{"id":"1f9d2219-da05-4fd9-9c49-db3afc42ba4a","arxiv_id":"2411.08129","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The τ-scaled spectral form factor of unorientable JT gravity agrees with the GOE random-matrix prediction through genus one, using new residue-based formulas for unorientable moduli volumes.","lead":"This paper computes the spectral form factor of unorientable Jackiw-Teitelboim gravity, the time-reversal symmetric version of a solvable two-dimensional quantum gravity, and shows it matches random matrix theory through genus one. It also supplies new analytic formulas for the moduli space volumes of unorientable hyperbolic surfaces, which were previously known only numerically or at lower genus.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The streamlined volume formulas (2.57)-(2.58) rest on an unproved cancellation of all O(b_i^{-1}) and z1=-z2 contributions; if any dropped term has a finite part, every volume and the SFF match through tau^3 fail.","rationale":"The reader's conditional verdict is appropriate. I focused on the streamlined volume computation because the non-Airy SFF terms, which are the genuinely new part of the paper, are direct integrals of v1,2, v1,1, and v1,0. The cancellation argument in Section 2.4 is the only support for discarding z1=-z2 poles and O(b_i^{-1}) terms; it is plausible but not a proof. The one-boundary case is simpler and the argument there is more convincing, since simple poles give pure b^{-1}e^{-bk/2} terms, but the two-boundary case involves theta functions and analytic continuation, where a finite leftover is not obviously impossible. The paper does have independent checks: agreement with Stanford's recursion for V_{1/2}, numerical symmetry of v1,1 and v1,0, and consistency with the Airy limit. These support but do not fully prove the streamline. The Airy tau^3 issue in (1.27) versus (1.31) is a second gap, but it is explicitly delegated to prior work [19] and does not affect the non-Airy terms that constitute the new claim; so I do not make it the primary concern. A direct symbolic check of the dropped terms would settle the matter, so the verdict should remain CONDITIONAL rather than ACCEPT.","tokens_in":42294,"tokens_out":12336,"duration_ms":126667,"concrete_test":"With a symbolic computation, evaluate the full inverse-Laplace transform of R_1^(p)(z1,z2) from (2.33), retaining the previously discarded last line (z1=-z2 residues) and the infinite sum, and extract the finite part for v1,1(b1,b2) and v1,0(b1,b2). If the finite part differs from (2.51)/(2.52) at any numerical point, e.g. b1=1, b2=2, the streamlining is invalid; if it agrees, the cancellation assumption is confirmed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim depends on the volumes v1,k(b1,b2) computed via the streamlined formulas (2.57)-(2.58), which in turn use the replacements R_1^(p)(z) -> -F_1^(p)(z)/(2y(z)) (eq. (2.27)) and R_{1/2}^(p)(z1,z2) -> -F_{1/2}^(p)(z1,z2)/(2y(z1)) (eq. (2.36)). The paper discards all O(b_i^{-1}) residues and all residues at z1=-z2 on the assertion that these must cancel: Section 2.3 argues that no combination of terms b^{-1} or b^{-1}e^{-bk/2} can be finite, and Section 2.4 argues that any z1=-z2 contribution has the structure (2.35) and must cancel entirely because its O(b_2^{-1}) part cannot cancel otherwise. This is a nontrivial statement about the full residue sum and is not proved. In particular, a Laurent piece c_{-1}/b_i + c_0 would contribute a finite c_0 if only the c_{-1}/b_i part cancels; the paper gives no argument excluding such pieces for g=1,n=2. Since v1,2, v1,1 and v1,0 (eqs. (2.50)-(2.52)) and the SFF terms (3.9)-(3.11) are all obtained after these discards, a surviving finite term would change the claimed equality (1.26)=(1.30).