{"id":"1aa412d7-1ca0-48d1-b68f-7280c8b26b33","arxiv_id":"1908.08907","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Known exact finite-volume KPZ fluctuation probabilities are expressed as traces on Riemann surfaces for half-integer polylogarithms, and prior formulas by Prolhac and by Baik and Liu are proved equivalent.","lead":"This paper rewrites known exact formulas for one-dimensional KPZ growth fluctuations on a ring as integrals of holomorphic differentials on infinite-genus Riemann surfaces, summed over the sheets of a covering map to a cylinder. It also shows that two earlier independent formulas for flat and sharp-wedge initial conditions are equivalent.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The equivalence proof between [39] and [40] depends on the unproved uniqueness of the solutions ν_P(s) of χ'_P(ν)=s stated at eq. (122); without it, the contour deformation in §5.1.1 that yields (4) for c>0 is not justified.","rationale":"I have read sections 5.1.1 and 5.1.2 carefully. The unconditional step is the expansion of the Baik-Liu Fredholm determinant (124) into (4) for c<0; the conditional step is the Fourier inversion of the [39] generating function (122). The substitution s=-iχ'_P(ν) requires global univalence of χ'_P on the half-plane Re ν>0 (or at least on a contour deformable to c+iℝ). This is exactly the uniqueness conjecture stated in (122) and (131). The paper explicitly labels it 'conjectured to be unique', so the limitation is self-flagged. I find no fatal internal inconsistency in the extensive analytic-continuation machinery; appendices and section 4 provide the needed identities, and the [40] derivation is algebraically clean. The same conjecture is needed for sharp wedge via (131). Thus the reader's weakest assumption is indeed the most load-bearing unresolved point, and the CONDITIONAL verdict with that condition is appropriate. No change to the reader's verdict is needed.","tokens_in":71469,"tokens_out":13940,"duration_ms":134175,"concrete_test":"Numerically test the uniqueness of ν_P(s) for flat-condition sets: for each P with |P|+=|P|- and |P|≤4, take the known real positive solution of χ'_P(ν)=s at large s (where χ'_P(ν)≃(2/√π)ν^{3/2}) and continue it down to s→0+ using a predictor-corrector homotopy, monitoring the argument principle on the boundary of {Re ν>0, |ν|<R} to detect any additional roots. If any second solution with Re ν>0 appears for some P, or if χ''_P vanishes at a solution, the change of variables in §5.1.1 is invalid and the claimed equivalence between [39] and [40] is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the [39] and [40] expressions agree is established by showing both equal the integral (4): §5.1.2 obtains (4) for c<0 unconditionally from [40], while §5.1.1 obtains (4) for c>0 from the generating function (122) of [39] by the substitution s = -iχ'_P(ν). This substitution requires that for each P the equation χ'_P(ν)=is have exactly one solution in Re ν>0 for each real s; equivalently, for s>0, χ'_P(ν)=s has a unique solution with Re ν>0. The paper states this as 'conjectured to be unique' at (122) and (131), inheriting it from [39] without proof. If the conjecture fails, the contour {c+iℝ} is not a bijective image of the s-integration contour, extra critical-point or branch-cut contributions appear, and (4) for c>0 need not equal the [39] expression; then the proof that [39]=[40] collapses (the direct [40]→(4) route alone establishes only c<0). Since this conjecture is explicitly unresolved and load-bearing for one of the two equivalence directions, the conditional verdict is appropriate.