{"id":"c7802661-1b6c-4664-8686-50dd309e9c40","arxiv_id":"2508.00714","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new a priori estimate for supercritical Navier-Stokes weak solutions yields local separation-rate bounds and Hölder time regularity at a singular time.","lead":"This paper builds a theory of weak solutions to the 3D Navier-Stokes equations for supercritical initial data in the Lorentz space Lp,∞ with 2 < p < 3, proving a quantified energy decay bound for the nonlinear part of the flow. The bound yields a local short-time asymptotic expansion with an upper bound on how quickly hypothetical non-unique solutions can separate, plus a new time-regularity estimate at a possible singular time.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 2.2's local energy inequality for w = u - V is asserted without proof, and its validity under the modified Definition 1.2 (which drops condition (1.4)) is unverified; Theorem 1.4 and both applications rest on it.","rationale":"The central claim of the paper is Theorem 1.4, the a priori L2-decay estimate for u - e^{t Delta} u0 in the supercritical class Lp,infinity, 2 < p < 3. That theorem is proved via Theorem 2.3, which in turn relies on Lemma 2.2. Lemma 2.2 is the only place where the local energy inequality for the perturbation w is established, and its proof is explicitly omitted ('this is an easy calculation and is omitted'), with the details deferred to [9, Lemma 3.3]. The paper's modified Definition 1.2 drops condition (1.4) that appears in the analogous L3,infinity definition from [9], and the authors argue this is harmless because the dropped term is not needed for L2-decay. However, the proof of Lemma 2.2 does not verify that [9, Lemma 3.3] remains valid without (1.4); the local energy inequality for w is used to justify the passage from a formal energy identity to the rigorous energy inequality, and this passage involves controlling pressure terms and cross terms involving V in the limit of test functions approaching 1. If those terms cannot be controlled using only the reduced Definition 1.2, then Theorem 1.4 is not proven, and the applications in Theorems 1.8 and 1.10 collapse because they use Theorem 1.4 as their engine. The reader identified this same weak point ('the local energy inequality for the perturbation w = u - V is asserted via an omitted calculation'), and I agree that it is the weakest load-bearing assumption. The concern does not by itself disprove the theorem; it identifies a missing verification. Therefore the reader's CONDITIONAL verdict is appropriate and should remain unchanged, provided the authors supply the omitted calculation or a precise adaptation of [9, Lemma 3.3]. The concrete test proposed here - writing out the limiting argument for the local energy inequality under Definition 1.2 - would settle whether the omitted step is indeed routine or whether it requires the full force of condition (1.4).","tokens_in":35573,"tokens_out":12218,"duration_ms":125675,"concrete_test":"Independently derive the local energy inequality for w = u - V under Definition 1.2, writing out the passage to the limit in (1.6) with test functions phi_R -> 1, and check that every term, in particular integral integral p_u u . grad(phi_R) and integral integral p_V V . grad(phi_R), vanishes as R -> infinity using only (1.7), V in L_infty(0,T;L^4), and w0 in L^2. If a term cannot be shown to vanish without condition (1.4), then Theorem 1.4 is not established in the modified class. As a secondary check, recompute N^{-(p-2)} in the proof of Corollary 2.4; the displayed exponent on ||u0||_{Lp,infinity} appears to have the wrong sign.