{"id":"650dcd1c-ef9b-43ab-933a-60b9572d96ab","arxiv_id":"2506.08394","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Randomly forced resistive magnetic relaxation yields, in the zero-resistivity limit, random MHS equilibria; in 2D the limit measure has zero mass on finite Fourier mode equilibria.","lead":"This paper proves that a randomly forced, slightly resistive version of the magnetic relaxation equations converges, as resistivity vanishes, to a random equilibrium state whose statistics are explicitly controlled. The authors also show in two dimensions that these random equilibria are essentially never finite Fourier mode solutions, so the equilibrium measure is infinite-dimensional.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 2D non-concentration theorem rests entirely on the non-degeneracy hypothesis b_j≠0 for all j; with degenerate noise the diffusion estimates in Lemmas 6.11 and 6.17 fail and the absolute-continuity argument cannot rule out finite-dimensional support.","rationale":"The reader's weakest_assumption correctly identifies the non-degeneracy condition as the load-bearing input for Theorem 1.3. I examined the alternative candidate, the passage in (5.27) of E||Bκ||^2_H to the limit; this is a genuine technical gap in Theorem 5.15, but it is not load-bearing for the central existence theorem: it can be closed using the uniform exponential moment estimate (5.2) plus weak convergence in H^{1-ε}, since the L^2-norm is continuous on H^{1-ε} and the exponential bound gives uniform integrability. It affects the nontriviality remark, not the existence/support claim. The non-degeneracy assumption, by contrast, is the engine of the non-concentration theorem; with finitely many forced modes the diffusion rank collapses and the Krylov-estimate argument fails. Therefore the reader's verdict CONDITIONAL stands, and my stress-test does not move it.","tokens_in":59665,"tokens_out":23866,"duration_ms":267613,"concrete_test":"Fix an integer N≥1 and set b_j=0 for j>N (keeping C0>0). For this truncated noise, compute the diffusion matrix σ(f'(φ)) in Lemma 6.17: its rank is at most N for every φ, so for any n>N, det σ≡0 and the estimate (6.24) gives no bound. This analytically verifies that the non-degeneracy assumption is indispensable: with N active modes the proof of Theorem 6.16 cannot establish absolute continuity of any Casimir law of dimension n>N, so Theorem 1.3's conclusion for compact sets of Hausdorff dimension ≥N is not obtained. A numerical check could also simulate the N-mode-truncated system and observe the stationary measure supported on the finite-dimensional span of the active eigenfunctions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the random MHS equilibrium is almost surely not a finite Fourier mode solution (Theorem 1.3) depends on b_j≠0 for every mode j. This assumption enters twice: Lemma 6.11 needs ς=C_{-1}-sup_j b_j^2/λ_j>0 for the small-ball estimate (6.13), and Lemma 6.17 needs det σ(f'(φ))>0 for φ≠0, which uses completeness of the basis {√λ_j d_j} together with b_j≠0 to conclude a·f'(φ)=0. If only N modes are forced, the matrix σ(g)=Σ_{j=1}^N b_j^2 (g_k,d_j)(g_l,d_j) has rank at most N for every g; for any Casimir map C of dimension n>N, det σ≡0. Then the Krylov-estimate step (6.24) becomes vacuous, and Theorem 6.16 cannot give absolute continuity of Dμ0(C). Since the proof of Theorem 1.3 chooses n>dim_H K, the method cannot exclude concentration on compact sets of Hausdorff dimension ≥N. The theorem is honestly stated under non-degeneracy, but this is the most fragile input: if the noise is degenerate, the paper provides no route to μ0(FFFM)=0, and the headline non-concentration result is unsupported. The reader astutely identified this assumption; the secondary issue in (5.27) about passing E||Bκ||^2_H to E||B||^2_H is fixable via the uniform exponential moment bound (5.2) and does not affect Theorem 1.1.