{"id":"e1072f70-470e-4bc5-857e-6172df2c1adb","arxiv_id":"2411.16401","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A kernel deformation makes finite-temperature sine kernel Fredholm determinants solvable through an explicit Riemann-Hilbert problem, giving exact large-distance asymptotics and reproducing Szego, Hartwig-Fisher, and Borodin-Okounkov formulas.","lead":"This paper builds a new method for computing how free-fermion correlation functions decay over long distances at finite temperature, by rewriting the relevant infinite-dimensional determinant as a solvable matrix problem. It reproduces three classical asymptotic formulas and adds exact subleading corrections, giving researchers a cheaper and more systematic route to results that previously required heavy steepest-descent analysis.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed exact identity (99) rests on Appendix C's Hankel non-degeneracy Δ_k≠0, which is x-dependent and unproven; the proof shows only equality of variations (C.53), and the overall constant is asserted without demonstration.","rationale":"The reader's weakest-assumption list already identifies the Hankel condition, meromorphicity, and the S1-to-R limit as the main restrictions, and the CONDITIONAL verdict is appropriate. I agree with that emphasis, and my stress-test pass sharpens one point: even granting meromorphicity and distinct zeros, the Hankel condition Δ_k ≠ 0 is x-dependent and therefore cannot be treated as a harmless generic assumption. The variational argument in Appendix C only proves equality of variations; the paper's assertion that the integration constant is fixed by comparing asymptotics is not demonstrated in the text. Thus the load-bearing concern is not merely a question of generality but a missing step in the proof of the claimed exact identity for all x > 0. I do not claim the final formula is wrong: the explicit solvable RHP, the recovery of Szegő, Hartwig-Fisher, and Borodin-Okounkov asymptotics, and the connection to the earlier effective form-factor method provide substantial independent support. But the claimed exactness of (99) requires either a proof of the Hankel non-degeneracy or a limiting argument that handles the degeneracies. The proposed numerical test would settle whether a vanishing Hankel determinant actually obstructs the RHP solution or is merely an artifact of the orthogonal-polynomial presentation. If the RHP solution survives at degenerate points, the concern dissolves and the proof can likely be repaired locally; if it fails, the statement needs to be restricted or supplemented with a degeneracy-removal argument.","tokens_in":52,"tokens_out":2701,"duration_ms":146926,"concrete_test":"Take a concrete meromorphic example with winding number n = −1, e.g. 1+θ(q) = A(q−z_1)/(q−p_1) with |p_1| < 1 < |z_1|, and compute μ_0(x) = (1/2πi)∮_{S^1} q^{−1} e^{−ω>(q)−ω<(q)} q^{−x+1} dq numerically for a sweep of x > 0. If μ_0(x) = 0 (equivalently Δ_1 = 0) at some x, solve the RHP (C.10)-(C.11) directly at that x; if a solution still exists, the Hankel assumption is unnecessary and the gap is cosmetic, whereas if no solution exists, (99) requires an explicit degeneracy-handling argument for all large x. Repeat for Δ_2 with a two-winding example to confirm the pattern.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that the effective form-factor expression (99) exactly equals the deformed Fredholm determinant for all x>0. The proof in Appendix C reduces this to an orthogonal-polynomial solution of the RHP (C.10)-(C.11). The decisive step is the orthogonality condition (C.12), which requires the Hankel determinants Δ_k = det(μ_{i+j−2}) to be nonzero for k = 1,...,n, with μ(q) = e^{−ω>(q)−ω<(q)} q^{−x+n}. The paper explicitly assumes Δ_k ≠ 0 after (C.13), then derives δ log τ_eff = δ log τ_HF (C.53). This is an equality of variations, hence equality up to a constant; the paper states that the constant is fixed by comparing asymptotics but does not exhibit that comparison. More importantly, Δ_k depends on x through the factor q^{−x+n}, so even for generic meromorphic θ, degeneracies can occur at isolated x. If Δ_k = 0, the Gram matrix is singular, the polynomials p_n are not defined, the resolvent formula in (C.19) fails, and (99) is not established. The paper says the distinct-zero restriction of Sec. 2 can be lifted by 'proper limiting procedures', but no such limiting procedure is shown for the Hankel degeneracies. The introduction promises smooth θ with fast decay, while the proof requires meromorphic θ and an unanalyzed S1-to-R limit; both gaps are acknowledged but not resolved.