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Lemma 2.2, the foundation of Theorem 1.4, asserts that the perturbation w = u - V satisfies a local energy inequality, with the proof deferred to the phrase 'this is an easy calculation and is omitted' and to [9, Lemma 3.3]. This is load-bearing because the rigorous energy estimate for w is obtained by passing to the limit R -> infinity in the local energy inequality; that passage requires controlling pressure-dependent terms such as integral integral p_u w . grad(phi_R), which are not obviously controlled in the supercritical class. Moreover, Definition 1.2 drops condition (1.4) (the energy inequality for u - e^{t Delta} u0) that is present in the L3,infinity definition in [9]; the authors assert the dropped term is 'harmless' for L2-decay, but they do not show that [9, Lemma 3.3] - which may rely on (1.4) - applies verbatim to the modified class. If the local energy inequality for w fails or requires (1.4), the a priori bound (Theorem 1.4), and therefore the separation-rate and time-regularity applications (Theorems 1.8 and 1.10), lose their proof.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper develops a theory of Lp,∞-weak solutions to the 3D Navier-Stokes equations for 2<p<3, extending the L3,∞ framework of Barker, Seregin and Šverák. The main technical result is an a priori estimate (Theorem 1.4) for the energy of u−e^{t∆}u0, which vanishes as t→0 at the explicit rate σ(p). The authors prove existence and weak-star stability in this class, and apply the a priori bound to two problems: a local short-time asymptotic expansion with a separation-rate corollary for hypothetical non-unique solutions (Theorem 1.8 and Corollary 1.9), and Hölder regularity in time at a singular time away from the singular point (Theorem 1.10).","tokens_in":35743,"tokens_out":20103,"duration_ms":221738,"significance":"If the proof gaps are closed, this is a substantial contribution. It moves the weak solution theory of [9] into a genuinely supercritical range, provides a parameter-free bound with explicit exponents, and the applications are non-vacuous: the time regularity result does not depend on the existence of non-unique solutions. The paper also correctly identifies the obstruction in Popkin's Besov setting and states a precise conjecture about which sub-scale might retain a decay estimate. The estimates are dimensionally balanced, the constants are explicit up to dependence on norms and p, and the main statements are falsifiable in the sense that the exponents are concrete. The principal weakness is that several load-bearing inequalities are asserted rather than proved.","major_comments":[{"comment":"The proof of Lemma 2.2 is the load-bearing step for Theorem 1.4, yet the local energy inequality for w = u − V is asserted with the sentence 'this is an easy calculation and is omitted' and a citation to [9, Lemma 3.3]. This is not sufficient for two reasons. First, Definition 1.2 drops condition (1.4) from the L3,∞ definition in [9], and the authors do not show that [9, Lemma 3.3] or its proof is independent of that condition. Second, Lemma 2.2 concerns heat extensions of arbitrary V0 ∈ L4, whereas [9, Lemma 3.3] is formulated for the heat extension of the initial data in the critical class. In particular, the passage R→∞ in the local energy inequality requires control of pressure terms such as ∫ p_u w·∇φ_R, and the manuscript does not provide that control under the modified definition. Please supply the omitted calculation, or state and prove a variant of [9, Lemma 3.3] with hypotheses that are verified by Definition 1.2.","section":"Definition 1.2 and Lemma 2.2"},{"comment":"The estimate |B(u−P0,(u−P0)χ0)(x,t)| ≲ ∫_0^t (t−s)^{-1/2} s^{2γ} ds is not justified by the preceding bounds. Lemma 3.2 gives |u−P0| ≲ s^{γ/2} (for q=∞), so the integrand should contain s^γ rather than s^{2γ}; if the extra power is intended, it requires a separate argument. The displayed estimate would then be O(t^{1/2+γ}), not O(t^{1/2+2γ}). Since the second and final steps of the iteration quote the first-step rates, the authors should verify explicitly that the final O(t^{1+σ−δ}) conclusion of Theorem 1.8 is unaffected by this correction.","section":"Section 3, first step"},{"comment":"The proof of stability under weak-star convergence states 'The local energy inequality is easy to prove and we omit the details.' Definition 1.2 requires the limit u to satisfy the local energy inequality, and this is not a purely cosmetic point: the term w·∇V wϕ, which the authors themselves mention, does not enjoy the cancellation used for the other nonlinear terms. The details should be included, or