollary 2.2 deduces convergence of stationary measures of the increment process, and Theorem 2.3 gives the half-space analog.

Significance. If the proof can be repaired, this is a substantial step: it removes Liggett's condition for open ASEP, removes product-invariance assumptions for speed-change boundary dynamics, answers the cited open questions of Corwin and Himwich, and introduces a useful technical framework (Kipnis-Varadhan estimates with respect to a non-invariant measure on fattened boundary intervals). The paper is largely self-contained in its main estimates and identifies A and B parameter-free through expectations under P0. However, the right-boundary evolution equation and the displayed formulas for A and B contain sign/type inconsistencies that currently break the proof at load-bearing points; the central convergence claim is therefore not established as written.

major comments (4)
  1. [Section 3, Lemma 3.5, Eqs. (3.11)-(3.13)] The jump factors at the right boundary are inconsistent with the height definition (2.5). Under (2.5), a flip at site N from η_N = -1 to +1 increases h^N_{t,N} by 2N^{-1/2}, so Z^N_{t,N} should multiply by Exp(-2λN^{-1/2}) - 1; the lemma instead uses Exp(+2λN^{-1/2}) - 1, and the two factors in (3.12)-(3.13) are reversed. With the stated factors, the O(N^{3/2}) part of the flip drift is -λ/2 η_N Z^N_{t,N}, while the discrete Robin Laplacian (3.1) at x=N contributes +λ/2 η_N Z^N_{t,N}; the difference contains an O(N^{3/2}) term that cannot be absorbed into the O(N) term λN f_right Z^N_{t,N}. Lemma 3.5 is therefore false as written, and the right-boundary error R^N_{2,t} in Proposition 4.4 is not controlled.
  2. [Theorem 2.1, Eq. (2.8)] The last expectation in (2.8) is printed as E0{η1 (α[η] - γ[η])}, but f_left in Lemma 3.4 is 3λ/4 + α - γ - η1(α + γ) - A/2. With the printed formula one obtains E0 f_left = -2E0[η1 γ], which is generically nonzero because γ may depend on η1 near the boundary. The mean-zero condition required by Proposition 5.6 for applying it to R^N_{1,t} therefore fails. The sentence immediately after (2.8) states the intended function is 2(α - γ) - 2η1(α + γ), so the last term in (2.8) should read -2E0{η1 (α + γ)}; this must be corrected in the statement of the main theorem.
  3. [Eqs. (2.9) and Lemma 3.5] Even taking the manuscript's stated f_right = 3λ/4 + δ - β - η_N(δ + β) - B/2 and the value of B in (2.9), one gets E0 f_right = 3λ/2 - 2E0β, which is not zero in general. Hence R^N_{2,t} cannot be handled by Proposition 5.6 either. This is not an isolated typo: the right-boundary sign error in (3.12)-(3.13) changes the derivation of f_right, so (2.9) and the definition of f_right must be rederived consistently with the height convention (2.5).
  4. [Section 3, Lemma 3.4 proof] The stress-test concern about the left boundary O(N^{3/2}) cancellation does not actually land: in (3.6) the term (1/2)N^2 Δ_{A,B} Z^N_{t,0} appears on the right-hand side, and both this term and the flip drift in (3.10) contain -λ/2 N^{3/2} η_{t,1} Z^N_{t,0}; their difference is O(N), which is exactly what the definition of f_left requires. However, the prose in the proof of Lemma 3.4 says that a flip from -1 to +1 makes h^N_{t,0} go up by 2λN^{-1/2}, whereas under (2.5) it goes down; the displayed jump factor in (3.7) is consistent with (2.5), but the sentence is misleading and should be corrected.
minor comments (4)
  1. [Lemma 3.5, bullet after (3.11)] The function f_right is labeled f_left in the displayed definition; the name should be f_right.
  2. [Section 6, Eq. (6.7)] The exponent written as N^{-1/2 ρ} is ambiguous; if it means N^{-ρ/2}, please write it as such.
  3. [Lemma 5.4 proof] The notation P0 is used both for the initial density and for the reference measure; switching to a different symbol for one of these would improve clarity.
  4. [Section 4, Eqs. (4.5)-(4.7)] The symbols w_left/w_right and m_left/m_right are used inconsistently for the same object; please unify the notation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; boundary parameters are computed from the model and the key stochastic estimates are proved in the paper.

