{"id":"89fa2d3e-01e7-44be-928b-7a81d7b64139","arxiv_id":"2608.02772","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A Darboux-dressing construction on Fredholm determinants yields closed matrix Hirota-Miwa equations for eighteen KPZ models across four scaling regimes.","lead":"This paper introduces a unified algebraic framework that produces explicit closed equations for the joint distribution functions of eighteen exactly solvable models in the KPZ universality class. A smart generalist should read it because it connects random growth and particle systems to the classical Hirota-Miwa integrability hierarchy, suggesting that the field's many exact formulas share one underlying structure.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The eighteen-model theorem rests on per-model dressing-compatibility and resolvent checks that are not shown in the available text and are explicitly deferred for the semi-discrete product graph (Remark 3.13); a single failed verification would break the corresponding closed equation.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing concern: the framework's central claim depends on per-model verification of dressing compatibility and analytic conditions that are not supplied in the text under review and are explicitly deferred in the semi-discrete setting. I agree with the CONDITIONAL verdict. The core Darboux mechanism is presented carefully and has independent plausibility: the compatibility calculation is direct, the resolvent algebra in Theorem 2.5 is explicit, and the scalar reduction to Hirota-Miwa form is derived with stated boundary conditions. Credit should also be given for honest acknowledgment of limitations, including the deferred semi-discrete construction and the non-invertibility of Euclidean-division C-weights. However, none of these virtues establishes Theorem 1.5 for all eighteen models, because the theorem's hypotheses are model-specific analytic facts. A single failure among the eighteen would invalidate one of the catalogued closed equations. The proposed concrete test addresses the most exposed case, continuous-time TASEP, where the missing semi-discrete product graph construction is essential. The overstatement about the Hirota-Miwa connection noted by the reader is a presentation issue rather than a load-bearing mathematical gap, so it does not change the verdict.","tokens_in":73138,"tokens_out":3937,"duration_ms":47659,"concrete_test":"Complete the semi-discrete product graph construction for continuous-time TASEP (Chapter 8.1): define the admissible propagator B_u satisfying the semi-discrete compatibility condition stated in Remark 3.13, and prove uniform convergence of the double sum defining K(u) in (2.45) together with existence of (I-zK(u))^{-1} on the relevant ℓ^2 space at every shift needed for the mixed diamond equation. If the propagator conditions cannot be satisfied or the resolvent estimates fail, then Theorem 1.5's Chapter 8 claims are unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Darboux theorem and scalar reductions are internally coherent, but Theorem 1.5 transfers these results to eighteen concrete models by asserting that each model's Fredholm kernel satisfies dressing compatibility, e.g. conditions (2.7), plus resolvent existence and convergence of the kernel sums. These are load-bearing hypotheses. The text provided here truncates in Section 6.2, so no complete per-model verification is visible for review. More structurally, the multipoint equations for the continuous-time models in Chapter 8 (continuous-time TASEP, Push-TASEP, ASEP) require the semi-discrete product graph construction, and Remark 3.13 explicitly defers that construction to future work. Without it, the closed matrix equations for those three models have not been derived. Example 2.4 and Definition 3.4 introduce dressing-compatible kernels only under 'suitable analytic assumptions'; if any of the eighteen kernels fails condition (2.7), or if the resolvent (I-zK)^{-1} fails to exist on the shift set required by the mixed diamond proof, the corresponding advertised equation is not established. This is not an internal inconsistency, but it is a gap between the framework and the claimed catalogue.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper introduces an algebraic framework, the 'diamond framework', for deriving closed nonlinear equations for distribution functions of exactly solvable KPZ models. An overdetermined linear problem on a lattice (2.1) with edge weights (Ci, Λi) is compatible exactly when the diamond equations (2.2)–(2.4) hold (Proposition 2.1); analogous compatibility systems are developed in three further regimes: semi-discrete (Chapter 3), parabolic (Chapter 4), and continuum (Chapter 5). A Darboux transformation (Theorems 2.5, 3.6, 4.6, 5.5) dresses Fredholm-kernel data into a finite-dimensional matrix observable M (or additive observable A) that satisfies the same diamond equations, and a scalar reduction yields variable-coefficient