a precise reference should be given that covers the supercritical Lp,∞ setting rather than the critical setting of [9].","section":"Theorem 1.5, Section 2.2"}],"minor_comments":[{"comment":"The corollary says 'Let u and v be L3,∞-weak solutions with data u0 and v0,' but u0,v0 are assumed to lie in Lp,∞ for p∈(2,3), which does not imply membership in L3,∞; this should be Lp,∞-weak solutions.","section":"Corollary 1.9"},{"comment":"For p=3 the phrase 'Lp,∞-weak solution' is ambiguous because Definition 1.2 is only made for 2<p<3; please state explicitly that the p=3 case refers to Definition 1.1.","section":"Theorem 1.10"},{"comment":"In the dimensional analysis following the choice of N, the displayed calculation gives a length scale to the power (α−p)/(p−α), which equals −1; the sentence 'which matches the left-hand side' is not transparent, since the left-hand side has dimension length squared. Please spell out the dimensional bookkeeping.","section":"Corollary 2.4 proof"},{"comment":"The notation L^{p′,1}_y for Lorentz spaces is used in the proof of Lemma 3.2 without definition; a one-sentence definition or a reference would improve readability.","section":"Section 3, inequalities (3.2)-(3.6)"}],"recommendation":"major_revision","confidential_remarks":"The main obstacle is the omitted local energy inequality for w in Lemma 2.2. If that calculation cannot be supplied under Definition 1.2, then Theorems 1.4-1.10 lose their foundation. I recommend major revision rather than rejection because the surrounding argument is careful and the gap appears fixable by expanding the calculation or by stating a precise variant of [9, Lemma 3.3] with verified hypotheses."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a real extension of the Barker–Seregin–Šverák L3,∞ theory to Lp,∞ for 2<p<3, not a routine rerun. The a priori bound (Theorem 1.4) with explicit t^σ decay is the engine, and the two applications—local asymptotic expansion/separation rate and time regularity at a singular time—are substantive. I agree with the reader's CONDITIONAL verdict.\n\nWhat is new and works: the modified definition (dropping the (1.4)-type convergence) is defended honestly, and the main estimates are dimensionally sound. The far-field strategy in Theorem 1.8—building higher-order terms from localized lower-order terms—is a genuine trick to bypass the supercritical degradation of Picard iterates. Theorem 1.10 is especially nice: it is non-vacuous and quantifies exactly how the nonlocal pressure limits time regularity.\n\nSoft spots, in proportion: (1) Lemma 2.2 is load-bearing and its proof is actually deferred. The 'easy calculation' plus the reference to [9, Lemma 3.3] is plausible, but Definition 1.2 drops condition (1.4), and the text does not show that [9, Lemma 3.3] applies verbatim to the modified class. The pressure terms in the R→∞ passage are exactly where this could fail. This is fixable with a page of calculation, but a referee should push on it. (2) The Section 3 near-field estimate has a clear exponent slip: from |u−P0| ≲ s^{γ/2}, the product gives s^γ, not s^{2γ}. The displayed t^{1/2+2γ} is wrong as written. Since it is an upper bound, the final t^{1+σ−δ} rate is not immediately destroyed, but the bootstrap in the later steps should be rechecked with the correct exponent. It is a typo-level error in a delicate argument and deserves a careful fix. (3) Minor: the proof of Theorem 1.5 has a small range slip ('true for p∈(0,6)') and the local energy inequality convergence is again hand-waved, though less load-bearing.\n\nThe citation pattern is fine: reliance on [9] and [13] is standard, and the self-citations point to the actual sources of the techniques. No circularity.