{"id":"a0cc7da9-95a3-40fe-8be9-fce5d262ece1","arxiv_id":"2412.16404","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Renormalized sine-Gordon measures are shown to be tight below a critical coupling via parabolic stochastic quantization, with a new pathwise global well-posedness result for the hyperbolic model in a smaller range.","lead":"This paper gives a new PDE-based construction of the sine-Gordon measure, a probability distribution used in two-dimensional quantum field theory, for couplings below a critical threshold. It also proves global well-posedness of a related stochastic wave equation in a smaller coupling range.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper proves compactness, not construction: Theorem 1.1 yields only a subsequential weak limit, and no argument identifies it with the sine-Gordon measure; the abstract and Section 4 overstate the result.","rationale":"The reader's verdict is CONDITIONAL, and the main reason given in the rationale is the gap between the compactness theorem and the claimed construction of the sine-Gordon measure. I agree that this identification gap is the most load-bearing issue for the central claim as advertised in the abstract and title. The reader's listed weakest_assumption, however, is Proposition 2.5 and the L-dependence of the imaginary Gaussian multiplicative chaos bound. That is a genuine further concern: the proof of Proposition 2.5 is genuinely sketched rather than fully written out, and the bounds (2.18)-(2.19) are the only quantitative estimates that carry the large-torus scaling into Theorem 1.2. But I do not see a concrete contradiction or false implication in the sketched proof; the omitted covering argument appears fillable. The more decisive issue is that even a completely rigorous proof of the technical estimates would still leave the paper proving only subsequential compactness, not the existence of a distinguished limit measure identifiable as the sine-Gordon measure. The paper itself uses the phrase 'the sine-Gordon measure ... constructed in Theorem 1.1' in Section 4, which is stronger than what the theorem states. A constructive proof should at least show that all subsequential limits coincide, or that the limits solve the characterizing equations of the sine-Gordon measure. Since no such argument is present, the advertised central claim is not established. The hyperbolic pathwise global well-posedness result in Appendix A is a separate, credible contribution and does not depend on this identification issue. The verdict should remain CONDITIONAL: the compactness results and the hyperbolic well-posedness are valuable and plausible, but the central interpretation as a construction of the sine-Gordon measure requires additional identification or a revised claim.","tokens_in":26030,"tokens_out":10485,"duration_ms":98778,"concrete_test":"To settle the identification issue, fix L and show that the full sequence of characteristic functionals Phi_N(phi) = E_{rho_{L,N}}[exp(i<u,phi>)] converges for every test function phi in a dense subspace of C^infty(T^2_L), and that the limit equals the characteristic functional of the known sine-Gordon measure (for example, the measure constructed by Dimock-Hurd for beta^2 < 8pi). If the limit depends on the subsequence chosen in Theorem 1.1, or if it disagrees with the known measure for some phi, then Theorem 1.1 cannot be legitimately described as a construction of the sine-Gordon measure. A concrete analytical route is to verify a DLR-type equation or the equivalent first-order cluster-expansion identities for any subsequential limit; a cheaper check is to compare the second moment of the limit against the Gaussian free field plus the first nontrivial cos(beta u) correction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central advertised claim is that the paper constructs the sine-Gordon measure. What Theorem 1.1 actually proves is the existence of a subsequence {rho_{L,N_j}} converging weakly to some limit rho_L; no uniqueness of the limit, no convergence of the full sequence, and no identification of any subsequential limit with the known sine-Gordon measure is