{"id":"0697f769-e7ab-46d3-b8f6-f345a1c82b7f","arxiv_id":"1908.06591","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The stationary O'Connell-Yor polymer increments converge to energy solutions of the stochastic Burgers equation without using the Cole-Hopf transform.","lead":"This paper proves that the fluctuating increments of a semi-discrete directed polymer model converge, under a specific scaling, to the solutions of the stochastic Burgers equation. The proof avoids the usual Cole-Hopf transform and introduces a second-order Boltzmann-Gibbs principle, offering a new route to universal KPZ-class scaling limits.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The transport constant c in Theorem 1 is never derived: Section 7.2 only says linear terms converge to transport terms, and Remark 1 states c = -9/10 without computation, so the claimed limit equation is not identified.","rationale":"The paper establishes a substantial scaling limit by a plausible energy-solution route, and the second-order Boltzmann-Gibbs principle is a genuine technical contribution. The reader's conditional verdict is appropriate. The most load-bearing gap is the transport constant. The theorem promises an explicit c, but the proof only asserts convergence to 'transport terms' and never performs the coefficient bookkeeping. Remark 1 is an in-text admission of this missing support. This matters because the limiting object in Theorem 1 is defined through the martingale problem in Definition 3, whose drift term depends on c; without c the limit equation is not identified, and the imported uniqueness theorem is applied to an incompletely specified object. The concern is not that c is controversial or that the external uniqueness result is unreliable; it is that a central output of the proof is absent. A conditional acceptance requiring the missing derivation is therefore the right posture, and the reader's verdict need not be changed.","tokens_in":22004,"tokens_out":20720,"duration_ms":197794,"concrete_test":"Re-derive the full coefficient of the transport term by collecting, with explicit signs, every linear contribution in Sections 6.3 through 7.2 and Appendix A: from β²u_j and (1/√n)W_j in the expansion of W_j - u_j, from the linear part of W_{j-1}W_j, from the β²u_j terms produced by the L-operators in the appendix, and from the order-three monomial reductions in equation (8). Sum these contributions to verify that the coefficient equals -9/10; if the sum differs, Theorem 1's limit equation is misidentified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The theorem's central claim is convergence to the unique energy solution of ∂_t u = 1/2 ∂_x² u + c ∂_x u - 1/2 ∂_x u² + ∂_x W with an explicit constant c. The proof never computes this constant. In Section 6.3 the anti-symmetric term is expanded into contributions including -1/2 W_{j-1}W_j, -1/3 W_{j-1}u_j², L-terms, -β²/2, -1/6 β²u_j, and an order-three remainder; Section 7.2 states only that 'all linear terms ... converge to transport terms' and the appendix says the cubic terms contribute 'a few transport terms.' No coefficient is assembled. Remark 1 explicitly acknowledges this: 'The precise value of c can be obtained by careful bookkeeping along the proof. We found it to be -9/10.' Because the drift coefficient is part of the energy-solution martingale problem in Definition 3, and because the uniqueness theorem imported from [23] concerns that same object, the phrase 'the unique energy solution' in Theorem 1 is not supported by the proof as written. If the unperformed bookkeeping yields any value other than -9/10, the stated limit equation is wrong, even though the qualitative energy-solution universality might survive after a Galilean shift. This is not a disagreement with external consensus; it is a missing computation internal to the proof of the stated result.