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the randomly forced resistive magnetic relaxation equations (1.3) on the flat torus T^d, with resistivity κ and a noise √κ∂tζ built from eigenfunctions e_j with amplitudes b_j. It proves pathwise global well-posedness (Theorem 3.8), existence of invariant measures μκ via Krylov–Bogoliubov (Lemma 4.8), and a non-resistive limit: after passing to a subsequence, statistically stationary solutions Bκ converge almost surely in C([0,T];H^{-ε})∩L^2(0,T;H^{1-ε}) to a time-independent random field B∈H∩H^1 which satisfies the MHS equilibrium equations (1.2), with law D(B)=μ0 (Theorem 1.1). Under the additional assumption that b_j≠0 for every j, the paper proves for d=2 that μ0 gives zero mass to every compact subset of H^1 of finite Hausdorff dimension, and in particular μ0(F_FFM)=0 (Theorem 1.3). The tools include Itô formulas, balance relations for energy/helicity/mean-square potential, compactness in Bochner spaces, Skorokhod's theorem, and Krylov's estimates for stationary processes.","tokens_in":59961,"tokens_out":12673,"duration_ms":158349,"significance":"If correct, the paper provides a rigorous statistical construction of MHS equilibria by a fluctuation-dissipation limit, extending Kuksin's method for 2D Euler to magnetic relaxation. The 2D non-concentration theorem is a strong structural statement about the limiting measure: it excludes support on finite Fourier mode equilibria, and the proof via Casimir invariants and absolute continuity is original in this context. Strengths include the self-contained treatment of well-posedness with cubic estimates, the explicit balance relations (5.1)–(5.4), the compactness argument for lifted invariant measures, and the honest statement of hypotheses. The main limitation is the non-degeneracy assumption on the noise for Theorem 1.3; the theorem is stated with that assumption, but the abstract and introduction would benefit from making this condition equally prominent.","major_comments":[{"comment":"Theorem 1.3 and the 2D statement in the abstract are proved only under the full non-degeneracy hypothesis b_j≠0 for all j∈N. This hypothesis enters essentially in Lemma 6.11 through positivity of ς=C_{-1}-sup_j b_j^2/λ_j and the small-ball estimate (6.13), and in Lemma 6.17 through the conclusion det σ(f'(φ))>0, which uses completeness of {√λ_j d_j} together with b_j≠0. If only finitely many modes are forced, the covariance matrix σ has rank at most N, the Krylov step (6.24) degenerates, and the argument gives no information about concentration on finite-dimensional compact sets. The theorem is honestly stated in §1.2.2, but the abstract presents the non-concentration result without the hypothesis; the authors should state the assumption in the abstract and explicitly flag the degenerate-noise case as open.","section":"Abstract / Theorem 1.3"}],"minor_comments":[{"comment":"The Skorokhod construction is written for a fixed T, while Theorem 1.1 asserts convergence in C([0,∞);H^{-ε})∩L^2_loc([0,∞);H^{1-ε}). A diagonal argument in T and consistency of the laws across different T should be stated explicitly.","section":"Lemma 5.12 / Theorem 1.1"},{"comment":"The equality E||B||^2_H=C_{-1}/2 for d=2 requires passing the second moment E||Bκ||^2_H to the limit. This does not follow from convergence in P(H^{1-ε}) alone; it is recoverable from the uniform exponential bound (5.2), which gives uniform integrability of ||Bκ||^2_H, but that argument is not written. The same remark applies to (5.26).","section":"Theorem 5.15, Eq. (5.27)"},{"comment":"In the proof of Theorem 1.1, the convergence (5.24) is stated as Bκ⊗Bκ→B⊗B in L^1(T^d×(0,T)) almost surely; this follows from the a.s. convergence Bκ→B in L^2(0,T;H) and the uniform bound on Bκ in L^2(0,T;H^1), but the intermediate step using the boundedness from (5.6) should be mentioned.","section":"Section 5.5 / Proposition 5.14"},{"comment":"There are many minor typographical errors, including 'satifies' in Proposition 5.2, 'elemenrary caluculation' and 'caluculation' in Appendix A, and the phrase '0 < Γ' in the proof of Theorem 6.1, which should be '0∉Γ'. A careful proofreading pass is needed.","section":"Various"}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"My take on arXiv:2506.08394: the fluctuation-dissipation machinery from Kuksin–Shirikyan is imported, but the target is genuinely new. The paper constructs random MHS equilibria as the non-resistive limit of randomly forced resistive MRE, proving H^1 regularity in d≥2 and, for d=2, that the limit measure gives zero mass to finite-Fourier-mode equilibria. The proof is long but detailed and honest: pathwise well-posedness, invariant measures via Krylov–Bogoliubov, tightness through the KT compact space, and the limit passage in the constitutive law are all written out. Balance relations and exponential moment bounds are used carefully. I found no circular reasoning; μ0 is a weak limit of invariant measures and its properties are derived from stationarity and Itô calculus, not from a target distribution. The constants C0, C_{-1}, C_{-1/2} are noise inputs, not fitted parameters.