{"id":"aed756a6-95a2-442e-907c-4ce7196fc0c2","arxiv_id":"2611.17000","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Smooth solutions to the 3D homogeneous Landau–Coulomb equation are instantaneously bounded in L∞(R3) by a constant depending only on the initial conserved quantities and entropy.","lead":"This paper proves that solutions to the Landau–Coulomb equation, a model of plasma collisions, become bounded in velocity space at positive times, with the bound depending only on the initial mass, momentum, energy, and entropy. This resolves an open problem in kinetic theory by showing that no higher velocity moments or additional regularity are needed to control the solution's maximum.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek/deepseek-v4-flash","headline":"Theorem 1.1 is an a priori estimate; the claim that it resolves GGL's open problem requires the bound to transfer to solutions from minimal-regularity data, a step deferred in §4.2 and never supplied.","rationale":"I read the paper in good faith and hand-checked the most delicate quantitative steps I could reach: the representation identity (6.2) in Proposition 6.1 (signs check out after expansion), the score and Hessian remainder bounds P and Q in §3, the integration-by-parts identity (4.4) closing the Riccati inequality (the Cauchy-Schwarz gives exactly (3J_4 ∫⟨v⟩²f)^{1/2} with ∫⟨v⟩²f = 4), the localized cell Nash estimate (6.6) from (6.3) and the anisotropic Sobolev embedding, and the radial scaling with critical exponent α = 3/2 in §7.2. All of these are internally consistent. I also re-examined the reader's flagged 'heuristic' lower ellipticity bound (8.3) and found the argument complete: the negative-entropy control gives ∫ f log_+ f ≤ C_H, the threshold K_0 with C_H/log K_0 ≤ 1/4 bounds ∫_{f>K_0} f, the surviving cylinder mass is 9/16, and the slab measure estimate yields the claimed ⟨v⟩^{-3}Π_v + ⟨v⟩^{-1}Π⊥_v anisotropy. The genuinely load-bearing gap is elsewhere: Theorem 1.1 is honestly scoped as an a priori estimate, and the paper states this plainly in [b22]. But the abstract and §1.2 claim this 'resolves the open problem' for the solution classes constructed by GGL from L^{3/2} data and by Guillen–Silvestre from finite-Fisher data. Between the a priori estimate and those solutions sits an approximation and trace-stability argument that §4.2 performs only for the unweighted Fisher inequality (4.5) via Fatou's lemma, and never for the nonlinear weighted tail iteration feeding (8.14). The tail masses Y_n, the threshold K_0, the base radius R, and the geometric base γ_0 could in principle depend on qualitative properties (Gaussian tails, positive lower bounds, cutoff radii) of the approximating sequence that are not controlled by H, τ, T. Since I could not rule this out by inspection, and since the paper itself identifies the transfer as open ([b22]), the verdict CONDITIONAL stays: the a priori theorem appears correct, but its advertised reach exceeds what is demonstrated. My agreement with the reader is partial because we share the a priori/existence concern as primary, while the reader's additional worry about (8.3) is, in my reading, already resolved by the text.","tokens_in":29865,"tokens_out":26754,"duration_ms":140904,"concrete_test":"Perform the §4.2 regularization for a one-parameter family f_{0,ε} of smooth, rapidly decaying data with fixed mass, momentum, energy, and entropy bound H, converging as ε → 0 to a minimal-regularity datum with only finite mass, energy, and entropy. Track the three quantities entering the tail De Giorgi iteration: the threshold K_0(H) in (8.3); the base radius R(H,τ,T) needed so that Y_0,Y_1 ≤ ε_c γ_0^{-6} in (8.10); and the geometric base γ_0 in (8.14). The transfer passes if K_0, R, γ_0 stay uniformly bounded as ε → 0 so the bound f ≤ 2K_0 survives the limit; the open-problem resolution fails if any diverges.