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the stationary O'Connell-Yor semi-discrete directed polymer model in a Brownian environment under intermediate-disorder scaling β = n^{-1/4}, θ = 1 + 1/(2√n). The main result, Theorem 1, states that the fluctuation field of the centered increments of the log-partition function converges in distribution in C([0,T], S'(R)) to the unique stationary energy solution of the stochastic Burgers equation ∂_t u = 1/2 ∂_x^2 u + c ∂_x u - 1/2 ∂_x u^2 + ∂_x W, where c is an explicit constant. The proof is based on the martingale decomposition of the system of SDEs satisfied by the increments, a second-order Boltzmann-Gibbs principle replacing local nonlinearities by block averages, and the energy-solution framework of Gonçalves-Jara and Gubinelli-Perkovski. The approach does not use the Cole-Hopf transform and does not rely on spectral gap estimates. The paper contains a detailed exposition of the static estimates, dynamical estimates, tightness, and identification of the limiting quadratic term via the energy solution formalism.","tokens_in":22335,"tokens_out":9179,"duration_ms":88236,"significance":"If the proof can be completed, the result is a significant contribution: it establishes a new instance of KPZ universality for the stationary O'Connell-Yor polymer using the energy-solution route, independent of the Cole-Hopf transform. The second-order Boltzmann-Gibbs principle for this non-polynomial model is a substantial technical advance, and the paper carefully separates the symmetric and antisymmetric parts of the generator. The manuscript also benefits from transparent static computations for the stationary measure and from the use of the existing uniqueness theorem for energy solutions instead of an ad hoc limiting argument. I see no circularity: the target equation is not assumed, and the estimates are independent of the value of c. However, the central identification of the limit equation is incomplete because the transport constant c is never derived, and several load-bearing L2 computations are only asserted. These gaps must be repaired before the result is fully supported.","major_comments":[{"comment":"The limit equation is not identified because the transport constant c is never computed. Remark 1 states that c = -9/10 \"can be obtained by careful bookkeeping along the proof,\" but no coefficient is assembled in Sections 6 or 7. Section 7.2 only states that the linear terms converge to transport terms, and the displayed comparison there involves β² u_j and (1/√n) W_j, not the full list of linear and constant terms produced in the expansion of Section 6.3, which includes -β²/2 and -(1/6)β² u_j. Since Definition 3 includes c in the martingale problem, and the imported uniqueness theorem of [23] applies to that fixed equation, the conclusion \"the unique energy solution of (4)\" is not supported as written. The missing bookkeeping is load-bearing: if the assembled coefficient differs from -9/10, the stated limit equation is wrong, even if the qualitative energy-solution universality survives after a Galilean shift.","section":"Section 1.3, Theorem 1 and Remark 1; Section 7.2"},{"comment":"Several error bounds that are essential to the proof are asserted rather than derived. In Proposition 1, the bound for the term involving β²/l times the block average is justified by a \"careful L2 computation\" whose dependency-blocking step is only sketched. In Appendix A, the bounds on the error terms E_j^{(i)}, the replacement of u_j^2 by W_j^2, the index-shift estimate for W_j^3 - W_{j-1}^3, and the removal of the centering in the order-three monomials are each delegated to \"simple L2 computation\" or \"straightforward adaptation\" without displaying the calculations. These estimates are precisely what forces the cubic terms to vanish or to contribute only transport terms, so the identification of the anti-symmetric part in Section 7.5 depends on them. The authors should display these computations or provide enough detail for a reader to verify them independently.","section":"Section 5.2, Proposition 1; Appendix A, eq. (8)"},{"comment":"The passage from the convergence of the fluctuation field to the convergence of the quadratic block averages involving the indicator ι_ε is delegated to a reference to [17], Section 5.3. Since ι_ε is not a Schwartz function and this step is needed to identify the limiting quadratic action A^ε_{s,t}, the adaptation should be summarized rather than only cited, especially because the paper otherwise takes care to prove or state each estimate used in the identification.","section":"Section 7.5"}],"minor_comments":[{"comment":"In the displayed estimate for the linear term, the notation ∇nϕ n_t is inconsistent with the surrounding equations, where the test function is denoted ϕ n_j; this should be corrected.","section":"Section 7.2"},{"comment":"The sentence \"The term β² u_j is easily seen to be tight\" is imprecise: the term is bounded in L2 uniformly in n, and later in Section 7.2 it converges to a transport term rather than vanishing. The wording should be adjusted to reflect the actual role of the term.","section":"Section 6.3"},{"comment":"The definition of an