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper computes moduli-space volumes of unorientable JT gravity from the loop equations of an orthogonal matrix model with spectral curve y(z) ∝ sin(2πz), regularized via the (2,2p+1) minimal string model. The authors introduce 'streamlined formulas' that reduce volume computations to sums of residues, and use them to obtain explicit one- and two-boundary unorientable volumes up to genus one, including the three-crosscap volume v_{1,0}(b1,b2). They then compute the τ-scaled spectral form factor (SFF) of unorientable JT gravity up to genus one and compare it with the SFF obtained from the universal GOE microcanonical form factor. The two expressions are claimed to agree in all non-Airy infinite-series terms through τ^3, with the Airy τ^3 coefficient reconciled in the authors' prior work [19]. The paper also conjectures a multiple-zeta/multiple-polylogarithm structure for the polynomial parts of the unorientable volumes.","tokens_in":42592,"tokens_out":4361,"duration_ms":45235,"significance":"If the central claim is correct, the paper provides the first direct check of the BGS quantum-chaos signature for time-reversal-symmetric JT gravity, showing that the τ-scaled SFF matches universal GOE random-matrix theory order by order in the topological expansion. The computation of unorientable volumes up to genus one goes beyond previous results, and the streamlined residue formulas, if valid, are a substantial technical advance over Stanford's recursion. The paper includes several independent checks: the one-boundary volumes reproduce [12], the two-boundary volume v_{1/2,0} matches Stanford's recursion (2.40), and the long formula (2.52) passes a numerical symmetry check. However, the load-bearing simplifying step—dropping all O(b^{-1}) and z1=−z2 residues without proof—means the central claim is conditional on an unverified cancellation.","major_comments":[{"comment":"The streamlined formulas are derived by discarding all O(b_i^{-1}) terms and all residues at z1 = −z2, based on the assertion that they must cancel in the full residue sum. The argument in §2.3 is that no combination of terms ∝ b^{-1} or ∝ b^{-1}e^{-bk/2} can be purely finite, and in §2.4 that any z1 = −z2 contribution has the structure (2.35) and must cancel entirely because its O(b_2^{-1}) part cannot cancel otherwise. This is not a proof for the cases that matter: a Laurent piece c_{-1}/b_i + c_0 would produce a finite c_0 if only the c_{-1}/b_i part cancels, and the paper provides no argument excluding such pieces at (g,n) = (1,2). Since v1,2, v1,1, v1,0 in Eqs. (2.50)-(2.52) and the SFF terms (3.9)-(3.11) are all obtained after these discards, a surviving finite term would change the claimed equality (1.26) = (1.30). Please either prove the cancellation for general residue sums or provide an independent verification of v1,0(b1,b2)—for example by direct numerical evaluation of the full residue sum at fixed b1,b2 or by a comparison with Stanford's recursion at selected values.","section":""},{"comment":"The central claim that the SFF agrees with universal RMT up to genus one includes the Airy τ^3 coefficient. Eqs. (1.27) and (1.31) show that the Airy τ^3 coefficients differ (log(2t/β) versus −γ − log(2βτ^2) − 1/3), and the reconciliation is deferred to prior work [19]. As written, the present paper does not establish the full agreement; it establishes agreement only for the non-Airy infinite-series terms. The reader should be told precisely which parts of (1.26) and (1.30) are proved here and which parts are imported from [19], and the imported result should be stated explicitly if the paper is to stand alone.","section":""}],"minor_comments":[{"comment":"The displayed equation in §2.5 reads `−4z1z2F_1^{(p)}(z1,z2)e^{b1 z2}e^{b2 z2}/(2b1b2 y(z1))`; the first exponential should presumably be `e^{b1 z1}`, not `e^{b1 z2}`. The same typo appears in Eq. (2.49).","section":""},{"comment":"The equation contains a stray period inside the theta function: `θ(b2 − b1. )` should read `θ(b2 − b1)`.","section":""},{"comment":"The notation `+ O(b^{-1})` is used in a nonstandard sense: it is stated that the correct volumes are found by dropping all terms of this order. This usage should be defined explicitly, since in the standard meaning an O(b^{-1}) remainder is not a license to discard all O(b^{-1}) terms without further argument.","section":""},{"comment":"The convergence of the infinite residue sums is only discussed for b1 > b2; the analytic continuation in (2.39) and (2.49) is asserted without a detailed justification. A short remark on why the resulting