Hirota–Miwa equations for the Fredholm determinant (Propositions 2.11, 3.11, and the parabolic/continuum trace-defect identities). Seed data is produced by a product graph construction with admissible propagators (Theorem 2.26) and by Euclidean division in a polynomial quotient algebra. The paper's central result, Theorem 1.5, asserts that for each of eighteen listed models (§1.7.1) the Fredholm determinant data satisfies the hypotheses of the appropriate Darboux theorem, yielding an explicit closed matrix equation per model, and that the diamond equations are gauge-equivalent to the non-abelian Hirota–Miwa system (Proposition 2.15).","tokens_in":73256,"tokens_out":31187,"duration_ms":239382,"significance":"The framework, if its per-model hypotheses all hold, would be a significant unification: a single mechanism producing closed equations for finite-time multipoint distributions across four scaling regimes, recovering the Quastel–Remenik matrix KP equation and the author's earlier one-point bilinear equations, and connecting the exactly solvable KPZ sector to the non-abelian Hirota–Miwa system. The manuscript's strengths are substantial: the core theorems are proved in detail and appear internally consistent; the Darboux theory needs only resolvent existence rather than trace-class structure; the scalar reductions fix their T-orbit constants through boundary conditions rather than parameter fitting, so the derivations are parameter-free in the relevant sense; and the catalogue of eighteen explicit equations is concrete, falsifiable content, since a single kernel failing the linear dressing conditions (2.7) would break its entry. The product graph construction is original, and the gauge equivalence with Nimmo's system is cleanly formulated.","major_comments":[{"comment":"The semi-discrete product graph construction is explicitly deferred. Remark 3.13 states that the direct verifications are 'deferred to a future work', giving only the statements of the modified ∂1-compatibility condition and the corrected continuous Λ-weight without proofs. The catalogue entries 13 and 14—continuous-time TASEP (Theorem 8.6) and Push-TASEP (Theorem 8.14)—are m×m matrix multipoint equations and require exactly this construction to build the dressing-compatible kernel on the product graph; the one-point ASEP entry (Corollary 8.22) is not affected. Consequently, Theorem 1.5's claim that all Chapter 8 models satisfy the hypotheses of the semi-discrete Darboux theorem is not established for these two models in the submitted text. The revision should either provide the product-graph analogues of Lemmas 2.23–2.25 for the semi-discrete regime, or restate Theorem 1.5 and the eighteen-model claim with the affected entries explicitly qualified.","section":"Remark 3.13; Theorem 1.5, Chapter 8"},{"comment":"Theorem 1.5 asserts for each of the eighteen models that the Fredholm data satisfies the dressing-compatibility conditions ((2.7), (3.9)–(3.11), (4.14)–(4.16), (5.7)–(5.9)) together with resolvent existence and convergence of the kernel sums. These are model-specific analytic checks, not consequences of the abstract framework: the Darboux theorems guarantee only that dressing-compatible data yields a closed equation, and Remark 2.9 shows the resolvent must exist on the particular shift sets used in the mixed diamond proof. The text provided for review contains Chapters 1–5 and the opening of Chapter 6 (through §6.2), and the verification is cut off before the kernel-sum convergence and resolvent checks for the directed-path propagator are carried out; Section 6.1 itself flags this regularity as a model-by-model issue (Remark 6.3, Definition 2.21(iv)), and Example 2.4 and Definition 3.4 introduce dressing-compatible kernels only 'under suitable analytic assumptions'. Since these checks are the load-bearing content of the paper's central claim, the manuscript should display them in full—for instance as a per-model checklist following the five-step procedure of §1.7, with the kernel estimates needed for convergence and resolvent existence. Without this, Theorem 1.5 is an extrapolation of the framework rather than an established result.","section":"Theorem 1.5; Definition 2.21(iv); Example 2.4; §6.1–6.2"},{"comment":"The parabolic Darboux theorem (Theorem 4.6), which governs the two Chapter 9 models, depends on the factorization identity (4.31) in Lemma 4.8. The displayed verification is compressed: after expanding the remainder R in (4.32) and the four blocks BA, BS, BM, BC, the proof asserts that seven cross-block pairs cancel and that the remaining terms 'match (4.32) term by term'—a check involving several dozen terms that is not actually shown. An algebraic error at this point would invalidate the dressed mixed diamond equation for the parabolic regime. The revision should display the complete cancellation table (or an independent symbolic verification) for (4.31).","section":"Lemma 4.8 (identity (4.31))"}],"minor_comments":[{"comment":"The nondegeneracy