energy solution is stated for a fixed constant c, and the proof relies on the uniqueness theorem of [23]. The paper should make explicit that the uniqueness theorem applies for every real c, or cite the precise statement, since c is not used in any of the estimates before Section 7.","section":"Section 2, Definition 3"},{"comment":"In the paragraph after eq. (8), the removal of the centering uses the tightness of order-two terms; the sentence \"as we know that the terms of order two are tight\" should give a specific reference to the estimates in Section 6.3 rather than referring to the preceding discussion in general terms.","section":"Appendix A"},{"comment":"There are minor typographical issues: \"an huge body of work\" should be \"a huge body of work,\" and the notation introducing h_{β,θ} = log Z_{β,θ} and u_{β,θ} is missing a comma after Z_{β,θ}.","section":"Section 1.1 and Section 1.3"}],"recommendation":"major_revision","confidential_remarks":"The missing derivation of the transport constant c is the main obstacle to accepting the paper. I would not reject the manuscript on the basis of its reliance on [23] or on the asserted L2 computations, but these points must be made checkable. The authors should be given the opportunity to add the explicit bookkeeping for c and to expand the omitted L2 estimates in Proposition 1 and Appendix A. If those additions are made, the paper would be a strong contribution to the energy-solution approach to KPZ universality."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper proves a genuinely new convergence statement: the stationary O'Connell–Yor polymer increments, under intermediate disorder scaling, converge to energy solutions of the stochastic Burgers equation. The technical core is a second-order Boltzmann–Gibbs principle adapted to the non-polynomial potential V(u)=e^{-u}-1+u. It avoids both the Cole–Hopf transform and spectral gap estimates, and it fits cleanly into the established energy-solutions framework. That is a meaningful advance for the KPZ universality program, and the authors deserve credit for carrying out this argument.\n\nThe main soft spot is exactly what the stress-test flags: the transport constant c is never computed in the proof. In Section 7.2 the linear terms are said to 'converge to transport terms,' and Remark 1 states that careful bookkeeping gives c = -9/10, but the coefficient is never assembled. Since Theorem 1 claims convergence to the unique energy solution for that specific equation, the statement as written is not fully supported. This is not a fatal flaw — a Galilean shift would absorb any value of c, and the qualitative universality likely survives — but it is a real gap in the proof of the stated result. A referee should ask for the bookkeeping to be done explicitly.\n\nThere is a second, smaller issue: several central estimates are relegated to 'careful L2 computations' without details. The T/l^2 term in Proposition 1 and the order-3 controls in Appendix A are the most notable. The displayed arguments around them are plausible, and the structure of the proof is sound, but a serious referee would want those computations expanded.\n\nThe citation pattern looks honest. The self-citations to [19] and [32] are to prior work in the same framework, and the overlap with the unpublished [38] is acknowledged. The circularity burden is low: the proof imports the energy-solution uniqueness theorem from [23] and the standard existence theory; nothing is fitted to a target result.\n\nWho is this for? People working on KPZ universality proofs, especially those using energy solutions or discrete polymer models. The paper deserves a serious referee, despite the incomplete bookkeeping. I would recommend accepting it for peer review with a request for revision: derive c explicitly and expand the omitted L2 estimates. If those check out, the result stands.","headline":"A substantial KPZ-universality result with a real but fixable gap: the transport constant in the stated limit is asserted, not derived.","tokens_in":22812,"tokens_out":1836,"would_cite":true,"duration_ms":20100,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60H15","60K35","82B44"],"pacs":[],"model":"deepseek-v4-flash","headline":"The stationary semi-discrete directed polymer in a Brownian environment converges, under intermediate disorder scaling, to the unique energy solution of the stochastic Burgers equation, with the proof carried by a second-order…","keywords":["KPZ equation","stochastic Burgers equation","directed polymers","energy