\n\nThe main soft spot is exactly what the reader flagged: the non-degeneracy hypothesis b_j ≠ 0 for all j. It drives Lemma 6.11 and Lemma 6.17, which power the absolute-continuity argument behind Theorem 1.3. With finitely many forced modes, the covariance matrix σ can be singular and the method gives no route to μ0(F_FFM)=0. That is not a hidden flaw—the theorem is honestly stated under non-degeneracy—but it is the most fragile input. If the result is meant to be robust, this is where to push. The stress-test note confirms this and I agree with it.\n\nThe secondary worry about (5.27) is minor. The H-norm is not continuous on H^{1-ε}, so the displayed equality needs justification, but the exponential moment bound (5.2) provides uniform integrability and stationarity plus the strong convergence in L²_loc should close it. That is a fixable gap, not a load-bearing one. The d=3 helicity absolute continuity is a nice bonus, and the paper is appropriately careful about what changes for d≥4.\n\nWho is this for? People working on stochastic PDEs and magnetic relaxation. It gives a rigorous construction and an explicit extension of the Kuksin–Shirikyan method beyond 2D turbulence. I would cite it for the 2D non-concentration result and the construction. Deserves a serious referee: it is substantial and, conditional on the small fix, correct.","headline":"A substantial and largely correct stochastic-construction paper whose headline 2D non-concentration result rests on a stated but fragile non-degeneracy assumption.","tokens_in":60571,"tokens_out":2239,"would_cite":true,"duration_ms":29233,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q31","35Q35"],"pacs":[],"model":"deepseek-v4-flash","headline":"Randomly forced resistive magnetic relaxation has a non-resistive limit: a random MHS equilibrium; in 2D the limiting law avoids all finite Fourier modes.","keywords":["magnetic relaxation equations","magnetohydrostatic equilibrium","invariant measure","non-resistive limit","stochastic partial differential equations","finite Fourier modes","Hausdorff dimension","random MHD equilibria"],"falsifier":"Take the 2D system with a degenerate Wiener process in which exactly one coefficient $b_j$ is zero and all others are nonzero. If the corresponding limit measure $\\mu_0$ assigns positive mass to a compact subset of $H^1$ with finite Hausdorff dimension, for example the span of the unstirred eigenfunction, then Theorem 1.3 is false; the paper's lemmas locate the exact step where this degeneracy would break the argument.","tokens_in":59401,"feed_emoji":"🧲","tokens_out":17117,"duration_ms":175313,"temperature":0.7,"pith_summary":"Randomly forcing the resistive magnetic relaxation equations with noise of strength $\\sqrt{\\kappa}$ gives pathwise global well-posedness and invariant measures $\\mu_\\kappa$ for every resistivity $\\kappa>0$. The paper's main theorem says that as $\\kappa\\to 0$ these statistically stationary solutions converge almost surely to a time-independent magnetohydrostatic (MHS) equilibrium $B\\in H^1(\\mathbb{T}^d)$ with law $\\mu_0$, so the random relaxation procedure constructs MHS equilibria from general initial data rather than from data with special symmetries. In two dimensions the limiting measure is infinite-dimensional in a strong sense: $\\mu_0$ assigns zero mass to every compact subset of $H^1(\\mathbb{T}^2)$ with finite Hausdorff dimension, and in particular almost every realization is not a finite Fourier mode solution. This matters because the statistical route reaches equilibria beyond the simple finite-mode states whose structure is explicitly classified.","feed_headline":"Stirred magnetic fields relax to random equilibria as resistivity dies","feed_subtitle":"In 2D the limit measure avoids finite Fourier-mode fields; the random equilibria are genuinely