supplied. The same issue recurs in Theorem 1.2, where the limit is obtained by passing to a further subsequence of tori, again without identifying the law. Section 4 nonetheless writes 'let rho_L be the sine-Gordon measure ... constructed in Theorem 1.1' near (4.22), and the abstract promises 'a simple PDE construction of the sine-Gordon measure'; this conditional compactness statement is weaker than the title. The technical heart, Proposition 2.5, is also only sketched in the last paragraph of Section 2.3 ('The rest of the proof follows from similar arguments'), and it is the only place where the L-dependence min(|B|,L^2) needed for Theorem 1.2 is derived; if that bound or its covering argument were to require an extra power of L or a logarithmic factor, the uniform estimates (3.10) and (4.22) would fail. Thus the proof of compactness is conditional on a partially delegated argument, and even granted it, the paper has not identified the limiting object as the sine-Gordon measure.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a stochastic-quantization construction of the Euclidean sine-Gordon measure for 0 < beta^2 < 4*pi, in finite volume and, by a large-torus limit, on R^2. The main results are Theorem 1.1 (subsequential weak convergence of the renormalized truncated measures rho_{L,N} to some probability measure rho_L on C^{-delta}(T^2_L)), Theorem 1.2 (an analogous subsequential limit of the torus-periodic measures in a weighted Besov space on R^2), and Theorem A.1 (pathwise global well-posedness of the damped hyperbolic sine-Gordon model for 0 < beta^2 < 2*pi). The proofs combine a deterministic global a priori bound for a nonlinear heat equation with distributional coefficients (Proposition 3.1), regularity estimates for the truncated stochastic convolution and the imaginary Gaussian multiplicative chaos (Propositions 2.4 and 2.5), and a weighted-space compactness argument in Section 4.","tokens_in":26298,"tokens_out":14227,"duration_ms":128494,"significance":"The deterministic a priori bound in Proposition 3.1 is clean and appears correct; the idea of exploiting the boundedness and Lipschitz property of the sine nonlinearity is a genuine simplification for the parabolic equation. The hyperbolic result in Appendix A is also a useful contribution. However, the paper's advertised claim of 'construction of the sine-Gordon measure' is stronger than what Theorems 1.1 and 1.2 establish: they provide only subsequential compactness, with no identification of any limit point as the sine-Gordon measure. In addition, the crucial L-dependence of the chaos bounds in Proposition 2.5 is only sketched. For these reasons the significance is conditional: if the missing identifications and estimates are supplied or the claims are revised to 'tightness' and 'subsequential limits', the paper would provide a valuable PDE route to compactness and well-posedness results, but in its current form the central interpretation overreaches the proofs.","major_comments":[{"comment":"Theorems 1.1 and 1.2 establish only subsequential weak convergence: Theorem 1.1 returns a subsequence {rho_{L,N_j}} converging to some probability measure rho_L, and Theorem 1.2 returns a further subsequence of torus sizes, again without uniqueness or identification. The text nevertheless refers to 'the sine-Gordon measure rho_L^beta constructed in Theorem 1.1' immediately after Theorem 1.1, and the proof of Theorem 1.2 near (4.22) uses the same phrase. No argument is supplied that identifies any subsequential limit with the known sine-Gordon measure: there is no proof of convergence of the full sequence, no analysis of correlation functions or characteristic functionals, and no uniqueness result for invariant measures of the limiting SPDE. Consequently, the abstract's phrase 'construction of the sine-Gordon measure' and the title's 'construction' overstate what is proven. The compactness statements themselves may be valuable, but the paper should either prove the identification or revise the claims and interpretation accordingly.","section":"Section 1.1, Theorems 1.1-1.2, and Section 4"},{"comment":"The proof of the uniform L-dependence