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies Fredholm determinants of deformed sine kernels on S^1 (Eqs. (1)-(3)) with symbol 1+θ, where θ is a weight, and connects them with Toeplitz determinants (Eq. (6)). It introduces a deformed kernel V_C (Eqs. (11)-(12)) whose resolvent is found explicitly via a 2x2 Riemann-Hilbert problem (Eqs. (23)-(27)). A variational argument gives the closed formula (36) for the deformed determinant τ_C[ν], which for zero winding reduces to the strong Szegő formula (Eqs. (39)-(40)), and for nonzero winding n<0 after contour deformation yields the Hartwig-Fisher leading asymptotics (Eq. (53)). The full Toeplitz determinant is expressed in Eq. (59) as τ_C[ν] det_C(1-K) with an explicit integrable kernel K (Eq. (60)), leading to Slavnov-type and Borodin-Okounkov formulas. Section 5 identifies the deformed determinant with the effective form-factor expression (99), claimed as an exact identity for all x>0, and Appendix C attempts a direct proof via orthogonal polynomials.","tokens_in":24829,"tokens_out":5272,"duration_ms":57356,"significance":"The paper's strengths are its explicit resolvent construction, the closed variational formula (36), and the systematic derivation of the subleading kernel (60), which together reproduce the classical Szegő, Hartwig-Fisher, and Borodin-Okounkov results from one framework. The claimed exact identity (99) between the Fredholm determinant and the effective form-factor/Hartwig-Fisher expression would be a valuable bridge between heuristic form-factor methods and rigorous RHP analysis. However, the proof as written has load-bearing gaps: the orthogonal-polynomial solution in Appendix C assumes unproven Hankel non-degeneracy, and the equality of variations (C.53) determines the tau function only up to an undemonstrated constant. These issues must be fixed before the exactness claim is established.","major_comments":[{"comment":"The proof of (99) hinges on the existence of monic orthogonal polynomials p_j with nonzero norms h_j (C.12), equivalently on the Hankel determinants Δ_k = det(μ_{i+j-2}) ≠ 0 for k=1,...,n. The measure μ(q)=e^{-ω_>(q)-ω_<(q)} q^{-x+n} depends on x, so Δ_k are functions of x and can vanish at isolated values even for generic meromorphic θ. The paper simply says 'We will assume that our measure function μ(q) is such that Δ_k ≠ 0' and provides no argument that this holds for the class of symbols considered, nor a limiting procedure to handle degeneracies. Without this, the RHP solution (C.14)-(C.16) and all subsequent formulas in Appendix C are not established. This is a load-bearing assumption for the exact identity (99); it needs to be either proved, explicitly incorporated as a hypothesis in the statement, or removed by an approximation argument.","section":"Appendix C.1, after Eq. (C.13)"},{"comment":"Eq. (C.53) proves only equality of variations, δ log τ_eff = δ log τ_HF, which implies equality up to a ν-independent constant. The text states that the constant is fixed by comparing asymptotics in the main text, but the comparison is not carried out anywhere; Eq. (53) is derived for τ_C, not for τ_eff, and the asymptotic analysis of y_n(x) in Section 5 is only sketched. Since (99) claims exact equality for all x>0, the missing constant evaluation is a genuine gap. The authors should exhibit the explicit limiting argument (e.g., x→∞ or a special ν) that fixes the constant.","section":"Appendix C.2, Eq. (C.53)"},{"comment":"The introduction promises results for smooth θ decaying fast enough on R, and states that the R case is 'completely equivalent' to S1. The proofs, however, start with 'Let θ(q) in (1) be a meromorphic function in CP1' and assume all zeros of 1+θ are distinct; the contour deformation leading to (36) relies on these meromorphic zeros. No approximation argument is given for smooth θ, and the S1→R limit is explicitly declared out of scope ('We are not going to discuss the possible peculiarities'). Consequently the claimed scope is broader than what is proven. Please either restrict the theorem statements to the meromorphic setting or supply the missing limiting procedure.","section":"Sec. 2, after Eq. (8)"}],"minor_comments":[{"comment":"There are several typos: 'As a sequence of that' should be 'As a consequence of that', and 'shrinked' in Eq. (37) should be 'shrunk'.","section":"Sec. 1 and Eq. (37)"},{"comment":"The product notation ∏_{j,k}(z_j-z_k) is ambiguous/incorrect; the Hartwig-Fisher formula requires a product over ordered pairs or over j<k with appropriate powers. Please clarify the intended index ranges.","section":"Eqs. (48)-(49) and (B.8)"},{"comment":"The branch choice for the contour Uhat S^1 with φ∈[-π,π) introduces a cut for the logarithm; the text does not explain how the final result depends on this choice for ν that is not single-valued on S^1, and a remark on the branch-invariance would be helpful.","section":"Sec. 3, Eq. (53)"},{"comment":"After Eq. (95), the statement that the kernel acts on