{"id":"2cf09e8a-ac8c-4ff0-af49-3b656091b8bf","arxiv_id":"2502.02164","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The authors rigorously justify small-noise expansions for stochastic traveling waves, showing wave-speed corrections can be computed to arbitrary order with quantified errors.","lead":"This paper proves rigorous small-noise expansions for how random noise changes the speed and shape of traveling waves in reaction-diffusion equations. It shows how to compute the noise-induced speed correction to any order in the noise strength, with explicit error bounds over exponentially long timescales.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the central expansion is internally consistent and its key hypotheses are explicit and standard.","rationale":"The reader identified the spectral gap (HTw) as the weakest assumption. I agree that this is the most load-bearing hypothesis: all exponential decay estimates, the limiting expectations in Proposition 2.2, and the stability time scales depend on it. However, a load-bearing assumption is not the same as a load-bearing objection. The assumption is stated explicitly, is standard in the stability theory of traveling waves, and the paper does not claim to cover cases where it fails. I checked the internal logic of the expansion: Proposition 2.1 gives moment bounds for Yj, Proposition 5.1 is proven by an induction using the convolution bounds of Section 3, and Section 6 reduces limits of multilinear forms to iterated semigroup integrals via a lexicographic descent. The residual estimates in Sections 7 and 8 are consistent with the claimed O(σ^{r−1/2}) conditional error. The conditioning step in Proposition 2.6 is a straightforward application of the high-probability bound on A_stb, and the variance of the stochastic speed term B(σ,δ) is controlled by the exponentially large time horizon T*. I did not find a circularity, an omitted case that breaks the proof, or an unjustified exchange of limits. The lack of machine-checked formalization and reliance on prior work [11] are reasons for moderate rather than full confidence, but they are not correctness objections. Hence the verdict should remain ACCEPT, with no adjustment.","tokens_in":52650,"tokens_out":31102,"duration_ms":311153,"concrete_test":"Verify the most delicate analytic step by independently re-deriving Proposition 2.2 for r = 3 and Λ[Y1, Y1]: the limit should equal σ²∫₀^∞ S(s)ρρ*S(s)*ds, with error decaying as e^{-βt/2}. If this direct computation confirms the reduction in §6.1, the coefficient c2 in Corollary 2.7 follows; if an extra trace term appears, the expansion framework would need revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After a careful review I do not find a load-bearing flaw in the central argument. The main theorems are conditional on explicit hypotheses: global Lipschitz smoothness (HNL), the spectral gap (HTw), the normalization (HV*), and the covariance condition (Hq). Of these, (HTw) is the most load-bearing: the exponential semigroup bounds (3.5), the exponential convergence of the limits in Proposition 2.2, and the high-probability stability event all rely on the decay rate β = min{βtw, λ1/2}. However, this is an explicitly stated assumption, standard for pushed fronts; if it fails the results simply do not apply, rather than becoming internally inconsistent. I checked the construction of the expansion terms Yj, the moment bounds in Proposition 5.1, the conditioning estimate in Proposition 2.6 via equation (2.66), and the tail estimate in Theorem 2.4. These steps are mutually consistent. The one delicate point is the exchange of expectation with the infinite Itô correction sum in Lemma 6.1, but the Hilbert-Schmidt estimates (2.20) and the pairing bound (6.14) control that sum, so no gap is apparent. I therefore see no reason to challenge the reader's ACCEPT verdict.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript develops a rigorous small-noise expansion for planar traveling-wave solutions of stochastically perturbed reaction-diffusion systems on cylindrical domains. Using a stochastic freezing procedure, the perturbation relative to the deterministic wave is decomposed as V = Y1 + ... + Y_{r-1} + Z, where the Yj are defined recursively from the SPDE and Z is the residual. The main results are moment bounds for the Yj (Proposition 2.1), exponential convergence of expectations of multilinear expressions (Proposition 2.2), Taylor expansion of expectations of smooth functionals (Proposition 2.3), an exponentially long high-probability stability