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a unified geometric reformulation of exact finite-volume KPZ fluctuation formulas with periodic boundary conditions. It constructs infinite-genus Riemann surfaces R, ˇR, and R∆ associated with half-integer polylogarithms, defines meromorphic functions χ, χ′, χ′′, e^{2J}, e^{2K}, etc. on these surfaces, and expresses the one-point probabilities for flat, sharp wedge, and stationary initial conditions, as well as the multiple-time joint distribution for sharp wedge initial condition, as integrals of traces of holomorphic differentials over the covering maps to an infinite cylinder. In Section 5, these formulas are shown to be equivalent to the earlier Bethe-ansatz-based expressions of [39]–[42], with the additional claim that the distinct formulas of [39] and [40] agree once their sheet sums are reinterpreted as traces on the same Riemann surface. The paper also discusses connections to stationary large deviations, particle-hole excitations, and KdV/KP soliton tau functions.","tokens_in":71690,"tokens_out":7229,"duration_ms":77604,"significance":"If the claimed equivalences hold, the paper gives a genuinely unifying picture of exact finite-volume KPZ results: complicated sums over particle-hole excitations become traces of holomorphic differentials over sheets of ramified coverings, and the previously separate [39] and [40] derivations appear as different choices of fundamental domain. The analytic-continuation identities are derived in detail in the appendices, and the manipulations from the known formulas are transparent and largely self-contained once the ingredients of [39]–[42] are accepted. The main caveat is that one direction of the equivalence relies on an unproved and explicitly labeled conjecture about the uniqueness of the solutions of χ′_P(ν)=s. This does not affect the reformulation itself, but it does affect the strength of the claimed proof that the [39] and [40] formulas agree.","major_comments":[{"comment":"The passage from the generating function (122) to the probability formula (123), and hence to (4) for c>0, requires that for every finite P the equation χ′_P(ν_P(s))=s have a unique solution with Re ν_P(s)>0 when s>0. The manuscript states at (122) that this is “conjectured to be unique” and does not supply a proof or a reference to one. The change of variables s=−iχ′_P(ν) is a bijective substitution only under this uniqueness; if multiple roots exist, the deformed contour {c+iℝ} is not the image of the s-integration contour, and extra contributions can appear. Consequently the claimed equivalence between the [39] expression and (4) for c>0 is not fully established, and the subsequent conclusion that the [39] and [40] flat-initial-condition formulas agree is conditional on this conjecture. This is a load-bearing gap that should be resolved by a proof, a precise citation of a proof, or a clearly stated weakening of the claim.","section":"§5.1.1, Eq. (122)"},{"comment":"The same unproved uniqueness issue appears for sharp wedge initial condition: ν_{P,H}(s) is defined as the solution of χ′_{P,H}(ν)=s and is again stated to be “conjectured to be unique.” The derivation of (133) from (131) and the later identification of (10) with the [39] formula for c>0 depend on this conjecture. The direct derivation from [40] in §5.2.2 gives (10) only for c<0. Since the paper’s stated goal includes showing that the [39] and [40] sharp-wedge expressions agree, this second instance of the same unresolved assumption should also be addressed explicitly. The manuscript would be strengthened either by proving the conjecture or by making the conditional nature of the [39]=[40] statement prominent throughout Section 5.","section":"§5.2.1, Eq. (131)"}],"minor_comments":[{"comment":"The definition of I0 is split across the text and depends on the sign of c, but this is stated only parenthetically in the paragraph after the display. Since (4) is claimed for both c<0 and c>0, a displayed convention for I0 with explicit branch choices would improve clarity.","section":"§2.2, Eq. (4)"},{"comment":"The sentence describing the automorphism T says that the map is a homeomorphism and “hence an automorphism since it is locally holomorphic.” Strictly speaking one should also note that the inverse is holomorphic, or cite the standard fact that a bijective locally biholomorphic map between Riemann surfaces is a biholomorphism.","section":"§3.8.2"},{"comment":"The sentence “The rest of the section is essentially a more detailed version of the derivation of equation (22), run backwards” is vague because equation (22) is presented later in §2.6.3. A forward reference and a one-sentence explanation of the logical structure would help the reader.","section":"§5.1.2"},{"comment":"The manuscript contains numerous small typographical and spacing issues, such as missing spaces around mathematical expressions in prose (e.g., “R∆”, “χ∆P”, “S0P”). A careful copyedit would noticeably improve readability.