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1.1 ([b20]) is stated and proved only for smooth, rapidly decaying solutions, and the authors explicitly defer the approximation and initial-trace stability argument that would extend the bound to solutions obtained from general initial data with finite mass, momentum, energy, and entropy ([b22], end of §1.1). Because the result is advertised as resolving the open problem posed by GGL [5], whose global smooth solutions are constructed from L^{3/2} data and need not be rapidly decaying, the load-bearing point is whether the macroscopic bound C(H,τ,T) actually controls the L^∞ norm of those solutions. The error terms entering the tail iteration are not shown to be stable under the approximation used inside §4.2: that section passes only the unweighted Fisher inequality (4.5) to the limit via Fatou's lemma, but the nonlinear weighted truncated-entropy iteration of §8 (the recurrence (8.13)–(8.14) on the tail masses Y_n, involving the level-set-truncated dissipation D_{p,k}, the low-density ellipticity (8.3) at threshold K_0, and the weighted Nash inequality (7.3) with its deficit terms δ^{-2}M_V and δ^{-2}A_0M_V) is never shown to be semi-continuous under the approximation that would produce a solution from data with only finite mass/energy/entropy. If K_0(H), R(H,τ,T), or the geometric base γ_0 depend on qualitative features of the approximating sequence (the ε_j Gaussian tails or the cutoff χ(v/j) in the §4.2 regularization), the claimed 'solely macroscopic' dependence would fail for the actual solution class. This is not a defect of the a priori theorem as scoped, but it is exactly where the central claim—the open-problem resolution—is least secure. I note that the reader's secondary worry about (8.3) appears unfounded: the mass-concentration argument for (8.3) is complete as written (C_H/log K_0 ≤ 1/4 via the negative-part control from §4, the cylinder measure bound on B_4, and mass ≥ 9/16); my concern is not there.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek/deepseek-v4-flash","summary":"The paper claims to prove an instantaneous global L∞ bound for smooth, rapidly decaying solutions to the three-dimensional homogeneous Landau–Coulomb equation, with the bound depending only on the initial mass, momentum, energy, and entropy. This is presented as resolving an open problem posed by Golding, Gualdani, and Loher. The proof combines a weighted Fisher information estimate with a truncated mixed Nash inequality and a De Giorgi iteration on the velocity tail.","tokens_in":33156,"tokens_out":2068,"duration_ms":32549,"significance":"If the result were established for all smooth solutions (or for solutions from minimal-regularity data), it would be a major advance in the regularity theory of the Landau–Coulomb equation, answering a question that has received significant attention. The techniques introduced, particularly the use of the negatively weighted Fisher information and the anisotropic mixed Nash inequality, are novel and likely to be influential. However, the paper's result is proved only for a restricted class of solutions (smooth and rapidly decaying), and the advertised resolution of the open problem is not fully achieved within the manuscript.","major_comments":[{"comment":"The paper claims to resolve the open problem posed by Golding, Gualdani, and Loher [5]. However, Theorem 1.1 is proved only for smooth, rapidly decaying solutions, while the open problem concerns smooth solutions (which may not be rapidly decaying). The paper acknowledges in §1.1 that upgrading to general initial data requires an additional approximation and initial-trace stability argument, which is not provided. Consequently, the claim of resolving the open problem is not supported by the results presented. The authors should either supply the missing argument or revise the statements to accurately reflect that Theorem 1.1 is an a priori estimate for a restricted class of solutions, leaving the full resolution of the open problem as a future step.","section":"§1.1 and Abstract","authors_acknowledge":"Upgrading this a priori estimate to a global existence theorem for general initial data with finite mass, energy, and entropy requires an additional approximation and initial-trace stability argument, which we do not pursue here.","authors_acknowledge_p":0.93},{"comment":"The approximation and cutoff justification in §4.2 establishes the weighted Fisher information estimates (Theorem 1.2) for smooth, rapidly decaying solutions via a limiting argument from regularized initial data. However, the tail De Giorgi iteration in §8, which is essential for the proof of Theorem 1.1, is not subjected to a similar stability analysis. The quantities appearing in the recurrence (8.13)–(8.14) (weighted truncated entropy, dissipation functionals, and the mixed Nash inequality estimates) are defined directly for the given smooth, rapidly decaying solution, and their behavior under the approximation that would produce such a solution from general initial data with only