infinite-dimensional.","key_machinery":"The load-bearing mechanism is a statistical energy balance for stationary solutions: every invariant measure of (1.3) satisfies $E(\\kappa\\|\\nabla B_\\kappa\\|_H^2+\\|u_\\kappa\\|_{\\dot H^\\gamma}^2)=\\kappa C_0/2$ together with the exponential bound $E\\exp(\\rho\\|B_\\kappa\\|_H^2)\\le (C_0+1)e^{\\rho(C_0+1)}$. After dividing by $\\kappa$ this forces the velocity field $u_\\kappa$ to vanish like $\\sqrt{\\kappa}$ while $B_\\kappa$ stays bounded, so the constitutive law $\\nabla p_\\kappa=B_\\kappa\\cdot\\nabla B_\\kappa-(-\\Delta)^\\gamma u_\\kappa$ closes in the limit to $\\nabla p=B\\cdot\\nabla B$. The limit passage itself rests on tightness of the laws of the stationary solutions in the path space $X_T^{-\\varepsilon}=C([0,T];H^{-\\varepsilon})\\cap L^2(0,T;H^{1-\\varepsilon})$, obtained by decomposing $B_\\kappa$ into three pieces with different temporal regularities and using compactness of the relevant Bochner-space embedding, and on realizing the limiting law on a common probability space so the convergence is almost sure. For the 2D theorem, the additional machinery is absolute continuity: stochastic calculus identities for the stationary processes $E(B)$, $M(B)$, and Casimir functionals $C(B)$, combined with a lower-bound estimate on the quadratic variation matrix $\\sum_j b_j^2(f'(\\phi),d_j)^2$, whose non-singularity for nonconstant $\\phi$ follows from $b_j\\neq 0$ for all $j$. Absolute continuity of these scalar laws under $\\mu_0$ is what rules out concentration on finite-Hausdorff-dimension compact sets.","core_discovery":"On its own terms, the paper establishes Theorem 1.1: for $d\\ge 2$, $\\gamma>d/2$, and noise with finite $C_0$, the system (1.3) has invariant measures $\\mu_\\kappa$. Along a subsequence $\\kappa\\to 0$ the measures converge weakly to $\\mu_0$ on $H^{1-\\varepsilon}(\\mathbb{T}^d)$, and statistically stationary solutions $B_\\kappa$ with law $\\mu_\\kappa$ converge almost surely, on a suitable probability space, to a random field $B$ in $C([0,\\infty);H^{-\\varepsilon})\\cap L^2_{\\mathrm{loc}}([0,\\infty);H^{1-\\varepsilon})$. The limit $B$ is time-independent, lies in $H\\cap H^1(\\mathbb{T}^d)$, and satisfies $\\nabla p=B\\cdot\\nabla B$ with $\\nabla\\cdot B=0$ almost surely, so it is an MHS equilibrium with law $D(B)=\\mu_0$. Theorem 1.3 adds the 2D conclusion: with $b_j\\neq 0$ for every $j$, $\\mu_0$ gives zero mass to every compact subset of $H\\cap H^1(\\mathbb{T}^2)$ with finite Hausdorff dimension, hence $\\mu_0(F_{\\mathrm{FFM}})=0$ for the set $F_{\\mathrm{FFM}}$ of finite Fourier mode MHS equilibria. The paper also records explicit mean identities for the limit measure in low dimensions, $E\\|B\\|_H^2=C_{-1}/2$ in $d=2$ and $E(\\nabla\\times B,B)_H=C_{-1/2}/2$ in $d=3$, and extends the construction to hyper-resistivity in Theorem 5.16.","pith_inferences":["A numerical test on $\\mathbb{T}^2$ with small $\\kappa$ and forcing supported on many modes should show that the sampled fields' Fourier spectra do not collapse to finitely many modes, with the mean-square potential approaching $C_{-1}/2$; this would directly confirm the 2D conclusion.","If one of the forcing coefficients $b_j$ is set to zero, the covariance matrix can become singular for nonconstant $\\phi$, so the 2D conclusion may fail or become open; degenerate forcing is a natural place to seek counterexamples.","In 3D the paper leaves the finite-Fourier-mode trichotomy unresolved; the absolute continuity of the helicity law suggests $\\mu_0$ may also avoid finite Fourier mode equilibria there, but proving it would require handling the different geometry of 3D finite Fourier equilibria.","The same $\\sqrt{\\kappa}$ scaling of noise and dissipation may apply to other relaxation-type PDEs with a convex conserved quantity, making the invariant-measure route a general equilibrium-construction scheme."],"forward_implications":["For every $d\\ge 2$ and $\\gamma>d/2$ the construction yields a probability measure $\\mu_0$ whose realizations are $H^1$-regular MHS equilibria, so the random relaxation procedure generates equilibria from arbitrary initial data rather than only from specially chosen data.","The velocity field of the statistically stationary solutions vanishes as $\\kappa\\to 