event (Theorem 2.4), and a conditional expansion for the observed wave speed (Corollary 2.7). The proof is organized into convolution estimates, smoothness of nonlinearities, moment bounds for expansion terms, limiting expectations, residual estimates, and the nonlinear stability argument. The paper advertises the expansion coefficients as universal, i.e. independent of the precise definition of the conditioning stability event, and provides an algorithmic route to compute them.","tokens_in":52777,"tokens_out":13482,"duration_ms":145323,"significance":"If the results are accepted, this is a substantial rigorous contribution to the stochastic reaction-diffusion literature. It turns the formal and numerical expansions of [11,31] into a theorem, gives explicit error bounds on exponentially long time scales, and provides a method to compute speed and shape corrections to arbitrary order under sufficient smoothness. The proof is structurally sound: the hypotheses (HNL), (HTw), (HV*), and (Hq) are explicit and standard for pushed-front stability, and the construction of the expansion terms is derived from the SPDE rather than fitted. The paper also improves on prior convolution estimates by allowing only moment bounds on integrands instead of uniform pathwise bounds. The main limitation is the heavy reliance on prior work, especially [11,16,30], for semigroup estimates, forward integrals, and the mild Itô formula; this is acceptable for a technical paper in this research line, but it means the reader must consult the companion papers to verify the foundational tools.","major_comments":[{"comment":"The proof of Corollary 2.7 states that one may apply Proposition 2.6 to the function v ↦ aσ(Φ0+v), but this function depends on σ through the explicit terms in (A.14). Proposition 2.6, as stated, guarantees an expansion with coefficients h∞;j that are constant only for a fixed functional φ independent of σ. If the coefficients are allowed to depend on σ, the conclusion that the c_j are fixed scalars with an O(σ^{r-1/2}) remainder does not follow directly. The proof should be repaired by decomposing aσ(Φ0+v) into σ-independent components, e.g. aσ = a0 + σ² a2, applying Proposition 2.6 to each component, and then recombining the corresponding polynomials; the terms beyond order r−1 are then absorbed into the O(σ^{r-1/2}) error. This is a local fix, but as written the advertised wave-speed expansion lacks a complete proof.","section":"§2.4, Corollary 2.7"}],"minor_comments":[{"comment":"The statement begins with the typo 'Suuppose' instead of 'Suppose'.","section":"Lemma 5.11"},{"comment":"There is a duplicated word in 'for any any 0 < σ ≤ δσ'; it should read 'for any 0 < σ ≤ δσ'.","section":"Corollary 2.5"},{"comment":"The sentence 'Note that the first half is exluded to avoid any transients' contains a spelling error; 'exluded' should be 'excluded'.","section":"§2.4, Eq. (2.73)"},{"comment":"The title contains the spacing error 'W ave'; it should be 'Wave'.","section":"Title page"},{"comment":"The exchange of the expectation with the infinite Itô correction sum is only sketched; given that the full justification appears in the cited mild Itô formula, a short sentence indicating that the sum is controlled by the Hilbert-Schmidt bound (2.20) and the pairing estimate (6.14) would improve readability.","section":"§6.1, Lemma 6.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong technical contribution and I do not see a fundamental flaw in the core stochastic analysis. My main concern is limited to the proof of Corollary 2.7, which is a central advertised application; the needed repair is elementary but should be stated explicitly. The heavy reliance on the companion paper [11] is understandable, but the authors should verify that all imported semigroup and forward-integral results are applied within their stated hypotheses. I would be willing to accept after the requested clarification and typographical corrections."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is the paper that finally proves what the same group's earlier work [11,31] only did formally and numerically. The central result: under a spectral gap hypothesis and smoothness assumptions, conditional on a high-probability stability event, smooth functionals of the frozen perturbation admit a Taylor expansion in the noise strength sigma, with coefficients that do not depend on the event. In particular, the observed wave speed has the expansion c0 + c2 sigma^2 + ... + c_{r-1} sigma^{r-1} with O(sigma^{r-1/2}) error. That is a genuine advance for stochastic reaction-diffusion wave theory.