full rationale

The central derivation is self-contained. The boundary parameters A and B in (2.8)-(2.9) are not fitted to the limiting open SHE; they are explicit expectations of the microscopic boundary functions under the product measure P0, and the proof identifies them through the homogenization of f_left and f_right. The main technical estimates -- entropy production (Lemma 5.2), local-equilibrium bounds (Lemma 5.4), and the non-invariant Kipnis-Varadhan bound (Proposition 5.6) -- are proved in the paper using external tools such as Kipnis-Landim [16] and Yau [28]. The author's self-citation [27] is described only as the source of a technique, not as an unproved load-bearing theorem: the paper states it must adapt the method and then supplies the needed estimates itself. Citations for tightness and the martingale problem go to [5,24], bulk evolution to [6], and uniqueness of the invariant measure in Corollary 2.2 rests on [17,25], none of which are the present author's work. The apparent sign mismatch between (2.8) and f_left in Lemma 3.4 is a correctness concern rather than circularity: it would affect whether f_left is mean-zero under P0, but it does not mean the argument reduces to its own input. No step in the derivation chain is equivalent by construction to the conclusion.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The paper introduces no fitted parameters and no new physical entities. The boundary parameters A and B are derived from the microscopic functions via explicit expectations (2.8)-(2.9), not tuned. The proof uses a standard stack of hydrodynamic-limit tools: relative entropy, log-Sobolev, Kipnis-Varadhan, heat kernel estimates, and the martingale-problem uniqueness for the open SHE. The main extra ingredient is the author's prior technique [27], which is unpublished and therefore carries reproducibility risk. All listed axioms are either standard background or explicit hypotheses of the theorems.

assumptions (7)
  • standard math Uniqueness of the martingale problem for the open SHE (Proposition 4.3, quoted from [5]).
    Used in Section 4.2 to identify subsequential limits of Z^N as the unique law of (1.3)-(1.4).
  • standard math Heat kernel estimates for the discrete Robin heat kernel (Proposition A.1, quoted from Propositions 3.15, 3.16, 3.21 of [24]).
    Used throughout Lemmas 4.1 and 8.5 to bound Duhamel terms; also half-space version Proposition A.2.
  • standard math Log-Sobolev inequality for simple exclusion and spin-flip dynamics (from [28], [20]).
    Used in Lemma 5.4 to convert relative entropy to Dirichlet form control (Eqs. 5.14, 5.16).
  • standard math Kipnis-Varadhan inequality and spectral gap for the symmetric local process (from [16], Appendix 1.6).
    Core of Proposition 5.6, bounding time averages of mean-zero local functions by the Dirichlet form of the symmetrized generator L^{L_fat,symm}.
  • domain assumption The boundary functions alpha, beta, gamma, delta are uniformly bounded and depend only on O(1) sites near the boundary (Section 2.1).
    The theorem's scope is the general local boundary asymmetry; all estimates with L_fat, |L| bounded, and the discrepancy coupling depend on this locality.
  • domain assumption The initial data converges in the sense of (2.7) and its continuum limit exists (Theorem 2.1 hypothesis).
    Standard hypothesis inherited from [5,24]; the proof propagates these bounds to positive times via Gronwall (Lemma 4.1).
  • standard math Uniqueness of the stationary measure for the open KPZ increment process (from [17,25]).
    Used in Corollary 2.2 to identify all subsequential limits of the stationary increment measures as the unique invariant measure of the SPDE increment process.

how reviews work

0 comments
Cite this review

Pith. "Pith review of KPZ equation from open ASEP with general boundary asymmetry." pith.science (2026). https://pith.science/paper/ELXXL6U7