0$ at the rate $\\sqrt{\\kappa}$ in $L^2(0,T;\\dot H^\\gamma)$, while the magnetic field remains bounded and converges almost surely in the stated spaces.","In two dimensions, $\\mu_0$ gives zero mass to every compact subset of $H^1(\\mathbb{T}^2)$ with finite Hausdorff dimension; in particular, $\\mu_0(F_{\\mathrm{FFM}})=0$, so almost every realization is not a finite Fourier mode solution.","For $d=2$ and $d=3$ the limit measure has explicit mean values, $E\\|B\\|_H^2=C_{-1}/2$ and $E(\\nabla\\times B,B)_H=C_{-1/2}/2$, respectively; no analogous formulas are obtained for $d\\ge 4$.","The same fluctuation-dissipation argument extends to hyper-resistivity with $(-\\kappa)(-\\Delta)^\\alpha$ and yields a random MHS equilibrium in $H^\\alpha(\\mathbb{T}^d)$ for every $\\alpha\\ge 1$."],"supporting_citations":[{"why":"Introduces magnetic relaxation as a route to MHS equilibria, the object the paper constructs statistically.","marker":"[Mof85]"},{"why":"Establishes global well-posedness of the deterministic MRE and decay of the velocity field, the base system the random forcing perturbs.","marker":"[BFV22]"},{"why":"Gives deterministic global well-posedness for the 2D resistive relaxation base system, cited for the unforced case.","marker":"[MRR14]"},{"why":"Gives deterministic global well-posedness for the 3D resistive relaxation base system, cited for the unforced case.","marker":"[JT21]"},{"why":"Supplies the fluctuation-dissipation template of taking an inviscid-style limit of invariant measures for stochastic hydrodynamics.","marker":"[Kuk04]"},{"why":"Supplies the invariant-measure toolkit used throughout: existence of stationary measures, exponential moments, stationary-process identities, and absolute-continuity estimates.","marker":"[KS12]"},{"why":"Provides the compactness theorem for Bochner spaces used to prove tightness of the lifted measures.","marker":"[Sim87]"},{"why":"Constructs complex plane Beltrami waves used to build the orthonormal basis for the 3D noise.","marker":"[DLS13]"},{"why":"Supplies Beltrami-wave constructions for the orthonormal system in the 3D rotation-operator basis.","marker":"[BV19]"},{"why":"Characterizes finite Fourier mode MHS/Euler equilibria, the set excluded by the 2D conclusion.","marker":"[EHˇS17]"}],"fun_headline_variants":["Random forcing yields infinite-dimensional magnetic equilibria","2D random magnetic equilibria avoid finite Fourier modes","Resistivity dies, random magnetic fields relax to infinite-dimensional equilibria","Non-resistive limit gives random MHS equilibria, 2D is infinite-dimensional","Stirred magnetic relaxation: limit equilibria are not finite Fourier"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The 2D conclusion needs the noise to stir every Fourier mode ($b_j\\neq 0$ for all $j$); if even one mode is unstirred, the covariance that drives the absolute-continuity proof can become singular and the infinite-dimensionality conclusion collapses.","fun_headline_variants_meta":{"raw":{"variants":["Random forcing yields infinite-dimensional magnetic equilibria","2D random magnetic equilibria avoid finite Fourier modes","Resistivity dies, random magnetic fields relax to infinite-dimensional equilibria","Non-resistive limit gives random MHS equilibria, 2D is infinite-dimensional","Stirred magnetic relaxation: limit equilibria are not finite Fourier"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000282,"raw_usage":{"total_tokens":1744,"prompt_tokens":1095,"completion_tokens":649,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":711,"completion_tokens_details":{"reasoning_tokens":560}},"tokens_in":711,"tokens_out":649,"duration_ms":8010,"temperature":1.0,"reasoning_tokens":560,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T05:16:13.311712+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the 2D system with a degenerate Wiener process in which exactly one coefficient $b_j$ is zero and all others are nonzero. If the corresponding limit measure $\\mu_0$ assigns positive mass to a compact subset of $H^1$ with finite Hausdorff dimension, for example the span of the unstirred eigenfunction, then Theorem 1.3 is false; the paper's lemmas locate the exact step where this degeneracy would break the argument.","supporting_citations":[],"review_version":1}