hypothesis of Proposition 2.11 (and Corollary 1.3) should explicitly require αji(u) ≠ 0 in addition to αij(u) ≠ 0 and αij(u) ≠ αji(u), including the T-limit conditions: the coefficient reduction in the proof divides by cj(Tu) and by λi(u), and the current wording guarantees neither division.","section":"Proposition 2.11 / Corollary 1.3"},{"comment":"The abstract and Theorem 1.5 describe 'closed matrix equations' for eighteen models, but six catalogue entries (§1.7.1 items 8–12 and 15) are scalar one-point equations rather than matrix multipoint equations; the wording should distinguish the matrix multipoint equations from their scalar one-point specializations.","section":"Abstract and §1.7.1"},{"comment":"The analytic hypotheses are introduced repeatedly as 'suitable analytic assumptions' (Example 2.4, Definition 3.4) or bundled into the regularity condition (Definition 2.21(iv)). Because Theorem 1.5 rests on these per-model conditions, a single consolidated statement of the analytic framework at the start of Section 2—specifying the Hilbert space, the trace-class/summability hypotheses, and the allowed formal manipulations—would make the per-model chapters a fixed checklist rather than a sequence of ad hoc verifications.","section":"Definition 2.21(iv); Example 2.4; Definition 3.4"},{"comment":"In Remark 6.4, the index and argument reversal in the Neumann-series identification is easy to misread; writing the block formula explicitly, for example δ_{ij}δ_{r,r'} + [B_u]_{ij}(r,r') = [(I+χ̄_a L χ̄_a)^{-1}]_{j,i}(r',r), would clarify the convention.","section":"Remark 6.4"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the framework chapters are sound and well written, and I verified several of the key scalar-reduction identities independently; the risk in this manuscript is completeness versus ambition. Theorem 1.5 and the abstract promise an eighteen-model catalogue, but the text provided for review verifies none of the chapters beyond §6.2, and the semi-discrete product graph construction—needed for the continuous-time TASEP and Push-TASEP entries—is explicitly deferred. I recommend insisting that the author either include the deferred verifications and a complete per-model checklist, or soften the claim to what the text proves. The author's self-citation [Rod25] is used appropriately, and I found no circularity."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the core mechanism is good. The diamond linear problem, the dressing compatibility conditions, the Darboux theorem, and the scalar reduction to Hirota–Miwa are worked out with real care. Theorem 2.5 is a detailed proof that uses only resolvent existence, and the gauge equivalence to Nimmo's non-abelian Hirota–Miwa system is explicit and useful. This is the most serious candidate I have seen for a common algebraic mechanism behind closed equations for KPZ distribution functions. The product-graph construction is also a genuine new device for building multipoint data, and the recovery of the author's one-point equations [Rod25] as a special case is correct and honestly presented.\n\nNow the soft spots, in proportion. The main gap is between the framework and the advertised catalogue. The abstract and Theorem 1.5 present the eighteen-model result as established, but the text made available to me truncates in §6.2, so the per-model verifications for almost all of the eighteen models are not visible. For the continuous-time models in Chapter 8, the situation is worse: Remark 3.13 explicitly defers the semi-discrete product graph construction to future work, which means the multipoint equations for continuous-time TASEP, Push-TASEP, and ASEP are not actually derived in this version. That is not an internal inconsistency; the paper is honest about the deferral. But Theorem 1.5 overstates what is proven, and any referee should require either the missing verifications or a revised claim.\n\nThe analytic hypotheses are the second soft spot. Example 2.4 and Definition 3.4 introduce dressing compatibility under 'suitable analytic assumptions', and each model verification must check resolvent existence and convergence of kernel sums. This is load-bearing, and in the visible text the checks are asserted rather than shown. It is the kind of thing that can fail in a specific model, so the broad claim should be scaled back until the checks are done.\n\nOne minor complaint: the introduction claims no prior connection between the Hirota–Miwa equation and KPZ, while crediting [MQR24] with the non-abelian 2D Toda lattice, which is a reduction of the same hierarchy. That is a rhetorical overstatement. The citation pattern is otherwise appropriate; self-citation to [Rod25] is justified because the earlier paper is the direct input.