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper is a substantial and mostly careful reformulation of known exact results, with detailed appendix calculations. The main obstacle is the unproved uniqueness conjecture behind the [39]-to-(4) and [39]-to-(10) directions. If the authors can supply a proof or a precise reference for the conjecture, or alternatively restrict the claimed full equivalence and state the [39]=[40] agreement as conditional on that conjecture, the paper would be suitable for publication. The novelty is conceptual rather than computational, but the geometric unification and the explicit checks against prior formulas give it value for the readership."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The one thing you should know: this is a careful, honest reformulation of known finite-volume KPZ exact formulas, and it does prove something genuinely new—that the Prolhac and Baik-Liu expressions for flat and sharp-wedge initial conditions agree. The Riemann-surface machinery is real, not decorative: the trace over sheets cleanly packages the complicated particle-hole sums, and the explicit analytic continuations in section 4 are worked out in exceptional detail. I checked the appendices against the quoted identities in [39] and [40]; the derivations are transparent and no steps look fudged. No fitting, no invented entities. This is a legitimate organizational step.\n\nThe soft spot is exactly what the stress-test flagged. Equation (122) needs the uniqueness of nu_P(s) for chi'_P(nu)=s, Re nu>0, to justify the contour deformation that converts the generating function into the integral representation for c>0. The paper states this as a conjecture, inherited from [39], and the same issue appears at (131) for sharp wedge. So the equivalence proof is conditional: the c<0 direction from Baik-Liu is unconditional, and the cylinder analyticity of the trace is established, but bridging to c>0 through the generating function relies on uniqueness. If that conjecture fails, the proof that [39]=[40] would need another route. To the paper's credit, the gap is explicit and it is not a new flaw introduced here—it comes from prior work.\n\nA minor point, not a flaw: the Riemann surfaces and trace formulas are tailored to reproduce the known sheet sums. That is the point of a unifications, and it means the paper predicts no new physics. The value is organizational and structural.\n\nWho is this for? Researchers in integrable probability and finite-volume KPZ who want a cleaner language for the exact results, plus anyone interested in infinite-genus Riemann surfaces attached to polylogarithms. It deserves a serious referee: the claims are checkable, the appendices support the hard analytic continuations, and the equivalence proof is worth verifying carefully.\n\nRecommendation: send it to a refereed journal. The conditional verdict is right. If the uniqueness conjecture is later resolved, or if the Baik-Liu route is treated as primary, the paper is fully acceptable. Even as it stands, it is a solid contribution with one openly-flagged gap.","headline":"Careful Riemann-surface reformulation of known KPZ formulas with a genuinely new equivalence proof between Prolhac and Baik-Liu; one explicitly-flagged uniqueness conjecture is load-bearing for half of that equivalence.","tokens_in":72222,"tokens_out":2454,"would_cite":true,"duration_ms":28510,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14H55","30F30","37K10","60K35"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that the exact finite-volume KPZ height probabilities for flat, wedge, and stationary initial conditions are traces of holomorphic differentials on one infinite-genus Riemann surface, and that two independent derivations…","keywords":["KPZ equation","periodic boundary conditions","Riemann surfaces","half-integer polylogarithms","Bethe ansatz","TASEP","Fredholm determinants","KdV equation"],"falsifier":"Find one finite set $P\\subset\\mathbb{Z}+1/2$ and one $s>0$ for which $\\chi'_P(\\nu)=s$ has two distinct solutions with $\\mathrm{Re}\\,\\nu>0$, or