\n\nWhat I like: the expansion terms are defined recursively from the SPDE itself, not fitted. The moment bounds for the Y_j and the residual estimates are the real substance, and the proof structure is coherent. The forward-integral machinery and convolution bounds are appropriate. The paper is honest about what it relies on: the spectral gap (HTw), global Lipschitz smoothness (HNL), the normalization (HV*), and the covariance condition (Hq). I checked the steps that matter: Proposition 5.1, Lemma 6.1, the conditioning estimate (2.66), and Theorem 2.4. I do not see a load-bearing gap. The delicate exchange of expectation with the infinite Ito correction sum in Lemma 6.1 is controlled by the Hilbert-Schmidt bound (2.20) and the pairing bound (6.14).\n\nSoft spots, in proportion: the paper imports a fair amount of machinery from [11,16] rather than reproving it. That is standard in this area, but it means the verification burden is spread across several papers. The conditioning on Astb is slightly awkward—you get universal coefficients, but the statement concerns expectations conditioned on not having blown up by a sigma-dependent time. That is the natural statement for this problem, and the error terms are honest about it. The spectral gap assumption is load-bearing, but it is standard for pushed fronts and clearly stated. The paper is not machine-checked; at this length, that is not surprising.\n\nWho is this for: people working on stochastic stability of patterns, especially the freezing approach. It deserves citation. A serious editor should send it to review; I would referee it myself.\n\nRecommendation: accept, with the usual request that the authors flag explicitly which technical results are imported from earlier papers and which are new at each proof stage.","headline":"Rigorous proof of previously formal noise-induced wave-speed corrections; heavy but coherent, and the central expansion holds together under explicit standard hypotheses.","tokens_in":53356,"tokens_out":1749,"would_cite":true,"duration_ms":20618,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35K57","35R60"],"pacs":[],"model":"deepseek-v4-flash","headline":"Small-noise corrections to travelling wave speeds are rigorous, universal, and computable to any order.","keywords":["stochastic reaction-diffusion equations","travelling waves","stochastic freezing","small noise expansions","conditional expectations","wave speed corrections","metastability","forward integrals"],"falsifier":"Simulate a concrete bistable reaction-diffusion system with small translation-invariant noise at several values of $\\sigma$, keep only trajectories that stay near the shifted wave for the paper's time $T(\\sigma;\\theta_*)$, and compare the empirical conditional mean of the observed speed with $c_0+c_2\\sigma^2+c_3\\sigma^3+\\cdots$; a systematic gap larger than the paper's $K\\sigma^{r-1/2}$ error bound would refute Corollary 2.7.","tokens_in":52379,"feed_emoji":"🌊","tokens_out":11975,"duration_ms":107328,"temperature":0.7,"pith_summary":"The paper sets out to make rigorous a claim that earlier work only supported numerically: when a planar travelling wave in a reaction-diffusion system is hit by small translation-invariant noise, the observed speed and shape of the wave acquire corrections that can be expanded in powers of the noise strength, with the first speed correction at order $\\sigma^2$. The main theorem states that, conditional on the wave remaining stable for an exponentially long time, the expectation of any sufficiently smooth functional of the frozen perturbation has a Taylor expansion whose coefficients do not depend on the precise definition of the 'stability event'. This matters because stochastic corrections to front speeds appear throughout excitable media and chemical wave theory, and before this paper there was no rigorous way to define $\\mathbb{E}[C_{\\mathrm{obs}}]$ or to know which power of $\\sigma$ dominates. The proof proceeds by stochastic freezing—choosing a moving coordinate $\\Gamma(t)$ so that the perturbation stays small—then splitting the perturbation into ordered Taylor terms plus a residual that is controlled with high probability.","feed_headline":"Wave speed noise corrections are universal to every order","feed_subtitle":"Conditional on the wave surviving, the speed correction series is explicitly computable and independent of stability conventions.","key_machinery":"The central object is the stochastic freezing decomposition $u(x+\\Gamma(t),x_\\perp,t)=\\Phi_0(x)+V(x,x_\\perp,t)$, in