\n\nWho is this for? Probabilists working on exact solvability in KPZ, and integrable-systems people interested in new solutions of non-abelian Hirota–Miwa. It deserves a serious referee: the framework is important if it holds, and the core proofs are substantial enough to warrant expert time. My recommendation is to send it to peer review with the clear instruction that the eighteen-model theorem must either be fully verified in the manuscript or reformulated as a conditional statement pending the deferred work.","headline":"A serious Darboux framework with genuinely careful core proofs, but the advertised 18-model catalogue is not yet established in the text under review.","tokens_in":73871,"tokens_out":2357,"would_cite":true,"duration_ms":24601,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60K35","37K10","37K60","82C22"],"pacs":[],"model":"deepseek-v4-flash","headline":"A single Darboux-dressing mechanism yields closed matrix equations for 18 solvable KPZ models.","keywords":["KPZ universality class","Fredholm determinants","non-abelian Hirota-Miwa","Darboux transformation","diamond equations","multipoint distribution functions","exactly solvable models","integrability"],"falsifier":"Find a single model among the eighteen where the dressing compatibility conditions fail at some finite parameter value--for instance, a parameter regime where the resolvent (I - zK(u))^{-1} does not exist or the kernel sums defining K diverge--and the closed matrix equation for that model is not established. For the semi-discrete models specifically, the product graph extension is deferred to a future work (Remark 3.13), so a failure of the claimed extension for continuous-time TASEP would falsify the chapter-8 verifications.","tokens_in":72775,"feed_emoji":"📈","tokens_out":3566,"duration_ms":26746,"temperature":0.7,"pith_summary":"The paper establishes that the Fredholm determinant data of eighteen exactly solvable KPZ models---discrete particle systems, vertex models, polymers, and the KPZ fixed point---can be embedded into one overdetermined linear problem on a directed lattice, whose compatibility conditions (the diamond equations) are preserved by a Darboux-type dressing transformation. The upshot is that each model's multipoint distribution function obeys an explicit closed matrix differential-difference or PDE, and verifying such an equation for a new model reduces to checking a handful of linear conditions on its kernel. The framework also shows the diamond equations are a gauge-equivalent reparametrization of the non-abelian Hirota-Miwa system, connecting the KPZ class to a central object of classical integrability.","feed_headline":"18 KPZ models obey one matrix integrability mechanism","feed_subtitle":"Fredholm determinant data dresses into closed matrix equations that reduce to the Hirota-Miwa system.","key_machinery":"The central object is the diamond linear problem, an overdetermined linear system for a wave function Psi on a lattice with commuting shifts T, S1, S2, whose compatibility conditions are the diamond equations: two non-mixed conditions on the C-weights and Lambda-weights separately, and one mixed condition coupling them. Seed data extracted from a Fredholm kernel's shift structure solves these equations, and the Darboux transformation M(u) = I + z Phi(u) R(u) Psi(u) with dressed weights M_i = M(Tu)^{-1} C_i M(S_i u) preserves the diamond equations, producing finite-dimensional matrix observables that encode multipoint distributions. The product graph construction with admissible propagators supplies dressing-compatible kernels in the fully discrete regime, Euclidean division in a polynomial quotient algebra supplies seed data for vertex and polymer models, and for invertible C-weights the diamond equations are gauge-equivalent to the non-abelian Hirota-Miwa system.","core_discovery":"The paper's central claim is Theorem 1.5: for each of the eighteen listed models, the model's Fredholm determinant data satisfies the hypotheses of the Darboux theorem appropriate to its scaling regime (discrete, semi-discrete, parabolic, or continuum), and therefore its dressed observable M (or additive A) is governed by a closed matrix equation---the mixed dressed diamond equation---in every case. The mechanism is that elementary seed data from the shift structure of a Fredholm kernel provide simple solutions to the diamond equations, and a Darboux transformation dresses this seed data into a finite-dimensional matrix observable while preserving the diamond structure. The distinct one-point equations of the author's earlier work become scalar specializations of the resulting Hirota-Miwa reduction, and the non-abelian Hirota-Miwa system of Nimmo is recovered as the underlying integrable system in a suitable gauge.","pith_inferences":["If the framework survives close scrutiny, it suggests that exact solvability in the KPZ class is not a zoo of unrelated methods but one algebraic structure acting in different scaling regimes; one could test this by taking a newly solved model and running the five-step verification procedure to predict its multipoint equation before any per-model computation.","The gauge equivalence with non-abelian Hirota-Miwa implies that the Fredholm determinants