in (2.18)-(2.19) is only sketched. The final paragraph of Section 2.3 reduces the problem to the local estimate (2.40) via a covering argument and then says 'The rest of the proof follows from similar arguments'; the actual covering argument converting the O(1) local bounds into the L^2 factor in (2.18), and the localization argument producing min(|B|,L^2) in (2.19), are not written out. These estimates are the only place where the L-dependence used in (4.19)-(4.20) for Theorem 1.2 is derived, so if the covering argument required an extra power of L or a logarithmic factor, the uniform bound (4.22) and hence Theorem 1.2 would fail. Please provide a complete proof of Proposition 2.5, or state the L-dependence explicitly as an assumption with a precise reference. In particular, the comparison in (2.42) invoking (2.18) for a fixed small L0 should be made fully explicit so that the argument is not circular.","section":"Section 2.3, Proposition 2.5"}],"minor_comments":[{"comment":"In the last display of Section 2.3, the factor 'e^{-i Psi^2_{L0,N}}' appears without the parameter beta; it should read 'e^{-i beta Psi^2_{L0,N}}' to match the preceding definitions.","section":"Eq. (2.42)"},{"comment":"The proof of Lemma 2.6 says 'By Plancherel's identity and Young's inequality' for what appears to be an L^p estimate with general p. Plancherel is specific to p=2; the estimate (2.25) follows from Young's convolution inequality and the Fourier multiplier bound, and the wording should be corrected.","section":"Lemma 2.6"},{"comment":"Proposition 3.1 is stated for (3.2) without the projector Pi_N, but (1.13) contains Pi_N in front of the Duhamel term. The application is justified if Pi_N is uniformly bounded on the relevant Besov spaces, but this should be stated explicitly.","section":"Proof of Theorem 1.1 and Eq. (1.13)"},{"comment":"The compact embedding lemma for the weighted Besov spaces is stated without proof ('the required modification is straightforward'). Since this embedding is used in the tightness argument for Theorem 1.2, a proof or a precise reference would be appropriate.","section":"Lemma 4.1"},{"comment":"Tightness on C^{-delta}(T_L^2) is deduced from the moment bound sup_N E ||u_{L,N}(0)||^p < infinity. Because C^{-delta} is not a separable space and balls are not compact, the argument relies on the fact that the same bound holds for a slightly smaller delta and on the compact embedding C^{-delta'} -> C^{-delta}; the authors should spell this out rather than say only that the choice of small delta was arbitrary.","section":"Section 3.2"}],"recommendation":"major_revision","confidential_remarks":"The paper's technical core, especially the deterministic a priori bound, is appealing, and the hyperbolic well-posedness result is a nice addition. The main concern is not internal inconsistency but the gap between the proven subsequential compactness and the advertised 'construction' of the sine-Gordon measure. This is fixable either by adding a genuine identification argument or by rewriting the title, abstract, and Section 4 to claim only tightness/precompactness. The missing details in Proposition 2.5 are also load-bearing; if they cannot be supplied, the claim of uniform L-dependence should be downgraded accordingly. I would be comfortable with publication after a major revision that addresses these two issues."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a solid, useful SPDE paper, and it delivers less than its title promises. What is genuinely new is the clean deterministic a priori bound (Proposition 3.1) giving global well-posedness of the parabolic sine-Gordon equation below 4π, the large-torus L-dependence estimates for imaginary Gaussian multiplicative chaos (Proposition 2.5), and the pathwise global well-posedness of the damped hyperbolic sine-Gordon model below 2π (Theorem A.1). The existence of the sine-Gordon measure itself was already known from Dimock–Hurd and Oh–Robert–Sosoe–Wang, which the authors acknowledge. The contribution is a simpler route via stochastic quantization, plus a compactness argument that works in infinite volume.