{"id":"4abfd4f1-ed63-4000-b9f1-20bb101e8b5b","arxiv_id":"1908.04879","paper_version":2,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"Existence and uniqueness of invariant measures are claimed for anisotropic degenerate parabolic-hyperbolic conservation laws driven by additive white noise, extending Debussche-Vovelle's first-order theory.","lead":"This paper proves existence and uniqueness of invariant measures for degenerate parabolic-hyperbolic conservation laws with additive white noise, extending a known first-order result to second-order anisotropic equations. It also surveys related stochastic PDE results, making it a useful reference for the long-time statistics of randomly forced transport models.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 5.2 Step 1 asserts a uniform L2 mollification bound for L1 initial data that is false; Theorem 5.1's uniqueness proof therefore lacks a key input.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing flaw: Lemma 5.2 Step 1 claims a quantitative L2 regularization of arbitrary L1 functions that is not true in general. I verified the counterexample: for a function concentrated on a set of volume δ^d, L1-distance approximation forces the approximant to retain mass on that set, forcing an L2 norm of order δ^{-d/2}, which can be arbitrarily larger than ε^{-d/2}. This is not a stylistic gap but a false assertion, and the subsequent Step 8 estimate depends on the asserted L2 bound. Since Theorem 5.1's uniqueness conclusion is the paper's central new claim, the result is not established as written. The existence theorem (Theorem 4.1) may be salvageable, but the uniqueness proof requires a genuinely different argument. I agree with the REJECT verdict and would not weaken it to CONDITIONAL without a substantial revision of Lemma 5.2.","tokens_in":34152,"tokens_out":3431,"duration_ms":33994,"concrete_test":"Set d = 1, κ = 1/2, ε = 10^{-6}, δ = 10^{-12}, and define u0(x) = δ^{-1} on [0, δ] and 0 elsewhere on the torus, so ||u0||_{L1} = 1 ≤ 2κ. Analytically compute inf_{f: ||u0 - f||_{L1} ≤ ε/8} ||f||_{L2} ≥ (1 - ε/8)δ^{-1/2} ≈ 10^6. Compare with the claimed bound C κ ε^{-d/2} = C * 10^3; for the explicit C in Lemma 5.2 the inequality fails. This directly disproves Step 1 of Lemma 5.2 and shows the proof of Theorem 5.1 cannot be patched by adjusting constants alone.","verdict_should_be":"REJECT","load_bearing_attack":"The uniqueness claim (Theorem 5.1) rests on Lemma 5.2, whose Step 1 requires: for every u0 with ||u0||_{L1} ≤ 2κ, there exists a mollified approximation u0_tilde with ||u0 - u0_tilde||_{L1} ≤ ε/8 and ||u0_tilde||_{L2} ≤ C κ ε^{-d/2}. This is false. Take u0 = δ^{-d} 1_{|x|<δ} on T^d with δ << ε, so ||u0||_{L1} = 1. For any f with ||u0 - f||_{L1} ≤ ε/8, the mass of f on the small ball is at least 1 - ε/8, so ||f||_{L2} ≥ (1 - ε/8)δ^{-d/2}. For δ sufficiently small relative to ε, this exceeds the claimed C κ ε^{-d/2}; no such approximation exists. This false bound is used in Step 8 of Lemma 5.2 to control the kinetic measure: |m_v + N_v| ≤ C e^{Cκt}(||u0_tilde||_{L2}^2 + 1) ≤ C e^{Cκt}(κ^2 ε^{-d} + 1). That estimate feeds into the M1 estimate (5.16) and into the final averaging bound in Step 9. Without the uniform L2 control of regularized initial data, the recurrence argument collapses, and the coupling proof of uniqueness in §5.3 has no valid basis.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an existence and uniqueness theory of invariant measures for stochastic anisotropic degenerate parabolic-hyperbolic conservation laws of second order driven by additive spatial white noise, equation (4.1). Section 4 develops an existence argument via the kinetic formulation, a regularized mild formulation, a four-term decomposition into u0, u♭, M1, M2, kernel estimates, and compactness in Sobolev spaces, followed by an invocation of the Krylov-Bogoliubov machinery. Section 5 aims to prove uniqueness through a coupling/recurrence argument: Lemma 5.1 provides almost-sure finite-time entry into a