arising in KPZ theory constitute a new class of solutions to the classical Hirota-Miwa system---solutions constructed from random growth rather than algebraic geometry---and the stochastic origin likely imposes spectral constraints worth investigating.","The trace-defect terms in the parabolic and continuum bilinear equations suggest that for m >= 2 observation points the multipoint distribution functions are governed by the same hierarchies as their scalar one-point limits only up to a deterministic correction built from the observable; this might be interpreted as a genuinely noncommutative footprint of multipoint statistics.","A plausible testable extension is to apply the product graph construction to models not on the list but with known Fredholm determinants (e.g. KPZ equation with narrow wedge, q-Whittaker processes) to predict their multipoint equations as a check of the framework's reach."],"forward_implications":["The finite-time multipoint distribution functions of all eighteen listed models satisfy explicit closed matrix equations, most for the first time; the six one-point bilinear equations of the author's earlier work are recovered as scalar specializations.","Verifying a closed nonlinear equation for a new exactly solvable KPZ model is reduced to checking a handful of linear conditions on its Fredholm kernel data, so the framework is a construction method rather than a per-model computation.","The scalar reduction produces variable-coefficient Hirota-Miwa equations for Fredholm determinants, so any model accommodated by the framework automatically produces a Hirota-form bilinear equation by the same mechanism.","The diamond equations are a gauge-equivalent reparametrization of the non-abelian Hirota-Miwa system, establishing a two-way connection between KPZ distribution theory and classical integrability: Painlevé II, KP, and 2D Toda appearances become reductions of one system.","For the KPZ fixed point, the framework independently derives the matrix KP equation of Quastel and Remenik and shows the multipoint distribution function satisfies the KP equation with a forcing term built from the matrix observable."],"supporting_citations":[{"why":"Supplies the non-abelian Hirota-Miwa system that the diamond equations are shown to be gauge-equivalent to.","marker":"[Nim06]"},{"why":"Proves the KP equation for KPZ fixed point multipoint distributions, the one prior result the framework re-derives as its continuum special case.","marker":"[QR22]"},{"why":"Provides the Fredholm determinant formulas for the KPZ fixed point that serve as input data in Chapter 10.","marker":"[MQR21]"},{"why":"The author's one-point bilinear equations, which the scalar reduction recovers as specializations.","marker":"[Rod25]"},{"why":"Supplies the multipoint Fredholm determinant formula for discrete-time TASEP that the Chapter 6 verification starts from.","marker":"[MR22]"},{"why":"Pre-limit non-abelian 2D Toda for polynuclear growth, the key prior example of a pre-limit integrable lattice equation.","marker":"[MQR24]"},{"why":"Painlevé II representation of Tracy-Widom, the historical root of the connection between KPZ distributions and integrable ODEs.","marker":"[TW94]"}],"fun_headline_variants":["18 KPZ models obey one matrix mechanism","One matrix equation governs 18 KPZ models","Diamond equations unify 18 KPZ models","Non-Abelian Hirota-Miwa for 18 KPZ models"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim depends on every one of the eighteen models satisfying the dressing compatibility conditions (2.7)--resolvent existence and convergence of the kernel sums--but the text introduces these only under suitable analytic assumptions and does not display the per-model verification for all eighteen.","fun_headline_variants_meta":{"raw":{"variants":["18 KPZ models obey one matrix mechanism","One matrix equation governs 18 KPZ models","Diamond equations unify 18 KPZ models","Non-Abelian Hirota-Miwa for 18 KPZ models"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00022,"raw_usage":{"total_tokens":1459,"prompt_tokens":972,"completion_tokens":487,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":588,"completion_tokens_details":{"reasoning_tokens":422}},"tokens_in":588,"tokens_out":487,"duration_ms":4636,"temperature":1.0,"reasoning_tokens":422,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:00:13.667027+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a single model among the eighteen where the dressing compatibility conditions fail at some finite parameter value--for instance, a parameter regime where the resolvent (I - zK(u))^{-1} does not exist or the kernel sums defining K diverge--and the closed matrix equation for that model is not established. For the semi-discrete models specifically, the product graph extension is deferred to a future work (Remark 3.13), so a failure of the claimed extension for continuous-time TASEP would falsify the chapter-8 verifications.","supporting_citations":[],"review_version":1}