\n\nWho this is for: specialists in Navier–Stokes weak solution classes, and people working on non-uniqueness or separation rates. It deserves a serious referee. My recommendation: send it to review, with the referee asked to verify Lemma 2.2 and correct the Section 3 exponents before publication.","headline":"Genuine supercritical extension of the L3,∞ weak-solution theory with two fixable gaps—a deferred energy inequality and an exponent slip—that do not sink the core results.","tokens_in":36345,"tokens_out":5239,"would_cite":true,"duration_ms":57312,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q30","76D05","35B65"],"pacs":[],"model":"deepseek-v4-flash","headline":"In the 3D Navier-Stokes equations, every weak solution with supercritical $L^{p,\\infty}$ data ($2<p<3$) approaches the heat flow at the algebraic rate $t^{\\sigma(p)}$, and this rate controls non-uniqueness separation and time regularity…","keywords":["Navier-Stokes equations","weak solutions","Lorentz spaces","supercritical regimes","energy decay","separation rates","time regularity","pressure nonlocality"],"falsifier":"Compute, for one nontrivial pair $(u,V)$ with $V$ the caloric extension of a subcritical component of the initial data, whether the perturbation $w=u-V$ satisfies the claimed local energy inequality; a direct failure would remove Lemma 2.2 and with it the a priori bound. Alternatively, exhibit an $L^{p,\\infty}$-weak solution whose separation satisfies $\\liminf_{t\\to 0} t^{-\\sigma(p)}\\|u(t)-e^{t\\Delta}u_0\\|_{L^2}^2>0$, or two solutions with the same data whose local $L^\\infty$ separation exceeds $C t^{1+\\sigma}$ on a nested ball; either would contradict Theorems 1.4 and 1.8.","tokens_in":35295,"feed_emoji":"🌊","tokens_out":10009,"duration_ms":102010,"temperature":0.7,"pith_summary":"This paper proves that in the 3D Navier-Stokes equations, every weak solution with initial data in the supercritical Lorentz space $L^{p,\\infty}$, $2<p<3$, separates from the heat flow at a quantitative algebraic rate: $\\|u(t)-e^{t\\Delta}u_0\\|_{L^2}^2 \\le C_p(\\|u_0\\|_{L^{p,\\infty}}^{2p/(4-p)} t^{\\sigma(p)}+t^{1/2})$ with $\\sigma(p)=\\frac12\\frac{p-2}{4-p}$, so the $L^2$-distance to the heat flow vanishes as $t\\to 0$. The paper develops a full solution theory for this supercritical class, including existence and stability under weak-star convergence. Two consequences follow. First, two weak solutions with the same data cannot separate locally faster than $t^{1+\\sigma-\\delta}$; if their data agree on a ball but differ at infinity, local separation is at most $O(t)$. Second, at a regular spatial point near a singular time, $\\partial_t u$ is H\\\"older continuous in time with exponent $\\sigma(p)/2$ (or any exponent below $\\sigma(3)/2$ at $p=3$), with the nonlocal pressure as the limiting factor. The result matters because supercritical data are rougher than the critical $L^{3,\\infty}$ scale, and this is the first quantitative control on separation and time regularity in that range.","feed_headline":"Supercritical Navier-Stokes flows decay to heat flow at a fixed rate","feed_subtitle":"The rate bounds how fast non-unique solutions can separate and how rough time regularity becomes at a singularity.","key_machinery":"The load-bearing mechanism is a Calder\\'on-type splitting of the initial data into a smooth subcritical piece and a small supercritical piece, combined with an energy estimate for the perturbation between the fluid and a heat flow. Concretely, for $u_0\\in L^{p,\\infty}$ one writes $u_0=\\bar u_0^N+\\tilde u_0^N$ with $\\bar u_0^N\\in L^\\alpha$, $\\alpha\\in(3,4]$, and $\\tilde u_0^N\\in L^2$, with norms controlled by a parameter $N$; the perturbation $w=u-e^{t\\Delta}\\bar u_0^N$ satisfies Lemma 2.2's energy inequality, whose Gr\\\"onwall factor is an exponential of $\\int_0^t \\|V\\|_{L^4}^8\\,ds$. Choosing $N$ proportional to $t^{(12-4\\alpha)/(8(\\alpha-p))}$ makes that exponential $O(1)$ and leaves exactly the powers $t^{\\sigma(p)}$ and $t^{1/2}$. In the applications the same splitting feeds a localized bootstrap with cut-off functions: the first step gives a local expansion whose data-determined term is $P_\\Omega=P_1+\\tilde P_2$, and the pressure estimate in the time-regularity theorem uses the same decay to control the far-field singular integral.","core_discovery":"The paper's central claim is that in the supercritical range $2<p<3$, every $L^{p,\\infty}$-weak solution obeys the dimensionally balanced a priori bound $\\|u-e^{t\\Delta}u_0\\|_{L^2(t)}^2 + \\int_0^t \\|\\nabla(u-e^{s\\Delta}u_0)\\|_{L^2}^2\\,ds \\le