ball, Lemma 5.2 gives temporal L1 averaging bounds under small noise, and Section 5.3 concludes by combining L1 contraction with a Borel-Cantelli argument. The paper also contains a broad survey of invariant measures for stochastic conservation laws, Burgers equation, Navier-Stokes equations, the KPP equation, and large deviations.","tokens_in":34470,"tokens_out":5969,"duration_ms":60524,"significance":"If established, Theorems 4.1 and 5.1 would extend the invariant-measure results of Debussche-Vovelle [35] from first-order stochastic balance laws to second-order degenerate parabolic-hyperbolic equations, a nontrivial and interesting step. The architectural ideas in Section 4, including the regularized semigroup, the four-term decomposition, and the use of kernel estimates to handle the stochastic and kinetic-measure terms, are potentially reusable and clearly presented. The survey part is broad and useful. However, the uniqueness proof in Theorem 5.1 rests on a false uniform mollification estimate in Lemma 5.2, and the existence proof in Section 4 never verifies the Feller property that the invoked Krylov-Bogoliubov theorem requires. These are load-bearing defects, so the central claims are not established in the present manuscript.","major_comments":[{"comment":"Step 1 claims that for every u0 with ||u0||_{L1} ≤ 2κ̂ there exists a mollified approximation ˜u0 satisfying ||u0 - ˜u0||_{L1} ≤ ε/8 and ||˜u0||_{L2} ≤ C κ̂ ε^{-d/2}. This statement is false. Take u0 = δ^{-d} 1_{B_δ} on T^d with ||u0||_{L1}=1 and δ << ε. For any f with ||u0 - f||_{L1} ≤ ε/8, the mass of f on B_δ is at least 1 - ε/8, so ||f||_{L2} ≥ (1 - ε/8)δ^{-d/2}. For δ sufficiently small this exceeds the claimed C κ̂ ε^{-d/2}. Thus no such uniform bound can hold. This estimate is essential in Step 8, where it produces the bound |m_v + N_v| ≤ C e^{Cκ̃t}(||˜u0||_{L2}^2 + 1) ≤ C e^{Cκ̃t}(κ̂^2 ε^{-d} + 1), which feeds directly into (5.16) and the final averaging estimate in Step 9. The recurrence and uniqueness proof in §5.3 therefore collapse.","section":"§5.2, Lemma 5.2, Step 1"},{"comment":"The proof of Theorem 4.1 invokes the Krylov-Bogoliubov mechanism of §2.2 without verifying the Feller property of the solution semigroup. Theorem 2.1 explicitly requires that P_s maps C(X) into C(X); the tightened compactness estimate (4.21) alone is insufficient for the argument as written. The paper mentions in §2.2 a replacement technique from [21] that avoids the Feller condition, but that technique is not developed or applied here. Consequently, Theorem 4.1 is incomplete as stated.","section":"§4.5, Completion of the existence proof"},{"comment":"The almost-sure L1 contraction and well-posedness stated in Lemma 5.3 are imported from the companion paper [12] without proof. The final uniqueness argument in §5.3 uses this contraction to transfer averaging estimates from the regularized solutions ˜u1, ˜u2 to the original solutions u1, u2, so the statement is load-bearing. The manuscript should either reproduce the proof or explicitly state that Theorem 5.1 is conditional on the results of [12]; as it stands, the dependency is not transparent to the reader.","section":"§5.2, Lemma 5.3"}],"minor_comments":[{"comment":"In the definition D0 := ||σ^2||_{L∞(T)}, the domain should presumably be T^d, not T.","section":"§4.3, Lemma 4.1"},{"comment":"In the paragraph beginning 'For α < 1/4', the symbol C(ρ,θ) appears where C4(γ,θ) seems intended; this is a typographical error.","section":"§5.2, Step 9"},{"comment":"The notation for the averaged integral is introduced as 'the symbol ffl', but the displayed formulas in Lemma 5.2 use a different typesetting; the intended meaning is clear but the typography should be unified.","section":"§5.2, Step 1"},{"comment":"The definition of the Sobolev regularity exponent (1-α)κ + α in (4.15) is not explicitly tied to the embedding constants later; a brief justification of the exponent's range would improve readability.","section":"§4.1"},{"comment":"In the kinetic formulation (3.6), the term ∂_ξ(m_u + n_u - p_u) is introduced formally; the footnote-like explanation of the Itô correction p_u is helpful but would benefit from a precise reference to [12] for the rigorous limiting procedure.","section":"§3.2"}],"recommendation":"reject","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know. First, the paper is really two papers: a survey of invariant measures for stochastic conservation laws, and an original proof of existence and uniqueness for anisotropic degenerate parabolic-hyperbolic equations with additive noise. The survey is competent and the existence theorem looks like a genuine extension of Debussche-Vovelle. Second, the uniqueness theorem has a genuine load-bearing gap, and the stress-test note is right.\n\nWhat's new and good: Section 4 builds the kinetic solution framework, adds regularizing operators γ(−Δ)^α + θI, splits the mild solution into four terms, and uses kernel estimates plus the non-degeneracy condition (4.2) to get compactness. The estimates seem plausible and the architecture is sound. The existence proof doesn't explicitly verify the Feller property, but L1 contraction from the companion paper [12] would give it in one line; that's a minor omission, not a fatal one.\n\nThe soft spot: Lemma 5.2, Step 1 (page 23) claims that every L1 function with ||u0||₁ ≤ 2κ has a mollified approximation with L1 error ≤ ε/8 and L2 norm ≤ Cκ ε^{-d/2}. That's false. Take u0 = δ^{-d} on a set of volume δ^d with δ << ε; ||u0||₁ = 1, and any f with ||u0−f||₁ ≤ ε/8 must retain most of its mass near that spike, forcing ||f||₂ ≥ (1−ε/8)δ^{-d/2}, which blows up as δ→0. The claimed uniform bound of order ε^{-d/2} cannot hold. This bound feeds directly into the Gronwall estimate for the kinetic measure, then into the M1 estimate (5.16), and finally into the averaging argument that drives the recurrence. Without it, the uniqueness proof of Theorem 5.1 loses its control. This is not a small technicality; it is the main step distinguishing the uniqueness argument from the first-order case.\n\nThe citation pattern is mostly fine. The paper leans on the authors' own [12] for well-posedness and L1 contraction, but that is a separate companion paper, not a circular assumption of the result. The survey portions are useful and cite the relevant literature.\n\nNet: as it stands, the uniqueness claim is not established. The existence theorem may be salvageable and is worth refereeing, but the paper needs major revision. I would send it to a serious referee with instructions to focus on Lemma 5.2, and I would flag that the uniqueness part cannot be accepted without a different argument—probably replacing the L2 mollification bound with an L1-based estimate or adding a smallness condition on the initial data.\n\nRecommendation: engage, but require revision; do not reject on the grounds that the topic is uninteresting.","headline":"The existence theorem is a credible extension of Debussche-Vovelle, but the uniqueness proof breaks on a false mollification bound in Lemma 5.2.","tokens_in":34996,"tokens_out":3685,"would_cite":false,"duration_ms":37806,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35B40","35K65","37-02","37A50","37C40","60H15"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper establishes existence and uniqueness of an invariant probability measure for anisotropic degenerate parabolic-hyperbolic conservation laws driven by additive white noise.","keywords":["invariant measures","stochastic conservation laws","anisotropic degenerate parabolic-hyperbolic equations","kinetic formulation","additive white noise","Krylov-Bogoliubov method","coupling method","long-time behavior"],"falsifier":"Take initial data $u_0$ that concentrate unit mass in a tiny subset of volume $\\delta^d$ with $\\delta$ much smaller than $\\varepsilon$, and compute the minimum $L^2$ norm among all functions within $L^1$ distance $\\varepsilon/8$ of $u_0$; that minimum is about $\\delta^{-d/2}$, which far exceeds the claimed $C\\hat\\kappa\\,\\varepsilon^{-d/2}$ bound and would invalidate Step 1 of Lemma 5.2.","tokens_in":33952,"feed_emoji":"🎲","tokens_out":6893,"duration_ms":70366,"temperature":0.7,"pith_summary":"The paper's central claim is that a second-order conservation law with degenerate diffusion, nonlinear flux, and additive white-in-time noise settles into a unique statistical steady state: there is exactly one invariant probability measure describing its long-run behavior. The equation studied is the anisotropic degenerate parabolic-hyperbolic balance law $\\partial_t u + \\nabla\\cdot F(u) = \\nabla\\cdot(A(u)\\nabla u) + \\sigma(x)\\partial_t W$ on the torus, with the noise having zero spatial average. Existence of an invariant measure is proven under a joint nonlinearity-diffusivity condition, and uniqueness is proven under the additional condition that $F''$ and $A'$ are bounded. A sympathetic reader cares because this turns the long-time behavior of stochastic conservation laws, not just deterministic ones, into a well-defined statistical question.","feed_headline":"White noise gives conservation laws a unique steady state","feed_subtitle":"Under a nonlinearity condition, solutions forget their initial data and settle into one invariant probability measure.","key_machinery":"The central object is the kinetic formulation of the equation, in which the indicator function $\\chi(\\xi,u)$ is transported in phase space, so the nonlinear flux becomes a linear first-order operator in $\\xi$ and $x$ at the cost of adding kinetic measures that carry the dissipation. The non-degeneracy condition (4.2) quantifies how much velocity averaging regularizes oscillations in $\\xi$, and the regularizing operators $\\gamma(-\\Delta)^\\alpha + \\theta I$ inserted into the kinetic equation make the semigroup estimates possible. These pieces together yield the compactness in $W^{s,q}$ needed for the Krylov-Bogoliubov existence argument, while the $L^1$-contraction property of the original equation is what lets the coupling stopping times force uniqueness.","core_discovery":"On the paper's own terms: for the stochastic anisotropic degenerate parabolic-hyperbolic equation with additive noise $\\sigma(x)\\partial_t W$, the solution semigroup admits at least one invariant probability measure, and under a boundedness condition this measure is unique. The proof treats the unknown through its kinetic formulation, decomposes the solution into deterministic, martingale, and Itô-correction parts, and uses regularizing operators to obtain compactness estimates that feed into the Krylov-Bogoliubov argument. Uniqueness is obtained by a coupling argument: two solutions starting from different data enter the same small ball infinitely often with probability one, and the $L^1$-contraction property of the equation then forces their $L^1$ distance to tend to zero almost surely.","pith_inferences":["Editorial: The uniqueness proof as written relies on Step 1 of Lemma 5.2 being able to replace any $L^1$ datum of size at most $2\\hat\\kappa$ by a mollified datum close in $L^1$ with $L^2$ norm $O(\\varepsilon^{-d/2})$; a concentrated initial datum violates this bound, leaving a gap in the proof of uniqueness as written.","Editorial: Because $L^1$ contraction is the only pathwise stability input, the same argument cannot mechanically extend to multiplicative noise; the natural test case is noise with a root, where the root is a fixed point and one should check whether more than one invariant measure appears.","Editorial: The recurrence probability depends only on a Brownian increment staying small in $W^{1,\\infty}$ over a fixed time window, so any noise with the same positive small-ball probability would likely inherit the coupling argument."],"forward_implications":["If the theorems are right, any two $L^1$ initial data give solutions whose $L^1$ distance tends to zero almost surely, so the initial state is asymptotically forgotten.","The unique invariant measure is ergodic, so statistics along a single long trajectory reproduce