\n\nWhere it is soft: the advertised \"construction\" is only subsequential compactness. Theorem 1.1 says there is a subsequence of ρ_{L,N} converging weakly to some ρ_L; there is no uniqueness, no full convergence of the sequence, and no identification of ρ_L with the known sine-Gordon measure. Section 4 then writes \"let ρ_L be the sine-Gordon measure constructed in Theorem 1.1\", which the theorem does not justify. Theorem 1.2 has the same issue: another subsequence of tori, again no identification. If the paper is read as showing that the stochastic quantization dynamics have a subsequential limit point, it is fine; read as constructing the sine-Gordon measure, it falls short. This is a presentation gap, not a fatal flaw, but it should be fixed.\n\nThe second soft spot is Proposition 2.5(ii), the bound min(|B|, L²) that drives the infinite-volume step. The proof is only sketched: after a covering argument, the text says \"the rest of the proof follows from similar arguments\". Since this is the estimate that makes Theorem 1.2 work, the omission is not cosmetic. I think the argument is probably right, but a referee should ask for the details.\n\nNo circularity here: the truncated measures are explicit, and the chaos bounds come from Gaussian estimates, not from assuming the target measure. The deterministic bootstrap in (3.10)–(3.11) and the weighted-Besov tightness in Section 4 look sound to me.\n\nThis is for readers working on stochastic quantization and singular SPDEs, especially those needing large-torus uniform estimates. It deserves a serious referee; the report should focus on Proposition 2.5 and the identification gap, both of which are fixable in revision.","headline":"A solid stochastic-quantization compactness result with a real presentation gap: the abstract oversells subsequential tightness as construction of the sine-Gordon measure.","tokens_in":26878,"tokens_out":2949,"would_cite":true,"duration_ms":26928,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T08","60H15","35K05","35L71"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves a simple PDE construction of the sine-Gordon measure for $0<\\beta^2<4\\pi$, in finite and infinite volume, via stochastic quantization.","keywords":["sine-Gordon model","stochastic quantization","imaginary Gaussian multiplicative chaos","tightness","parabolic sine-Gordon","hyperbolic sine-Gordon","renormalization"],"falsifier":"Compute the $p$-th moment of the $C^{-\\alpha}$ norm of $\\chi_B\\Theta_{L,N}$ on a large torus with $L\\gg 1$, at $\\beta^2$ close to $4\\pi$, and compare it with $C\\min(|B|,L^2)$: if the moment grows faster than $L^2$ or fails to be uniform in $N$, Proposition 2.5 is false and the proof of Theorems 1.1 and 1.2 does not go through. A direct check of the sketched final paragraph of Section 2.3 would also settle whether the proof of this estimate is complete.","tokens_in":25811,"feed_emoji":"🧮","tokens_out":17384,"duration_ms":129618,"temperature":0.7,"pith_summary":"The paper proves that for $0<\\beta^2<4\\pi$, the renormalized truncated sine-Gordon measures $\\rho_{L,N}$ on a two-dimensional torus are tight, so they admit weak limit points; the same compactness argument, applied along growing tori, yields a limiting probability measure on $\\mathbb{R}^2$. The proof works through the parabolic sine-Gordon equation, a stochastic heat flow for which the sine-Gordon measure is the formal invariant measure, and controls the flow by a single uniform bound on a renormalized exponential of the stochastic convolution, the imaginary Gaussian multiplicative chaos. The same deterministic estimate, run through the damped wave propagator, gives pathwise global well-posedness of the hyperbolic sine-Gordon model for $0<\\beta^2<2\\pi$. The appeal of the construction is that it uses only standard heat-flow smoothing, product estimates for rough distributions, and an invariance argument, rather than a dedicated measure-theoretic construction.","feed_headline":"Stochastic heat flow builds sine-Gordon measure below 4π","feed_subtitle":"One uniform bound on a renormalized chaos term yields compactness on tori and on R².","key_machinery":"The load-bearing object is