which the phase $\\Gamma$ is chosen to keep the perturbation $V$ from drifting along the translational direction of the wave, so that $V$ stays bounded. The linearized operator $L_{\\mathrm{tw}}=\\partial_x^2+c_0\\partial_x+Df(\\Phi_0)$ carries the stability information: a simple zero eigenvalue from translation and a spectral gap $\\beta_{\\mathrm{tw}}>0$ ensure exponential decay once the neutral direction is projected out. The expansion writes $V=Y_1+\\cdots+Y_{r-1}+Z$, with $Y_j$ defined recursively through convolutions of the evolution family with deterministic and stochastic integrands; the stochastic convolutions are forward integrals because their integrands are anticipating. Two engine results do the work: the moment bounds $\\mathbb{E}\\sup_{0\\le t\\le T}\\|Y_j(t)\\|_{H^{k_j}}^{2p}\\le K^{2p}[(\\sigma^{2p})^{jp}+(\\sigma^2\\ln T+\\delta^2)^{jp}]$, and the convergence of multilinear expectations to explicit limits built from iterated semigroup integrals, which produces the universal coefficients. The stability event $A_{\\mathrm{stb}}=\\{\\|V^{(j)}\\|\\le\\sigma^{-1/2},\\|Z\\|\\le\\sigma^{r-1/2}\\text{ on }[0,T_*]\\}$ has probability close to one for $T_*$ exponentially long in $\\sigma^{-1}$, and conditioning on it perturbs expectations only by $O(\\sigma^{r-1/2})$.","core_discovery":"The central claim is that for every sufficiently smooth functional $\\varphi$ of the frozen perturbation $V(t)=U(\\cdot+\\Gamma(t),\\cdot,t)-\\Phi_0$, the conditional expectation given the stability event $A_{\\mathrm{stb}}$ admits the expansion $\\mathbb{E}[\\varphi(V(T_*))\\mid A_{\\mathrm{stb}}]=h_{\\infty;0}+\\sigma h_{\\infty;1}+\\cdots+\\sigma^{r-1}h_{\\infty;r-1}+O(\\sigma^{r-1/2})$, with coefficients that are independent of how the stability event is defined. In particular, the observed average speed satisfies $|\\mathbb{E}[C_{\\mathrm{obs}}\\mid A_{\\mathrm{stb}}]-(c_0+c_2\\sigma^2+\\cdots+c_{r-1}\\sigma^{r-1})|\\le 2K\\sigma^{r-1/2}$, so the first stochastic correction enters at order $\\sigma^2$. The paper proves that the expansion terms $Y_j$ are well-defined with explicit moment bounds, that multilinear limits $\\lim_{t\\to\\infty}\\mathbb{E}\\Lambda[Y_{i_1}(t),\\dots,Y_{i_\\ell}(t)]$ exist and equal explicit iterated semigroup integrals, and that the residual $Z$ stays small with high probability on exponentially long timescales.","pith_inferences":["The paper leaves implicit that the same universal coefficients should appear under different phase-pinning conventions; comparing two conventions numerically would provide a gauge-invariant test of the expansion.","The effective expansion parameter $\\sigma\\sqrt{\\ln T}$ means corrections grow logarithmically in observation time; the paper's estimates stop at $\\ln T_*\\ll\\sigma^{-2}$, so longer-time behaviour would need a different argument.","Because the only structural inputs are a one-dimensional translational neutral mode and a spectral gap, the machinery should transfer to other pattern-forming systems such as spiral waves or cellular fronts, although the paper does not carry out such applications.","A concrete next step would be to compute the higher-order coefficients $c_3,c_4,\\dots$ for a standard bistable model and benchmark them against direct simulations at moderate $\\sigma$, giving a quantitative check of the metastability picture."],"forward_implications":["For any smooth observable $\\varphi$ of the frozen perturbation, $\\mathbb{E}[\\varphi(V(T_*))\\mid A_{\\mathrm{stb}}]$ equals a universal polynomial in $\\sigma$ up to an error $O(\\sigma^{r-1/2})$ (Proposition 2.6).","The observed average wave speed satisfies $|\\mathbb{E}[C_{\\mathrm{obs}}\\mid A_{\\mathrm{stb}}]-(c_0+c_2\\sigma^2+\\cdots+c_{r-1}\\sigma^{r-1})|\\le 2K\\sigma^{r-1/2}$, identifying $\\sigma^2$ as the leading noise correction (Corollary 2.7).","The coefficients $h_{\\infty;j}$ and $c_j$ are algorithmically computable as iterated semigroup integrals, so in principle the speed and shape corrections can be obtained to any order, limited only by the smoothness of $f$ and $g$.","The decomposition $V=Y_1+\\cdots+Y_{r-1}+Z$ holds with probability at least $1-2\\exp(-\\tfrac12\\mu\\sigma^{-2\\theta/r})$ on times exponentially long in $\\sigma^{-1}$, supplying a rigorous metastability statement for stochastic travelling waves (Corollary 2.5)."],"supporting_citations":[{"why":"The preceding multidimensional stability framework; supplies the stochastic freezing phase process and the moment estimates that this paper refines.","marker":"[11]"},{"why":"Establishes