the measure's long-run statistics.","The existence theorem covers fully degenerate hyperbolic cases, not only uniformly parabolic equations, as long as the nonlinearity-diffusivity condition (4.2) holds.","Taking $A=0$ recovers the first-order scalar balance law case with additive spatially dependent noise, with uniqueness under bounded $F''$.","The zero-spatial-average assumption on $\\sigma$ is used throughout, so the result is tied to noise that does not directly inject mass into the conserved quantity."],"supporting_citations":[{"why":"Supplies the invariant-measure strategy and the kinetic decomposition that this paper extends to the second-order degenerate equation.","marker":"[35]"},{"why":"Provides the well-posedness theory of stochastic kinetic solutions on which the invariant-measure argument is built.","marker":"[12]"},{"why":"Gives the chain-rule relation that identifies parabolic defect measures inside the kinetic formulation.","marker":"[13]"},{"why":"Provides the averaging lemma used for the compactness estimates on the deterministic solution components.","marker":"[6]"},{"why":"Motivates the nonlinearity-diffusivity condition (4.2) and the compactness mechanism in the deterministic large-time setting.","marker":"[14]"},{"why":"Establishes kinetic solutions to scalar conservation laws with stochastic forcing, the framework reused here.","marker":"[34]"}],"fun_headline_variants":["Noise pins down a unique steady state for conservation laws","Conservation laws with white noise have one invariant measure","Unique invariant measure for noisy conservation laws","White noise forces a unique long-time measure for conservation laws","Stochastic conservation laws settle into a single invariant measure"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that every $L^1$ initial datum inside the recurrence ball can be mollified into a nearby function whose $L^2$ norm is bounded by a constant times the radius times $\\varepsilon^{-d/2}$; the uniqueness theorem's recurrence estimate collapses if this approximation bound cannot be enforced.","fun_headline_variants_meta":{"raw":{"variants":["Noise pins down a unique steady state for conservation laws","Conservation laws with white noise have one invariant measure","Unique invariant measure for noisy conservation laws","White noise forces a unique long-time measure for conservation laws","Stochastic conservation laws settle into a single invariant measure"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000971,"raw_usage":{"total_tokens":4019,"prompt_tokens":726,"completion_tokens":3293,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":342,"completion_tokens_details":{"reasoning_tokens":3218}},"tokens_in":342,"tokens_out":3293,"duration_ms":25086,"temperature":1.0,"reasoning_tokens":3218,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:30:55.231712+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take initial data $u_0$ that concentrate unit mass in a tiny subset of volume $\\delta^d$ with $\\delta$ much smaller than $\\varepsilon$, and compute the minimum $L^2$ norm among all functions within $L^1$ distance $\\varepsilon/8$ of $u_0$; that minimum is about $\\delta^{-d/2}$, which far exceeds the claimed $C\\hat\\kappa\\,\\varepsilon^{-d/2}$ bound and would invalidate Step 1 of Lemma 5.2.","supporting_citations":[{"cited_title":"Debussche and J","cited_arxiv_id":null,"evidence_quote":"Supplies the invariant-measure strategy and the kinetic decomposition that this paper extends to the second-order degenerate equation."},{"cited_title":"Nonlinear Anisotropic Degenerate Parabolic-Hyperbolic Equations with Stochastic Forcing","cited_arxiv_id":"1903.02693","evidence_quote":"Provides the well-posedness theory of stochastic kinetic solutions on which the invariant-measure argument is built."},{"cited_title":"Debussche and J","cited_arxiv_id":null,"evidence_quote":"Establishes kinetic solutions to scalar conservation laws with stochastic forcing, the framework reused here."}],"review_version":1}