none; that would falsify the conjectured uniqueness behind (122) and require modifying the contour deformation leading to (4). Short of that, truncating the sheet sum in (4) at increasing $|P|$ and checking that the $c<0$ and $c>0$ evaluations agree, or comparing with Monte Carlo data for TASEP on a ring at intermediate times, would test the claimed equivalence numerically.","tokens_in":71196,"feed_emoji":"📐","tokens_out":9810,"duration_ms":91968,"temperature":0.7,"pith_summary":"The paper is trying to establish that the known exact formulas for one-dimensional KPZ fluctuations in a finite ring—flat, sharp-wedge, and stationary initial conditions, and even multiple-time joint distributions—are not separate calculations but different faces of one geometric object: a holomorphic differential on a Riemann surface associated to half-integer polylogarithms, integrated around an infinite cylinder and summed over all sheets of a covering map. If correct, this unification turns the large sums over Bethe-ansatz particle-hole excitations in the earlier formulas into traces of a single differential, and it supplies the missing proof that the two known derivations of the flat and wedge distributions agree. The same surface already appears in stationary large deviations, so the paper suggests that the full finite-time dynamics can be recovered from static large-deviation data by analytic continuation. The flat-initial-condition probability is also identified as a KdV soliton tau function averaged over the soliton velocity, with the integration variable playing the role of a moduli parameter for a degenerate hyperelliptic surface.","feed_headline":"One Riemann surface unifies exact KPZ probabilities on a ring","feed_subtitle":"Flat, wedge, and stationary height probabilities become traces of one holomorphic differential on an infinite-genus surface.","key_machinery":"The Riemann surface $R$ is the natural single-valued domain of the half-integer polylogarithm $\\chi_{\\emptyset}(\\nu)=-\\mathrm{Li}_{5/2}(-e^{\\nu})/\\sqrt{2\\pi}$; its sheets are indexed by finite subsets of $\\mathbb{Z}+1/2$, which gives it the topology of an infinite-dimensional hypercube. Quotients by translation automorphisms produce $\\check{R}$, while quotients by involutions that remove chosen branch points produce $R_{\\Delta}$. The key operation is the trace of a holomorphic differential over the covering map to the cylinder $C$, $(\\mathrm{tr}_{\\rho}\\,\\omega)(q)=\\sum_{p:\\rho(p)=q}\\omega(p)$; because the trace of a holomorphic differential is holomorphic on $C$, the loop of integration in the probability formula may be moved freely. The paper proves that the exponential building blocks $e^{2I}$, $e^{I+J}$, $e^{2J}$, and $e^{2K}$ are well-defined meromorphic functions on these surfaces once the Vandermonde determinants and powers of $i/4$ are supplied by analytic continuation.","core_discovery":"Equations (1), (6), (11), and (13) are claimed to be exactly equivalent to the known formulas of [39]–[42]. The mechanism is that the finite subsets of half-integers labelling the sheets of $\\check{R}$ and $R_{\\Delta}$ are the same objects as the particle-hole excitation sets in the Bethe-ansatz sums, so what looked like a discrete sum over excitations is actually a trace over a covering map from the Riemann surface to the cylinder $C$. Carrying out the trace reproduces the earlier formulas, and comparing the two choices of fundamental domain shows that the expressions in [39] and [40] agree, a fact that had previously only been checked numerically. The paper also rewrites the multiple-time formula (13) from [42] and analyses the pole structure of the integrand, showing that poles occur exactly when two points coincide on the same Riemann surface.","pith_inferences":["If the paper's picture holds, the entire finite-time transition probability is an analytic continuation of stationary large-deviation data, making the excited-state spectrum a derived object rather than an input.","A direct numerical check at small time that the KdV tau function approaches the Painlevé II scaling solution would test the soliton-gas interpretation of the flat probability.","The trace formalism looks transferable to other solvable