finite mass, energy, and entropy is not examined. If Theorem 1.1 is to be extended beyond the class of smooth, rapidly decaying solutions, the stability of the tail iteration under the approximation must be addressed. As it stands, the proof of Theorem 1.1 is valid only for solutions that are already smooth and rapidly decaying, which is a significant restriction relative to the advertised scope.","section":"§8 and §4.2","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The derivation of the lifted Fisher identity could be clarified by explicitly stating the assumptions on the weight matrix M and the solution required for the integration by parts.","section":"§2, Eq. (2.4)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The justification of the unweighted dissipation bound (4.5) relies on the existence of global smooth solutions with Gaussian bounds from [7]. The paper should cite the specific theorem or appendix in [7] that guarantees this.","section":"§4.2","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The step involving equation (6.2) could benefit from a more detailed explanation of how the weak identity Fσ = ∇g is used to obtain the representation.","section":"§6, proof of Proposition 6.1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The constant b0 is introduced as a universal geometric base (e.g., b0 = 16). The paper should verify that the induction argument works for this specific value and that the choice is independent of the solution.","section":"§8","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The disclosure of AI assistance is commendable but unusual for a mathematics journal. The authors should ensure this complies with the journal's policy.","section":"General","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Typographical issues: In the abstract, 'la{v}' should be '⟨v⟩'. In the introduction, 'GGL' is used without definition; it is defined later in §1.2.","section":"General","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"major_revision"},"author_rebuttal":null,"desk_editor":{"model":"deepseek/deepseek-v4-flash","letter":"Lingda Xu has written a serious paper. The main result—an L∞ bound for the Landau–Coulomb equation that depends only on mass, momentum, energy, and entropy—is exactly the open problem Golding, Gualdani, and Loher posed. The proof is a genuine piece of analysis: the weighted Fisher information generation at the second-moment threshold (Theorem 1.2) is clever, and the truncated mixed Nash inequality combined with the De Giorgi iteration in the tail is a substantial technical achievement. The paper is well-organized and the estimates are carried out with care.\n\nThe soft spot is the one the authors flag themselves: Theorem 1.1 is proved only for smooth, rapidly decaying solutions. The paper says explicitly that upgrading it to general initial data requires an additional approximation and initial-trace stability argument, which is not supplied. The stress-test note worries that the nonlinear weighted iteration of §8 might not be semicontinuous under the approximation that would produce a solution from minimal-regularity data. I think that concern is legitimate but not a fatal flaw of the paper as written—the authors are upfront about the limitation, and the a priori estimate is the hard part. The missing step is a separate project, and the paper does not claim to have done it.\n\nI also note that the lower ellipticity bound (8.3) is argued heuristically but the argument is complete enough for the purpose; the reader's worry about it seems misplaced. The citation pattern is clean; the paper builds on Guillen–Silvestre and Ji as black boxes, which is appropriate.\n\nWho is this for? Anyone working on regularity for the Landau equation or kinetic equations with singular kernels. It deserves a serious referee. The paper should be published, with the understanding that the a priori estimate is the contribution and the extension to general data remains open.