solutions","Boltzmann-Gibbs principle","intermediate disorder","log-partition increments","fluctuation field"],"falsifier":"Go through Section 7 keeping every constant in the expansions of $W_j-u_j$ and the generator's antisymmetric part; the value of $c$ is the coefficient of $\\int_0^t X_s(\\partial_x\\phi)\\,ds$ that survives in the limit. If that coefficient is not $-9/10$, the equation in Theorem 1 is wrong as stated.","tokens_in":21834,"feed_emoji":"🌊","tokens_out":12019,"duration_ms":109876,"temperature":0.7,"pith_summary":"The paper establishes that, in the intermediate disorder regime, the stationary semi-discrete directed polymer in a Brownian environment has the increments of its log-partition function converging to the unique energy solution of the stochastic Burgers equation. This matters because it connects a polymer model to the Burgers/KPZ equation without using the Cole-Hopf transform and without spectral gap estimates. The proof's central estimate is a second-order Boltzmann-Gibbs principle, which controls the nonlinear term by replacing local products of the fluctuating field with block averages. The result is one more instance of KPZ universality obtained through the energy-solution route.","feed_headline":"Polymer increments converge to the Burgers energy solution","feed_subtitle":"A second-order Boltzmann-Gibbs principle proves the limit without Cole-Hopf or spectral gap estimates.","key_machinery":"The load-bearing object is the second-order Boltzmann-Gibbs principle: a quantitative replacement of local products of the centered field by block averages. Proposition 1 shows that sums of $\\sqrt n \\sum_j \\{W_{j-1}W_j - \\tau_j Q(l,\\cdot)\\}\\phi_j$ vanish in a suitable sense with error $O(l/\\sqrt n + T/l^2)$, and Proposition 2 applies the same idea to cubic products $W_{j-1}W_jW_{j+1}$. The quadratic replacement becomes the term $\\partial_x u^2$ in the limit; the cubic replacement shows higher-order terms vanish. Around this principle the proof organizes a one-block estimate, Kipnis-Varadhan dynamical estimates, Mitoma tightness, and the energy-solution martingale problem that identifies the limit.","core_discovery":"Under the scaling $\\beta = n^{-1/4}$ and $\\theta = 1 + 1/(2\\sqrt n)$, the fluctuation field $X^n_t(\\phi) = \\sum_j (u_j(tn)-\\rho_n)\\phi((j-nt-a_n)/\\sqrt n)$, where $u_j$ are the log-partition increments, converges in distribution in $C([0,T],\\mathcal S'(\\mathbb R))$ to the unique stationary energy solution of $\\partial_t u = \\frac12 \\partial_x^2 u + c\\,\\partial_x u - \\frac12 \\partial_x(u^2) + \\partial_x W$, with $c=-9/10$ and $W$ space-time white noise. The convergence is for the polymer analogue of the Burgers slope, not the height itself, and the limit is characterized by the energy-solution martingale problem: at each time the field is a white noise, the quadratic term is defined by a limit of block averages, and both forward and time-reversed martingale conditions hold. The proof does not use the Cole-Hopf transform and avoids spectral gap estimates; the nonlinear term emerges from the second-order Boltzmann-Gibbs principle.","pith_inferences":["The estimates in the proof are local in space, so a local-equilibrium version for non-stationary initial data is a plausible extension, although the paper explicitly restricts to the stationary setting.","Because $c=-9/10$ is stated without derivation, an independent computation of the drift from the generator's antisymmetric part would either confirm Theorem 1 or show that only the centering of test functions needs adjustment.","The block size $l \\sim \\sqrt{(t-s)n}$ used in the proof suggests the limiting antisymmetric term has a $3/2$ Hölder modulus in time; checking this exponent would be a sharper quantitative test of the convergence mechanism."],"forward_implications":["The stationary polymer's log-partition increments and the stochastic Burgers equation share the same equilibrium scaling limit, so the polymer inherits the SPDE's white-noise spatial structure and martingale characterization.","Convergence to Burgers for this model does not need a discrete Cole-Hopf transform, so the energy-solution route is available for polymer models lacking an explicit integrable transform.","The same proof transfers to systems of coupled diffusions with a potential that is quadratic at zero, as noted in Remark 3, whenever the dynamics is well defined.","The drift $c=-9/10$ is a centering artefact: a change of coordinates in the equation, or a more careful discrete