l^2(x,x+1,...) after a shift of n and m is asserted without showing the index shift explicitly. A few more lines would make this step transparent.","section":"Sec. 4, Borodin-Okounkov formula"}],"recommendation":"major_revision","confidential_remarks":"The paper fits the scope of mathematical physics and contains elegant constructions. My recommendation is driven by the two technical gaps in Appendix C: the unproven Hankel non-degeneracy and the missing constant-fixing argument after Eq. (C.53). If the authors can supply these, the result would likely be acceptable. I would also encourage a careful pass over the hypotheses in Section 2, since the current statements overclaim smooth θ while the proofs require meromorphic θ."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real technical contribution. The authors deform the finite-temperature sine kernel to a kernel with an exactly solvable matrix Riemann-Hilbert problem, get the resolvent in closed form, and from it recover Szego, Hartwig-Fisher, and Borodin-Okounkov asymptotics with a handle on subleading terms. The derivation is mostly self-contained and the final formulas match the classical benchmarks. The effective form factor identity (99), previously heuristic, is given a direct RHP proof in Appendix C.\n\nThe main body is in good shape. Equations (36) and (59) are the load-bearing exact statements: the tau function for the deformed kernel is an explicit exponential, and the original Toeplitz determinant equals that exponential times a Fredholm determinant of an explicit integrable kernel. I checked the algebra in the variational step and the resolvent formula (23); it is coherent. The contour-deformation arguments for nonzero winding are standard and the recovery of the Hartwig-Fisher leading term is clean.\n\nThe soft spots are all at the edges of the proof, and they are acknowledged but not resolved. The paper assumes theta is meromorphic with distinct zeros of 1+theta, then says the restriction can be lifted by limiting procedures without showing them. Appendix C assumes the Hankel determinants Δ_k are nonzero for k=1..n; that is x-dependent, so even for generic theta there may be isolated bad x-values. More importantly, Appendix C only proves equality of variations, δ log τ_eff = δ log τ_HF, so τ_eff = const·τ_HF; the constant is said to be fixed by comparing asymptotics, but the comparison is not exhibited. That is a genuine gap in the proof as written, though a fix is likely straightforward. The S1-to-R limit is explicitly deferred, so the paper's promise of smooth fast-decaying weights on R is stronger than what is actually shown.\n\nNone of this makes me doubt the central claim. The method is plausible, the internal algebra checks out, and the formulas reproduce known results in the limits where they are checked. The gaps are technical conditions and a missing constant-fixing argument, not a contradiction.\n\nWho should read this: anyone working on Fredholm determinants of sine-type kernels, Toeplitz determinants, or finite-temperature correlators in integrable models. It deserves a serious referee. I would send it to review and ask the authors to restate the theorems under the meromorphic, non-degenerate hypotheses they actually use, to prove or at least discuss the limiting procedures for the distinct-zero and Hankel conditions, and to write out the asymptotic comparison that fixes the constant in Appendix C.","headline":"Solid, technically useful paper; the RHP-based proof of the effective form factor identity is new and mostly sound, but the assumptions (meromorphic theta, nonzero Hankel determinants) and the missing constant-fixing argument in Appendix C need tightening before it is fully rigorous.","tokens_in":25444,"tokens_out":5668,"would_cite":true,"duration_ms":58175,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["47B35","45M05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that finite-temperature sine-kernel Fredholm determinants reduce to an explicit exponential of a quadratic phase functional times a Fredholm determinant of an integrable kernel, with the resolvent obtained from an exactly…","keywords":["Fredholm determinant","Toeplitz determinant","sine kernel","Riemann-Hilbert problem","finite temperature","resolvent","asymptotic expansion","Hartwig-Fisher formula"],"falsifier":"Take a concrete rational symbol, for instance $\\theta(q)= c\\, q^m$ or another meromorphic choice with known nonzero winding, compute the Toeplitz determinant numerically at several $x$, and compare with the right-hand side of (59) with $K$ from (60); any discrepancy beyond round-off would falsify the exact identity. Alternatively, for a smooth non-meromorphic $\\theta$, check whether the contour deformation to zero winding actually preserves the determinant; if the deformed