\n\nRecommendation: accept with minor revisions. The authors should add a remark clarifying that the approximation argument is not provided and that the result is conditional on the existence of smooth, rapidly decaying solutions. They might also discuss whether the De Giorgi iteration constants depend on qualitative features of the approximating sequence, as the stress-test note suggests. But these are clarifications, not corrections.","headline":"A technically impressive a priori estimate that resolves the open problem under the assumption of smooth, rapidly decaying solutions; the gap to general initial data is real but clearly scoped.","tokens_in":31020,"tokens_out":550,"duration_ms":7011},"pith_extraction":{"msc":["35Q20","82C40","35B65"],"pacs":[],"model":"deepseek/deepseek-v4-flash","headline":"The Landau–Coulomb equation's $L^\\infty$ bound depends only on mass, energy, and entropy.","keywords":["Landau–Coulomb equation","instantaneous $L^\\infty$ regularization","weighted Fisher information","entropy dissipation","mixed Nash inequality","De Giorgi iteration"],"falsifier":"If one could construct a sequence of smooth, rapidly decaying solutions to the Landau–Coulomb equation with uniformly bounded initial mass, momentum, energy, and entropy, but with $\\|f_n(t)\\|_{L^\\infty(\\mathbb{R}^3)}$ unbounded for some fixed $t>0$, then Theorem 1.1 would be false. Alternatively, if the constant $C(H,\\tau,T)$ in Theorem 1.1 could be shown to depend on a higher moment (e.g., the fourth moment) in any essential way, the claim that the bound depends only on the conserved quantities and entropy would be refuted.","tokens_in":33592,"feed_emoji":"","tokens_out":2782,"duration_ms":30298,"temperature":0.7,"pith_summary":"The paper proves that any smooth, rapidly decaying solution to the three-dimensional Landau–Coulomb equation is instantaneously bounded in $L^\\infty(\\mathbb{R}^3)$ for positive times, with the bound depending only on the initial mass, momentum, energy, and entropy. This resolves an open problem posed by Golding, Gualdani, and Loher, showing that no higher velocity moments, Fisher information, or Lebesgue integrability are needed to control the velocity tail. The proof combines two main innovations: (1) a weighted Fisher identity on the two-particle phase space that closes a Riccati inequality at the second-moment threshold using a comparison with a collision-invariant pair weight, and (2) a truncated anisotropic mixed Nash inequality that recovers longitudinal regularity from two transverse Fisher directions and the nonlinear high-density part of the Coulomb entropy dissipation, enabling a superlinear De Giorgi recurrence in the velocity tail. The result is a purely macroscopic a priori estimate; upgrading it to a global existence theorem for general initial data with only finite mass, energy, and entropy would require an additional approximation and stability argument that the paper does not provide.","feed_headline":"One bound rules the Landau–Coulomb velocity tail","feed_subtitle":"Proof that $L^\\infty$ depends only on mass, energy, and entropy, resolving a 2025 open problem.","key_machinery":"Two main analytical tools carry the proof. First, on the lifted two-particle phase space $\\mathbb{R}^6$, the paper compares the separable weight matrix $S(v,w)=\\operatorname{diag}(\\langle v\\rangle^{-1}I_3,\\langle w\\rangle^{-1}I_3)$ with the collision-invariant pair weight $\\Omega(v,w)=(1+(|v|^2+|w|^2)/2)^{-1/2}$. Because $\\Omega$ is constant on collision spheres, Ji's sharp $\\Gamma_2$ inequality on the real projective sphere provides an $11/9$ coercivity ratio between the spherical dissipation $\\mathcal{D}_\\Omega$ and the curvature integral $\\mathcal{R}_\\Omega$, and the ratio $\\omega(v)/\\Omega(v,w)\\ge 1/\\sqrt{2}$ leaves a strictly positive coercive margin $\\kappa_0=\\sqrt{2}-9/11>0$. The commutator from the perturbation $T=S-\\Omega I_6$ is controlled near the collision diagonal by reversing one rotational integration by parts and using quadratic local mass decay from the conserved energy, turning the Hessian error into $C\\delta J_4(f)$ without invoking higher moments. Second, to overcome the $\\langle v\\rangle^{-3}$ radial degeneracy of the diffusion matrix at large velocities, the paper proves a truncated anisotropic mixed Nash inequality on upper level sets of the density: $\\int_{\\mathbb{R}^3} h^{5/3}\\,dv\\le C(\\mu^{2/3}X+\\mu^{1/2}X^{1/2}D^{1/3})$, where $h=(\\sqrt{f}-\\sqrt{K})_+^2$, $\\mu$ is the mass of the upper level set, $X$ is the transverse Fisher energy, and $D$ is the high-density Coulomb entropy dissipation. This inequality recovers longitudinal regularity from two transverse Fisher directions and the nonlinear entropy dissipation, bypassing the degenerate radial eigenvalue entirely.","core_discovery":"The central claim is Theorem 1.1: for every entropy upper bound $H\\in\\mathbb{R}$ and any $0<\\tau<T$, there exists a constant $C(H,\\tau,T)$ such