the renormalized imaginary Gaussian multiplicative chaos $\\Theta_{L,N}=e^{\\frac{\\beta^2}{2}\\sigma_{L,N}}e^{i\\beta\\Psi_{L,N}}$, where $\\Psi$ is the stationary solution of the linear heat equation with space-time white noise and $\\sigma_{L,N}=\\mathbb{E}[(\\Pi_N\\Psi)^2]\\approx(2\\pi)^{-1}\\log N$ is the renormalization counterterm. Proposition 2.5 supplies the crucial quantitative estimate: for $\\beta^2<4\\pi$, $\\Theta_{L,N}$ is Cauchy in $L^p(C([0,T];C^{-\\alpha}))$ and its $C^{-\\alpha}$ norm localized by a disk $B$ has $p$-th moment bounded by $C\\min(|B|,L^2)$, uniformly in $N$ and $L$. This estimate feeds a deterministic global-in-time a priori bound (Proposition 3.1) for the equation $\\partial_t v+(1-\\Delta)v=\\sum_{\\kappa\\in\\{+, -\\}}e^{i\\kappa\\beta v}\\Theta_\\kappa$, proved with heat-kernel smoothing, a product estimate for distributions of opposite regularity, and the boundedness of $e^{\\pm i\\beta v}$. Combining the a priori bound with invariance of the truncated measure under the truncated flow (Lemma 1.3) gives uniform moment bounds on the initial data and hence tightness.","core_discovery":"The central claim is that below the first threshold, $\\beta^2<4\\pi$, the sequence of renormalized truncated sine-Gordon measures $\\rho_{L,N}$ on a fixed two-dimensional torus is tight in $C^{-\\delta}$, so it has weakly convergent subsequences; and along a subsequence of growing tori these limits converge weakly on $\\mathbb{R}^2$ to a probability measure $\\rho^\\beta$. The proof transfers the measure renormalization $\\gamma_{L,N}=e^{\\frac{\\beta^2}{2}\\sigma_{L,N}}$ to the dynamics, writes the truncated solution as $u=\\Psi+v$ with $\\Psi$ the stationary stochastic convolution of the linear heat equation, and reduces everything to a deterministic global a priori bound for a nonlinear heat equation forced by products $e^{\\pm i\\beta v}\\Theta_{L,N}$. The only stochastic input is a uniform $L^p$ bound, uniform in the truncation $N$ and the torus size $L$, for the imaginary Gaussian multiplicative chaos $\\Theta_{L,N}=e^{\\frac{\\beta^2}{2}\\sigma_{L,N}}e^{i\\beta\\Psi_{L,N}}$ and for its spatial localizations. The same a priori bound, adapted to the damped wave propagator, yields pathwise global well-posedness of the hyperbolic sine-Gordon model for $\\beta^2<2\\pi$.","pith_inferences":["Because the deterministic bound only uses the boundedness and Lipschitz character of $e^{\\pm i\\beta v}$, the same scheme should produce invariant measures for stochastic heat equations with other bounded Lipschitz interactions, with the admissible range set by the regularity of the corresponding multiplicative chaos.","The uniform volume growth $\\min(|B|,L^2)$ in Proposition 2.5 suggests the method is not special to sine-Gordon: exponential (Liouville-type) interactions on large tori should be treatable by the same localization argument.","The paper leaves open whether the subsequential limits coincide with the sine-Gordon measure constructed by other methods; showing convergence of characteristic functionals or matching Dirichlet forms would close the gap.","For the hyperbolic model, the restriction $\\beta^2<2\\pi$ comes from the single degree of smoothing of the damped wave propagator; reaching $4\\pi$ would likely require negative temporal regularity or additional stochastic objects."],"forward_implications":["For every $0<\\beta^2<4\\pi$ and every fixed torus size $L$, the family $\\{\\rho_{L,N}\\}_N$ is tight in $C^{-\\delta}$, so a subsequence converges weakly to a probability measure $\\rho_L^\\beta$ on $C^{-\\delta}(\\mathbb{T}_L^2)$.","Along a suitable subsequence of growing tori, the measures $\\rho_{L_j}^\\beta$ converge weakly on $\\mathcal{D}'(\\mathbb{R}^2)$ to a probability measure $\\rho^\\beta$, giving an infinite-volume sine-Gordon measure below the first threshold.","The parabolic sine-Gordon model is pathwise globally well-posed for $\\beta^2<4\\pi$: the truncated dynamics converge in probability to a limiting solution $u=\\Psi+v$.","The damped hyperbolic sine-Gordon model is pathwise globally well-posed for $0<\\beta^2<2\\pi$ with deterministic initial data