the exponentially long timescale stability approach and the supremum bounds for stochastic convolutions used here.","marker":"[30]"},{"why":"Gives sharp supremum and Hölder bounds for stochastic integrals indexed by a parameter, used for the $p^p+[\\ln T]^p$ scaling.","marker":"[15]"},{"why":"Introduces the stochastic freezing and time-transformation machinery for multiplicative-noise travelling waves, including the mild residual representation.","marker":"[28]"},{"why":"Derives the explicit $\\sigma^2$ speed correction and numerical validation that this paper generalizes to arbitrary order.","marker":"[31]"},{"why":"Supplies the mild Itô formula used to reduce multilinear expectations to iterated semigroup integrals.","marker":"[16]"},{"why":"Defines forward integrals for anticipating integrands, the tool needed for the stochastic convolutions in the expansion.","marker":"[41]"},{"why":"Provides the critical variational framework used for local well-posedness and stopping times in the residual analysis.","marker":"[5]"},{"why":"Introduces the deterministic freezing principle that the stochastic phase method adapts.","marker":"[8]"}],"fun_headline_variants":["Wave speed noise corrections: universal to every order","Stochastic wave speed: explicit corrections to any order","Noisy waves: speed correction series is universal","Rigorous noise expansions for traveling wave speed","Universal speed corrections for stochastic reaction-diffusion waves"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the wave is stable in every direction except pure side-to-side translation, with all other small disturbances decaying at a fixed exponential rate; without that spectral gap the semigroup estimates and the entire expansion framework collapse.","fun_headline_variants_meta":{"raw":{"variants":["Wave speed noise corrections: universal to every order","Stochastic wave speed: explicit corrections to any order","Noisy waves: speed correction series is universal","Rigorous noise expansions for traveling wave speed","Universal speed corrections for stochastic reaction-diffusion waves"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00075,"raw_usage":{"total_tokens":3324,"prompt_tokens":914,"completion_tokens":2410,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":530,"completion_tokens_details":{"reasoning_tokens":2351}},"tokens_in":530,"tokens_out":2410,"duration_ms":17191,"temperature":1.0,"reasoning_tokens":2351,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T13:05:09.246682+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate a concrete bistable reaction-diffusion system with small translation-invariant noise at several values of $\\sigma$, keep only trajectories that stay near the shifted wave for the paper's time $T(\\sigma;\\theta_*)$, and compare the empirical conditional mean of the observed speed with $c_0+c_2\\sigma^2+c_3\\sigma^3+\\cdots$; a systematic gap larger than the paper's $K\\sigma^{r-1/2}$ error bound would refute Corollary 2.7.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the exponentially long timescale stability approach and the supremum bounds for stochastic convolutions used here."},{"cited_title":"Sharp supremum and H\\\"older bounds for stochastic integrals indexed by a parameter","cited_arxiv_id":"2409.13615","evidence_quote":"Gives sharp supremum and Hölder bounds for stochastic integrals indexed by a parameter, used for the $p^p+[\\ln T]^p$ scaling."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the stochastic freezing and time-transformation machinery for multiplicative-noise travelling waves, including the mild residual representation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Derives the explicit $\\sigma^2$ speed correction and numerical validation that this paper generalizes to arbitrary order."},{"cited_title":"Da Prato, A","cited_arxiv_id":null,"evidence_quote":"Supplies the mild Itô formula used to reduce multilinear expectations to iterated semigroup integrals."},{"cited_title":"Russo and P","cited_arxiv_id":null,"evidence_quote":"Defines forward integrals for anticipating integrands, the tool needed for the stochastic convolutions in the expansion."},{"cited_title":"Agresti and M","cited_arxiv_id":null,"evidence_quote":"Provides the critical variational framework used for local well-posedness and stopping times in the residual analysis."},{"cited_title":"Beyn and V","cited_arxiv_id":null,"evidence_quote":"Introduces the deterministic freezing principle that the stochastic phase method adapts."}],"review_version":1}