centering, removes it and leaves the standard stochastic Burgers equation."],"supporting_citations":[{"why":"Introduces the energy-solution definition and the reconstruction theorem for the quadratic term that Section 2 uses.","marker":"[17]"},{"why":"Supplies the uniqueness theorem for energy solutions that Theorem 1 invokes.","marker":"[23]"},{"why":"Defines the semi-discrete directed polymer model and its stationary measure used throughout.","marker":"[40]"},{"why":"Establishes the intermediate disorder regime and the $\\beta=n^{-1/4}$ scaling.","marker":"[2]"},{"why":"Provides the Kipnis-Varadhan estimate and fixed-time convergence arguments adapted in Sections 5-7.","marker":"[14]"},{"why":"Develops the energy-solution theory for stochastic Burgers equations that the paper builds on.","marker":"[20]"},{"why":"Provides the tightness criterion for distribution-valued processes used in Section 6.","marker":"[37]"},{"why":"Gives the system of SDEs for the log-partition increments that is the proof's starting point.","marker":"[44]"},{"why":"Supplies the polynomial-product identities adapted in the appendix to control cubic terms.","marker":"[6]"}],"fun_headline_variants":["Polymer increments hit Burgers energy solution","New Gibbs principle yields polymer-Burgers convergence","Burgers energy solution from polymer increments","Polymer-Burgers limit without Cole-Hopf transform","Stationary polymer increments reach energy solution"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof relies on a previously established uniqueness theorem for stationary energy solutions of the stochastic Burgers equation; if that uniqueness failed, different subsequences of the polymer model could converge to different limits, and the phrase 'the unique energy solution' in Theorem 1 would be unjustified.","fun_headline_variants_meta":{"raw":{"variants":["Polymer increments hit Burgers energy solution","New Gibbs principle yields polymer-Burgers convergence","Burgers energy solution from polymer increments","Polymer-Burgers limit without Cole-Hopf transform","Stationary polymer increments reach energy solution"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000637,"raw_usage":{"total_tokens":2884,"prompt_tokens":842,"completion_tokens":2042,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":458,"completion_tokens_details":{"reasoning_tokens":1973}},"tokens_in":458,"tokens_out":2042,"duration_ms":15834,"temperature":1.0,"reasoning_tokens":1973,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:57:31.352178+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Go through Section 7 keeping every constant in the expansions of $W_j-u_j$ and the generator's antisymmetric part; the value of $c$ is the coefficient of $\\int_0^t X_s(\\partial_x\\phi)\\,ds$ that survives in the limit. If that coefficient is not $-9/10$, the equation in Theorem 1 is wrong as stated.","supporting_citations":[{"cited_title":"and Jara, M","cited_arxiv_id":null,"evidence_quote":"Introduces the energy-solution definition and the reconstruction theorem for the quadratic term that Section 2 uses."},{"cited_title":"and Perkowski, N","cited_arxiv_id":null,"evidence_quote":"Supplies the uniqueness theorem for energy solutions that Theorem 1 invokes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the semi-discrete directed polymer model and its stationary measure used throughout."},{"cited_title":"and Quastel, J","cited_arxiv_id":null,"evidence_quote":"Establishes the intermediate disorder regime and the $\\beta=n^{-1/4}$ scaling."},{"cited_title":"and Perkowski, N","cited_arxiv_id":null,"evidence_quote":"Provides the Kipnis-Varadhan estimate and fixed-time convergence arguments adapted in Sections 5-7."},{"cited_title":"and Jara, M","cited_arxiv_id":null,"evidence_quote":"Develops the energy-solution theory for stochastic Burgers equations that the paper builds on."},{"cited_title":"(1983) Tightness of probabilities in C([0,1], Y′) and D([0,1], Y′), Ann","cited_arxiv_id":null,"evidence_quote":"Provides the tightness criterion for distribution-valued processes used in Section 6."},{"cited_title":"(2014) KPZ scaling theory and the semidiscrete directed polymer mo del, Random Matri- ces, MSRI Publications, Vol","cited_arxiv_id":null,"evidence_quote":"Gives the system of SDEs for the log-partition increments that is the proof's starting point."},{"cited_title":"and Simon, M","cited_arxiv_id":null,"evidence_quote":"Supplies the polynomial-product identities adapted in the appendix to control cubic terms."}],"review_version":1}