and original Fredholm determinants differ, the leading asymptotic (53) fails.","tokens_in":24315,"feed_emoji":"📐","tokens_out":5931,"duration_ms":52832,"temperature":0.7,"pith_summary":"This paper gives an elementary derivation of the large-distance asymptotics of finite-temperature sine kernel Fredholm determinants on a circle, including the case where the phase shift has nonzero winding. The authors deform the original kernel into an effective form-factor kernel whose resolvent can be written exactly from a matrix Riemann-Hilbert problem. From that resolvent they obtain the deformed determinant as an explicit exponential of a quadratic functional of the phase shift, and express the original Toeplitz determinant as that exponential times a Fredholm determinant of an integrable kernel. For nonzero winding, the same machinery reproduces the Hartwig-Fisher asymptotics and, through a second contour deformation, the Borodin-Okounkov formula. If correct, the paper turns the previously heuristic effective form-factor method for static correlations into a systematic and fully controllable asymptotic expansion.","feed_headline":"Solvable kernel yields full sine-kernel asymptotics","feed_subtitle":"A contour-and-kernel deformation makes the resolvent explicit, recovering Szego, Hartwig-Fisher, and Borodin-Okounkov expansions.","key_machinery":"The central mechanism is the deformed kernel $V_C(q,p)$ of Eqs. (11)-(12), obtained from the original sine kernel by moving the contour and adding a correction built from $w_C(q)$. Its special feature is that the resolvent $R=(1+\\hat V_C)^{-1}-1$ has the same integrable form, so the vector Riemann-Hilbert problem for the $2\\times 2$ matrix $\\chi_C$ can be solved explicitly as a product of elementary matrices (Eq. (23)) in terms of the scalar functions $\\Omega_C$, $\\varphi_C$, and $b_C$. This explicit resolvent is what turns the variational formula for $\\ln\\tau_C[\\nu]$ into the exact exponential (36), and the difference $\\hat\\Delta = \\hat V_C - \\hat S$ between the deformed and original kernels into the Fredholm factor $\\det_C(1-\\hat K)$ in (59).","core_discovery":"The load-bearing assertions are Eq. (36) and Eq. (59): for a meromorphic phase shift with zero winding on a contour $C$, the deformed determinant is $$\\tau_C[\\nu] = \\exp\\!\\left( x\\oint_C \\frac{dq}{q}\\nu(q) - \\frac{1}{2}\\iint_C \\!\\left(\\frac{\\nu(k)-\\nu(q)}{k-q}\\right)^2\\! dk\\, dq \\right),$$ and the original Toeplitz determinant equals $\\det_C(1-\\hat K)$ times this exponential, with the integrable kernel $K$ given in Eq. (60). When the winding on the original circle is negative, deforming the contour to encircle the extra zeros of $1+\\theta(q)$ and summing over contours gives the Hartwig-Fisher leading asymptotics, Eq. (53), with the product over the zeros $z_k$ of $(z_j-z_k)$ and $z_k^{-x}/\\theta'(z_k)$. The subleading corrections are produced systematically by expanding the explicit resolvent, and the effective form-factor formula (99) is proven without recourse to the heuristic series.","pith_inferences":["The theory is stated for meromorphic $\\theta$; if the intended smooth symbols are obtained as limits of meromorphic ones, the same formulas should hold, but the limiting procedure is not exhibited and could in principle move corrections into the exponent.","The $S^1$ to $\\mathbb{R}$ limit, used to reach the physical sine kernel, is explicitly left unanalyzed; a natural check is to verify that the real-line asymptotics match the known mobile-impurity results in that limit.","The same solvable-resolvent mechanism should apply to Toeplitz+Hankel determinants and to kernels obtained by deforming $v_\\pm$ by $e^{\\pm g(q)}$, where the full expansion may fail but the leading asymptotics still follow.","The identity (110) for the full sine kernel suggests a direct Fredholm proof of the winding-shift relation, which the paper only proves in the zero-winding case; establishing it in general would give a purely determinant-theoretic derivation of the form-factor sum rule."],"forward_implications":["Large-$x$ asymptotics, including all subleading orders, follow from one explicit resolvent instead of a nonlinear steepest-descent analysis.","The effective form-factor heuristic for static two-point functions is placed on a firm footing: its thermodynamic-limit expression equals a determinant whose Riemann-Hilbert problem is exactly solvable.","The Borodin-Okounkov formula emerges as the $\\ell^2$ version of the same determinant, so the known subleading expansion is recovered as a special case.","The second Slavnov contour-sum formula (70) expresses the full Toeplitz determinant as a sum over tau functions of zero-winding contours, giving a closed expansion in the zeros of $1+\\theta$.","The rank-one determinant identities (109)-(110) follow in the zero-winding case and are conjectured by the authors to hold generally."],"supporting_citations":[{"why":"supplies the effective form-factor series and the kernel $V_C$ whose thermodynamic limit is the determinant under study; the paper's main result is to solve its resolvent exactly.","marker":"[22]"},{"why":"states the Hartwig-Fisher Toeplitz asymptotic that the nonzero-winding formulas (53) and (99) reproduce.","marker":"[35]"},{"why":"gives the Borodin-Okounkov Fredholm determinant presentation recovered in Section 4 as an $\\ell^2$ kernel.","marker":"[37]"},{"why":"presents the generalized sine-kernel Riemann-Hilbert approach that the explicit resolvent construction extends.","marker":"[31]"},{"why":"is the strong Szego formula recovered as the zero-winding case in Eq. (39).","marker":"[33]"},{"why":"supplies the vector Riemann-Hilbert notation and determinant formalism used to write the resolvent.","marker":"[38]"},{"why":"provides the orthogonal-polynomial solution of the Riemann-Hilbert problem used in Appendix C for the alternative proof.","marker":"[47]"},{"why":"backs the assumption that the Hankel determinants of the auxiliary measure are nonzero in the orthogonal-polynomial argument.","marker":"[48]"}],"fun_headline_variants":["Deformed kernel cracks finite-T Fredholm asymptotics","Explicit resolvent via effective form factors","Zero-winding path to Szego, Hartwig-Fisher, Borodin-Okounkov","Contour deformation makes resolvent explicit","Finite-T Fredholm determinants solved on contours"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation goes through only when the auxiliary measure $\\mu(q)=e^{-\\omega_>(q)-\\omega_<(q)} q^{-x+n}$ has nonzero Hankel determinants $\\Delta_k$ for $k=1,\\dots,n$, and when $1+\\theta(q)$ has distinct zeros; if a Hankel determinant vanishes, the orthogonal-polynomial solution collapses, and if $\\theta$ is merely smooth rather than meromorphic, the contour deformation that removes the winding number is not justified.","fun_headline_variants_meta":{"raw":{"variants":["Deformed kernel cracks finite-T Fredholm asymptotics","Explicit resolvent via effective form factors","Zero-winding path to Szego, Hartwig-Fisher, Borodin-Okounkov","Contour deformation makes resolvent explicit","Finite-T Fredholm determinants solved on contours"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000297,"raw_usage":{"total_tokens":1706,"prompt_tokens":911,"completion_tokens":795,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":716}},"tokens_in":527,"tokens_out":795,"duration_ms":6061,"temperature":1.0,"reasoning_tokens":716,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:10:06.209914+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a concrete rational symbol, for instance $\\theta(q)= c\\, q^m$ or another meromorphic choice with known nonzero winding, compute the Toeplitz determinant numerically at several $x$, and compare with the right-hand side of (59) with $K$ from (60); any discrepancy beyond round-off would falsify the exact identity. Alternatively, for a smooth non-meromorphic $\\theta$, check whether the contour deformation to zero winding actually preserves the determinant; if the deformed and original Fredholm determinants differ, the leading asymptotic (53) fails.","supporting_citations":[{"cited_title":"Gamayun, N","cited_arxiv_id":null,"evidence_quote":"supplies the effective form-factor series and the kernel $V_C$ whose thermodynamic limit is the determinant under study; the paper's main result is to solve its resolvent exactly."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"states the Hartwig-Fisher Toeplitz asymptotic that the nonzero-winding formulas (53) and (99) reproduce."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"presents the generalized sine-kernel Riemann-Hilbert approach that the explicit resolvent construction extends."},{"cited_title":"Szeg ˝o, Ein Grenzwertsatz uber die Toeplitzschen Determinanten einer reellen positiven Funktion, Mathematis- che Annalen 76 (4) (1915) 490–503","cited_arxiv_id":null,"evidence_quote":"is the strong Szego formula recovered as the zero-winding case in Eq. (39)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the vector Riemann-Hilbert notation and determinant formalism used to write the resolvent."},{"cited_title":"Bertola, The malgrange form and fredholm determinants, Symmetry, Integrability and Geometry: Methods and Applications (Jun","cited_arxiv_id":null,"evidence_quote":"provides the orthogonal-polynomial solution of the Riemann-Hilbert problem used in Appendix C for the alternative proof."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"backs the assumption that the Hankel determinants of the auxiliary measure are nonzero in the orthogonal-polynomial argument."}],"review_version":1}