@misc{pith2026250711537,
  author       = {Pith},
  title        = {Pith review of: KPZ equation from open ASEP with general boundary asymmetry},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ELXXL6U7}},
  note         = {Machine review of arXiv:2507.11537}
}
read the original abstract

We consider generalizations of open ASEP in the interval and half-space, where the speed of the reservoir dynamics can depend on the local particle configuration. We show that their height functions have a continuum limit given by the open KPZ equation. This removes the assumption of Liggett's condition in Corwin-Shen '18 and Parekh '19, thus answering a question of Corwin '22 and Himwich '25, and it removes the assumption of product invariant measures in Goncalves-Perkowski-Simon '20. In the case of the interval, we also show convergence of the stationary measure for the height function increment process to that of the increment process for the open KPZ equation.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 27 canonical work pages

  1. [4]

    Stationary measure for the open KPZ equation

    I. Corwin, A. Knizel, “ Stationary measure for the open KPZ equation”,Communications on Pure and Applied Mathematics, 77, 4, 2183-2267, 2023. 28

    work page 2023

  2. [1]

    Stochastic Burgers and KPZ Equations from Particle Systems

    L. Bertini, G. Giacomin, “Stochastic Burgers and KPZ Equations from Particle Systems”,Communications in Mathematical Physics, 183, 3, 571-606, 1997

    work page 1997

  3. [2]

    Billingsley,Convergence of Probability Measures

    P. Billingsley,Convergence of Probability Measures. New York: John Wiley & Sons, Inc. ISBN 0-471-19745-9, 1968

    work page 1968

  4. [3]

    Some recent progress on the stationary measure for the open KPZ equation

    I. Corwin, “Some recent progress on the stationary measure for the open KPZ equation”,Toeplitz Operators and Random Matrices, 321-360, 2022

    work page 2022

  5. [5]

    Open ASEP in the weakly asymmetric regime

    I. Corwin, H. Shen, “Open ASEP in the weakly asymmetric regime”,Communications on Pure and Applied Mathematics, 71, 10, 2065-2128, 2018

    work page 2018

  6. [6]

    Weakly asymmetric non-simple exclusion process and the Kardar-Parisi-Zhang equation

    A. Dembo, L.-C. Tsai, “Weakly asymmetric non-simple exclusion process and the Kardar-Parisi-Zhang equation”,Communications in Mathematical Physics, 341, 219-261, 2016

    work page 2016

  7. [7]

    Exact solution of a 1D asymmetric exclusion model using a matrix formulation

    B. Derrida, M. R. Evans, V . Hakim, V . Pasquier, “Exact solution of a 1D asymmetric exclusion model using a matrix formulation”, Journal of Physics A: Mathematical and General, 26, 7, 1493

    work page

  8. [8]

    Hydrodynamic Limit of one-dimensional exclusion processes with speed change

    T. Funaki, K. Handa, K. Uchiyama, “Hydrodynamic Limit of one-dimensional exclusion processes with speed change”,Annals of Probability, 19, 1, 245-265, 1991

    work page 1991

Show all 28 references
  1. [9]

    Singular SPDEs in domains with boundaries

    M. Gerencser, M. Hairer, “Singular SPDEs in domains with boundaries”,Probability Theory and Related Fields, 173, 3, 697-758, 2019

    2019

  2. [10]

    Nonlinear fluctuations of weakly asymmetric interacting particle systems

    P. Goncalves, M. Jara, “Nonlinear fluctuations of weakly asymmetric interacting particle systems”,Archive for Rational Mechanics and Analysis, 212, 597-644, 2014

    2014

  3. [11]

    Non-equilibrium fluctuations of one-dimensional boundary driven weakly asymmetric exclu- sion processes

    P. Goncalves, C. Landim, A. Milanes, “Non-equilibrium fluctuations of one-dimensional boundary driven weakly asymmetric exclu- sion processes”.Annals of Applied Probability, 27, 1, 140-177, 2017

    2017

  4. [12]

    Derivation of the stochastic Burgers equation with Dirichlet boundary conditions from the W ASEP