function is analytic in (b1,b2) would improve rigor.","section":""},{"comment":"The display of v1,1(b1,b2) in (1.19) has a formatting issue: `b2 2b1 2` should be `(b2^2 b1)/2`, and the same expression in (2.54) has the factor 1/2 on the b1^3 and b2^3 terms, which is correct but not clearly reflected in the introduction's version.","section":""},{"comment":"The notation `O(t^{-1/2})` is used for subleading terms in t; since τ = t e^{-S0} is fixed, it would be helpful to explain explicitly which quantities are held fixed when t → ∞ in this expansion.","section":""}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid contribution to the JT-gravity/RMT correspondence, but its central claim rests on an unproved cancellation of subleading residues in the volume computation. The authors have already verified the simpler cases (one-boundary volumes and v_{1/2,0}) against known results, and the numerical symmetry check on v1,0(b1,b2) is encouraging. The missing piece is a verification or proof of the cancellation for (g,n) = (1,2). I would not recommend reject, because the gap is local and potentially fixable. I would also mention to the editor that the Airy τ^3 agreement is imported from the authors' previous paper [19]; if the present paper is intended as the self-contained announcement, that dependency should be made more prominent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a careful read. The paper computes the tau-scaled spectral form factor for unorientable JT gravity through genus one and matches it to the GOE random-matrix prediction, using the loop equations of the sin(2πz) orthogonal matrix model. Along the way it produces the first analytic unorientable volumes at (g,n)=(1,1) and (1,2), including v1,0(b) which was previously numerical, and conjectures a multiple-polylog/zeta structure for the volumes. The central SFF matching is real: the non-Airy terms come from two genuinely independent computations — volume integrals versus a Laplace transform of the GOE form factor — with no fitted parameters, and the divergence cancellation that leaves a finite, log(p)-independent result is an actual check on the machinery. The low-genus volumes also agree with the existing results in [12]. Credit where due: this is substantial and mostly credible.\n\nThe soft spot is exactly the one flagged in the stress test. The streamlined formulas (2.57)-(2.58) drop all O(b_i^{-1}) residues and all z1=-z2 residues on the assertion that they cancel. The O(b^{-1}) argument is fine: simple poles give only b^{-1} or b^{-1}e^{-bk/2}, and no combination of those can survive in a finite volume. The z1=-z2 claim is weaker. The paper argues any such residue has the form (2.35) with a θ(b2-b1), and since its O(b2^{-1}) part must cancel, the whole expression cancels. That does not exclude a surviving finite c0 part if only the c_{-1}/b2 piece cancels against something else. The authors do not prove this. I think the conclusion is probably right, because the final results pass several internal checks, but the derivation has a real gap that a referee should ask them to close — ideally by proving the cancellation or by numerically evaluating the dropped contributions at finite b.\n\nOne more caveat: the advertised agreement up to genus one is only literal after importing a pseudo-renormalization from the authors' own [19] for the Airy τ^3 coefficient. They are transparent about this, so it is not a hidden flaw, but it does mean the headline result leans on prior work for one term.\n\nBottom line: this is for people working on 2D gravity, matrix models, or spectral form factors. It deserves a serious referee, and I would probably cite it, but I would not treat the streamlined volume formulas as fully proven until the cancellation question is settled.","headline":"Strong computation with a real but fixable gap: the tau-scaled SFF match is credible, but the streamlined volume formulas rest on an unproved cancellation of z1=-z2 residues.","tokens_in":43190,"tokens_out":4843,"would_cite":true,"duration_ms":46099,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":null,"created_at":"2026-08-12T21:56:30.245214+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}