that every smooth, rapidly decaying solution $f$ of the homogeneous Landau–Coulomb equation (with normalized mass, momentum, and energy) satisfies $\\sup_{\\tau\\le t\\le T} \\|f(t)\\|_{L^\\infty(\\mathbb{R}^3)}\\le C(H,\\tau,T)$. This is the first $L^\\infty$ bound that depends only on the physical conservation laws and the initial entropy, without any higher moments, Fisher information, or $L^p$ integrability assumptions. The paper thus resolves a problem raised in the introduction of Golding, Gualdani, and Loher (2025).","pith_inferences":["The paper's reliance on the existence of smooth, rapidly decaying solutions with the given normalization (1.3) means that the result is an a priori estimate, not a global existence theorem. A natural next step is to construct such solutions for initial data with only finite mass, energy, and entropy, possibly via a suitable approximation and stability argument that the paper sketches but does not ","The critical weight exponent $\\alpha=3/2$ in the exterior De Giorgi iteration (Proposition 7.2) is exactly the threshold that balances the $\\langle v\\rangle^{-1}$ transverse metric factor with the Sobolev embedding constants, suggesting that this exponent is sharp for the method and that any attempt to prove $L^\\infty$ bounds with only second moments must hit this exponent.","The commutator estimate near the collision diagonal (Section 3.1) uses the quadratic local mass decay $\\int_{\\{r\\le 2\\delta\\rho\\}} f(w)\\,dw\\le C\\langle v\\rangle^{-2}$ derived from the conserved energy. This technique may be transferable to other kinetic equations with singular collision kernels where the conserved energy provides enough control of nearby particles.","The paper leaves open whether the $t^{-1}$ rate in the bound for $I_{-1}(f)$ is sharp; the introduction mentions this as a separate open problem from Golding, Gualdani, and Loher (2025). A numerical or analytical investigation of the optimal rate could refine the smoothing estimate."],"forward_implications":["If the paper is correct, any smooth, rapidly decaying solution to the Landau–Coulomb equation is globally bounded in $L^\\infty$ for any positive time, with the bound controlled solely by the initial mass, energy, and entropy, confirming that no additional regularity or moment assumptions are needed for $L^\\infty$ regularization.","The instantaneous generation of the negatively weighted Fisher information $I_{-1}(f)$ at the $t^{-1}$ rate (Theorem 1.2) becomes a quantitative a priori estimate that can be used as a building block for further regularity theory, such as higher-order derivative bounds or Schauder estimates.","The truncated anisotropic mixed Nash inequality (Proposition 6.1) provides a new functional inequality for kinetic equations with anisotropic diffusion, potentially applicable to other soft-potential or degenerate-diffusion settings beyond Landau–Coulomb.","The superlinear De Giorgi recurrence in the velocity tail (Section 8) shows that the tail entropy can be driven to zero using only the conserved quantities, establishing a purely macroscopic control of the large-velocity region without moment closures."],"supporting_citations":[{"why":"Provides the lifted two-particle framework and the monotonicity of the unweighted Fisher information, which is the starting point for the weighted Fisher identity.","marker":"[7]"},{"why":"Posed the open problem that the paper resolves and established $L^\\infty$ estimates down to the critical $L^{3/2}$ threshold, setting the baseline for the current work.","marker":"[5]"},{"why":"Supplies the sharp $\\Gamma_2$ inequality on the real projective sphere that gives the $11/9$ coercivity ratio between spherical dissipation and curvature integral, essential for Lemma 2.2.","marker":"[8]"},{"why":"Proved a dissipation bound for the unweighted Fisher information under a fifth moment assumption, which the paper improves by working with $I_{-1}$ and $J_4$ at the second-moment threshold.","marker":"[9]"},{"why":"Provides the conditional parabolic regularization theorem used to justify the uniform $L^\\infty$ bound for the approximating sequence in Section 4.2.","marker":"[10]"},{"why":"Establishes uniqueness of bounded solutions to the Landau–Coulomb equation, used to identify the limit of the approximating sequence with the given solution.","marker":"[4]"}],"fun_headline_variants":["Landau–Coulomb tail tamed by entropy alone","No extra moments needed for Landau–Coulomb L∞ bound","Velocity tail uniformly bounded via entropy and conserved quantities","Instantaneous L∞ bound for Landau–Coulomb from basic invariants","Weighted Fisher info opens Landau–Coulomb tail bound"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The proof assumes the existence of smooth, rapidly decaying solutions with the given normalization (1.3) and that the initial entropy is bounded above by $H$; it does not prove that such solutions exist for all initial data with only finite mass, energy, and entropy.","fun_headline_variants_meta":{"raw":{"variants":["Landau–Coulomb tail tamed by entropy alone","No extra moments needed for Landau–Coulomb L∞ bound","Velocity tail uniformly bounded via entropy and conserved quantities","Instantaneous L∞ bound for Landau–Coulomb from basic invariants","Weighted Fisher info opens Landau–Coulomb tail bound"]},"model":"deepseek/deepseek-v4-flash","effort":"low","cost_usd":9.4e-05,"raw_usage":{"cost":9.41122e-05,"is_byok":false,"cost_details":{"upstream_inference_cost":9.41122e-05,"upstream_inference_prompt_cost":7.82782e-05,"upstream_inference_completions_cost":1.5834e-05},"total_tokens":1152,"prompt_tokens":1065,"completion_tokens":87,"prompt_tokens_details":{"audio_tokens":0,"video_tokens":0,"cached_tokens":256,"cache_write_tokens":0},"completion_tokens_details":{"audio_tokens":0,"image_tokens":0,"reasoning_tokens":0}},"tokens_in":1065,"tokens_out":87,"cost_source":"openrouter_usage","duration_ms":2229,"image_pages":[],"temperature":1.0,"billed_attempts":1,"reasoning_tokens":0,"transport_provider":"openrouter","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-13T11:17:01.790092+00:00","model_set":{"reader":"deepseek/deepseek-v4-flash"},"pipeline_version":"pith-pipeline@v1.4.1-daily-deepseek-or-medium","input_kind":null,"input_chars":null,"input_total_chars":null,"passes_missing":null,"published_standing":null,"published_confidence":null,"objection_audit":null,"run_record":null,"basis":{"kind":"not_recorded","source":null,"chars_given":null,"chars_total":null},"receipt":null,"falsifier":"If one could construct a sequence of smooth, rapidly decaying solutions to the Landau–Coulomb equation with uniformly bounded initial mass, momentum, energy, and entropy, but with $\\|f_n(t)\\|_{L^\\infty(\\mathbb{R}^3)}$ unbounded for some fixed $t>0$, then Theorem 1.1 would be false. Alternatively, if the constant $C(H,\\tau,T)$ in Theorem 1.1 could be shown to depend on a higher moment (e.g., the fourth moment) in any essential way, the claim that the bound depends only on the conserved quantities and entropy would be refuted.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the lifted two-particle framework and the monotonicity of the unweighted Fisher information, which is the starting point for the weighted Fisher identity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Posed the open problem that the paper resolves and established $L^\\infty$ estimates down to the critical $L^{3/2}$ threshold, setting the baseline for the current work."},{"cited_title":"Bounds for the optimal constant of the Bakry-Émery $Γ_2$ criterion inequality on $ RP^{d-1}$","cited_arxiv_id":"2408.13954","evidence_quote":"Supplies the sharp $\\Gamma_2$ inequality on the real projective sphere that gives the $11/9$ coercivity ratio between spherical dissipation and curvature integral, essential for Lemma 2.2."},{"cited_title":"Dissipation estimates of the Fisher information for the Landau equation","cited_arxiv_id":"2410.09035","evidence_quote":"Proved a dissipation bound for the unweighted Fisher information under a fifth moment assumption, which the paper improves by working with $I_{-1}$ and $J_4$ at the second-moment threshold."},{"cited_title":"Silvestre,Upper bounds for parabolic equations and the Landau equation, J","cited_arxiv_id":null,"evidence_quote":"Provides the conditional parabolic regularization theorem used to justify the uniform $L^\\infty$ bound for the approximating sequence in Section 4.2."},{"cited_title":"Fournier,Uniqueness of bounded solutions for the homogeneous Landau equation with a Coulomb potential, Comm","cited_arxiv_id":null,"evidence_quote":"Establishes uniqueness of bounded solutions to the Landau–Coulomb equation, used to identify the limit of the approximating sequence with the given solution."}],"lane_accuracy":{"status":"not yet measured","lane":{"model":"deepseek/deepseek-v4-flash","input_kind":null},"url":"https://pith.science/accuracy"},"review_version":1}