    P. Goncalves, N. Perkowski, M. Simon, “Derivation of the stochastic Burgers equation with Dirichlet boundary conditions from the W ASEP”,Annales Henri Lebesgue, 3, 87-167, 2020

    2020

  5. [13]

    Nonlinear diffusion limit for a system with nearest neighbor interactions

    M.Z. Guo, G.C. Papnicolaou, S.R.S. Varadhan, “Nonlinear diffusion limit for a system with nearest neighbor interactions”.Commu- nications in Mathematical Physics, 118, 31-59, 1988

    1988

  6. [14]

    Convergence of the open W ASEP stationary measure without Liggett’s condition

    Z. Himwich, “Convergence of the open W ASEP stationary measure without Liggett’s condition”,Stochastic Processes and Applica- tions, 185, 104634, 2025

    2025

  7. [15]

    Dynamic scaling of growing interfaces

    M. Kardar, G. Parisi, Y .-C. Zhang, “Dynamic scaling of growing interfaces”,Physical Review Letters56, 9, 889, 1986

    1986

  8. [16]

    Kipnis, C

    C. Kipnis, C. Landim,Scaling Limits of Interacting Particle Systems, Springer-Verlig Berlin Heidelberg, 320, 1999

    1999

  9. [17]

    The strong Feller property of the open KPZ equation

    A. Knizel, K. Matetski, “The strong Feller property of the open KPZ equation”, arXiv:2211.04466, 2022

    2022 arXiv

  10. [18]

    Hydrodynamic limit for asymmetric mean zero exclusion processes with speed change

    K. Komoriya, “Hydrodynamic limit for asymmetric mean zero exclusion processes with speed change”.Annales de l’Institut Henri Poincare, 34, 6, 767-797, 1998

    1998

  11. [19]

    Equilibrium fluctuations for exclusion processes with speed change

    C. Landim, M.E. Vares, “Equilibrium fluctuations for exclusion processes with speed change”.Stochastic Processes and Applications, 52, 107-118, 1994

    1994

  12. [20]

    Concentration of measure and logarithmic Sobolev inequalities

    M. Ledoux, “Concentration of measure and logarithmic Sobolev inequalities”,S ´eminaire de Probabilites XXXIII, 120-216, 2006

    2006

  13. [21]

    Ergodic theorems for the asymmetric simple exclusion process

    T. Liggett, “Ergodic theorems for the asymmetric simple exclusion process”,Transactions of the American Mathematical Society, 213, 237-261

  14. [22]

    Ergodic theorems for the asymmetric simple exclusion process II

    T. Liggett, “Ergodic theorems for the asymmetric simple exclusion process II”,Annals of Probability, 5, 5, 795-801, 1977

    1977

  15. [23]

    Bulk diffusion in a system with site disorder

    J. Quastel, “Bulk diffusion in a system with site disorder”,Annals of Probability, 34, 5, 1990-2036, 2006

    1990

  16. [24]

    The KPZ Limit of ASEP with Boundary

    S. Parekh, “The KPZ Limit of ASEP with Boundary”,Communications in Mathematical Physics, 365, 569-649, 2019

    2019

  17. [25]

    Ergodicity results for the open KPZ equation

    S. Parekh, “Ergodicity results for the open KPZ equation”, arXiv:2212.08248, 2022

    2022 arXiv

  18. [26]

    Kardar-Parisi-Zhang Equation from non-simple variations on open ASEP

    K. Yang, “Kardar-Parisi-Zhang Equation from non-simple variations on open ASEP”.Probability Theory and Related Fields. 183, 415-545, 2022

    2022

  19. [27]

    KPZ equation from ASEP plus general speed-change drift

    K. Yang, “KPZ equation from ASEP plus general speed-change drift”, arXiv:2409.10513, 2024

    2024 arXiv

  20. [28]

    Logarithmic Sobolev inequality for generalized simple exclusion processes

    H.-T. Yau, “Logarithmic Sobolev inequality for generalized simple exclusion processes”,Probability Theory and Related Fields, 109, 507-538, 1997. 29

    1997

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.