in $H^s\\times H^{s-1}$."],"supporting_citations":[{"why":"Constructs the imaginary Gaussian multiplicative chaos $\\Theta$ below $8\\pi$ and provides the regularity estimates that Proposition 2.5 extends to the large-torus setting.","marker":"[31]"},{"why":"Gives the sine-Gordon measure and hyperbolic Gaussian multiplicative chaos below $4\\pi$, serving as the base case for the large-torus bounds.","marker":"[36]"},{"why":"Supplies the standard invariance argument for truncated stochastic quantization equations used in Lemma 1.3.","marker":"[15]"},{"why":"Provides the invariance result for stochastic nonlinear heat equations with exponential nonlinearities, referenced in Lemma 1.3.","marker":"[37]"},{"why":"Introduces weighted Besov spaces and large-torus localization techniques adapted in Section 4 for the infinite-volume limit.","marker":"[33]"},{"why":"Gives the variance asymptotics $\\sigma_{L,N}=(2\\pi)^{-1}\\log N+O(1)$ uniformly in $L$, used to define the renormalization constant.","marker":"[30]"},{"why":"Establishes the threshold structure of the dynamical sine-Gordon model and the renormalization framework in which the proof works below the first threshold.","marker":"[13]"}],"fun_headline_variants":["Sine-Gordon measure built via stochastic heat flow below 4π","Stochastic quantization yields sine-Gordon measure below threshold","New PDE construction yields sine-Gordon measure on torus and R²","Heat flow gives sine-Gordon measure; hyperbolic well-posedness for β²<2π"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the renormalized exponential $\\Theta_{L,N}$ of the stochastic convolution has $p$-th moments bounded by the volume of the spatial region, uniformly in the frequency cutoff $N$ and the torus size $L$; if that bound grows faster than $\\min(|B|,L^2)$, the tightness argument in Theorems 1.1 and 1.2 collapses.","fun_headline_variants_meta":{"raw":{"variants":["Sine-Gordon measure built via stochastic heat flow below 4π","Stochastic quantization yields sine-Gordon measure below threshold","New PDE construction yields sine-Gordon measure on torus and R²","Heat flow gives sine-Gordon measure; hyperbolic well-posedness for β²<2π"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00053,"raw_usage":{"total_tokens":2517,"prompt_tokens":875,"completion_tokens":1642,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":491,"completion_tokens_details":{"reasoning_tokens":1563}},"tokens_in":491,"tokens_out":1642,"duration_ms":11644,"temperature":1.0,"reasoning_tokens":1563,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T10:38:35.638293+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the $p$-th moment of the $C^{-\\alpha}$ norm of $\\chi_B\\Theta_{L,N}$ on a large torus with $L\\gg 1$, at $\\beta^2$ close to $4\\pi$, and compare it with $C\\min(|B|,L^2)$: if the moment grows faster than $L^2$ or fails to be uniform in $N$, Proposition 2.5 is false and the proof of Theorems 1.1 and 1.2 does not go through. A direct check of the sketched final paragraph of Section 2.3 would also settle whether the proof of this estimate is complete.","supporting_citations":[{"cited_title":"Hairer, H","cited_arxiv_id":null,"evidence_quote":"Constructs the imaginary Gaussian multiplicative chaos $\\Theta$ below $8\\pi$ and provides the regularity estimates that Proposition 2.5 extends to the large-torus setting."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the sine-Gordon measure and hyperbolic Gaussian multiplicative chaos below $4\\pi$, serving as the base case for the large-torus bounds."},{"cited_title":"Da Prato, A","cited_arxiv_id":null,"evidence_quote":"Supplies the standard invariance argument for truncated stochastic quantization equations used in Lemma 1.3."},{"cited_title":"Mourrat, H","cited_arxiv_id":null,"evidence_quote":"Introduces weighted Besov spaces and large-torus localization techniques adapted in Section 4 for the infinite-volume limit."},{"cited_title":"Hairer, M","cited_arxiv_id":null,"evidence_quote":"Gives the variance asymptotics $\\sigma_{L,N}=(2\\pi)^{-1}\\log N+O(1)$ uniformly in $L$, used to define the renormalization constant."}],"review_version":1}