exclusion processes and to finite asymmetry, with the half-integer polylogarithm replaced by the corresponding special function."],"forward_implications":["The flat formula (4) and the wedge formula (10) are provably the same whether one starts from the derivation in [39] or the derivation in [40], closing the previous gap.","For sharp wedge initial condition, the particle-hole constraints $|P|_+=|H|_-$ and $|P|_-=|H|_+$ emerge automatically from the sheet structure of $R_{\\Delta}$, rather than being imposed by hand.","The flat probability is an $N=\\infty$ KdV soliton tau function, and the integration variable $\\nu$ is the common soliton velocity; higher KdV time variables appear as derivatives of $\\chi_{\\emptyset}$.","The stationary-initial-condition probability costs only the factor $-\\sqrt{2\\pi}e^{-\\nu}\\partial_u$ applied to the wedge differential, so the same Riemann-surface machinery covers it.","In the multiple-time formula, poles in $\\nu_{\\ell+1}=\\nu_{\\ell}+2i\\pi m$ appear exactly when $\\Delta_{\\ell+1}=\\Delta_{\\ell}+m$ and $P_{\\ell+1}=P_{\\ell}+m$, i.e. when the points coincide on the same Riemann surface."],"supporting_citations":[{"why":"supplies the stationary large-deviation generating function whose half-integer polylogarithms motivate the Riemann-surface construction.","marker":"[36]"},{"why":"furnishes the Bethe-ansatz generating functions (122) and (131) from which the flat and wedge formulas (4) and (10) are derived.","marker":"[39]"},{"why":"provides the independent Fredholm-determinant derivation whose agreement with [39] is one of the paper's main results.","marker":"[40]"},{"why":"supplies the stationary-initial-condition formula from which (11) is obtained.","marker":"[41]"},{"why":"provides the multiple-time joint distribution from which (13) and (16) are derived.","marker":"[42]"},{"why":"basis for the Bethe-ansatz spectrum of TASEP underlying the excited-state sums.","marker":"[44]"},{"why":"supplies the first excited states of the periodic TASEP spectrum needed for the sheet sums.","marker":"[45]"}],"fun_headline_variants":["Riemann surface encapsulates KPZ ring fluctuations","Infinite-genus surface yields exact KPZ formulas","KPZ on periodic boundaries: a Riemann surface proof","Polylogarithm surface explains KPZ ring statistics","Covering map connects KPZ and known exact results"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that for every sheet $P$ and every $s>0$ the equation $\\chi'_P(\\nu_P(s))=s$ has exactly one solution with $\\mathrm{Re}\\,\\nu_P(s)>0$—a uniqueness property the paper states as conjectured, not proved.","fun_headline_variants_meta":{"raw":{"variants":["Riemann surface encapsulates KPZ ring fluctuations","Infinite-genus surface yields exact KPZ formulas","KPZ on periodic boundaries: a Riemann surface proof","Polylogarithm surface explains KPZ ring statistics","Covering map connects KPZ and known exact results"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000627,"raw_usage":{"total_tokens":2824,"prompt_tokens":796,"completion_tokens":2028,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":412,"completion_tokens_details":{"reasoning_tokens":1954}},"tokens_in":412,"tokens_out":2028,"duration_ms":15884,"temperature":1.0,"reasoning_tokens":1954,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:25:54.550839+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find one finite set $P\\subset\\mathbb{Z}+1/2$ and one $s>0$ for which $\\chi'_P(\\nu)=s$ has two distinct solutions with $\\mathrm{Re}\\,\\nu>0$, or none; that would falsify the conjectured uniqueness behind (122) and require modifying the contour deformation leading to (4). Short of that, truncating the sheet sum in (4) at increasing $|P|$ and checking that the $c<0$ and $c>0$ evaluations agree, or comparing with Monte Carlo data for TASEP on a ring at intermediate times, would test the claimed equivalence numerically.","supporting_citations":[{"cited_title":"Prolhac, Spectrum of the totally asymmetric simple exclusion proces s on a pe- riodic lattice - first excited states , J","cited_arxiv_id":null,"evidence_quote":"supplies the first excited states of the periodic TASEP spectrum needed for the sheet sums."}],"review_version":1}