C_p(\\|u_0\\|_{L^{p,\\infty}}^{2p/(4-p)}t^{\\sigma(p)}+t^{1/2})$, with $\\sigma(p)=\\frac12\\frac{p-2}{4-p}\\in(0,1/2)$. The proof separates the initial data into a subcritical $L^\\alpha$ part and a supercritical $L^2$ part, applies an energy inequality to the perturbation, and chooses the splitting scale so that the Gr\\\"onwall factor becomes time-independent. The bound is the engine for the paper's two applications: a local short-time expansion in which every solution agrees with a data-determined term up to $O(t^{1+\\sigma-\\delta})$, and a time-regularity theorem at points away from a singularity, where the pressure's far-field contribution limits $\\partial_t u$ to a $C^{0,\\sigma/2}_t$ class. The paper also proves existence of these supercritical weak solutions and stability of the class under weak-star convergence, extending the known critical theory.","pith_inferences":["If the exponent $\\sigma(p)$ in the a priori bound is sharp for $p<3$ — a question the paper leaves open — then the true separation rate for supercritical non-uniqueness would degrade as $p\\to 2^+$, interpolating between the critical $t^{1/4}$ rate and the formally unlimited rate of the finite-energy class; this would make non-uniqueness progressively harder to detect.","The splitting-plus-dimension-balancing strategy should transfer to supercritical Besov spaces near $L^{p,\\infty}$; the paper explicitly conjectures such a range, and a testable extension is to prove a version of Theorem 1.4 there.","Corollary 1.9 offers a concrete diagnostic: if two solutions whose data agree on a ball but differ outside it are observed to separate locally faster than linearly in time, then the local data profile, not the far field, is the driver of the difference.","The time-regularity theorem suggests that near a Type I singularity, away from the singular point, the velocity is $C^{1,\\sigma/2}$ in time; a sharper estimate on the far-field pressure could raise the exponent toward 1, isolating nonlocality as the only obstruction."],"forward_implications":["Every $L^{p,\\infty}$-weak solution, $2<p<3$, has quantitative $L^2$-decay to the heat flow near $t=0$, with the same bound extended to space-time norms $L^r(0,T;L^q)$, $r=2q/(2q-3)$.","Two weak solutions with identical supercritical data cannot separate locally faster than $t^{1+\\sigma-\\delta}$ in $L^\\infty$ on a fixed ball; if the data agree on a ball but differ in the far field, the local separation is at most $O(t)$.","At a singular time, away from the singularity, $\\partial_t u$ is H\\\"older continuous in time with exponent $\\sigma(p)/2$ (or any exponent below $\\sigma(3)/2$ at $p=3$), so the nonlocal pressure is the only source of limited time regularity.","The supercritical solution class is closed under weak-star limits of approximating finite-energy weak solutions, so the decay bound survives the approximation process."],"supporting_citations":[{"why":"supplies the critical $L^{3,\\infty}$ weak-solution framework, the existence and stability template, and the perturbation local energy inequality on which Lemma 2.2 relies.","marker":"[9]"},{"why":"introduces the Calder\\'on splitting into subcritical and supercritical parts that Theorem 1.4's a priori bound is built on.","marker":"[22]"},{"why":"provides the local smoothing theorem that anchors the short-time local asymptotic expansion in Theorem 1.8.","marker":"[36]"},{"why":"gives the critical-case asymptotic expansion and separation-rate theorem that the supercritical Theorem 1.8 extends.","marker":"[13]"},{"why":"defines the Caffarelli-Kohn-Nirenberg local energy inequality that membership in $L^{p,\\infty}$-weak solutions requires.","marker":"[21]"},{"why":"supplies existence of finite-energy weak solutions used as approximating solutions in the proof of Theorem 1.3.","marker":"[47]"},{"why":"provides the $L^p$-$L^q$ estimates for the heat semigroup used in the mild formulation and the a priori bounds' proof.","marker":"[68]"},{"why":"supplies the classical spatial-regularity estimates used for the local terms in the time-regularity application.","marker":"[58]"}],"fun_headline_variants":["Supercritical Navier-Stokes: sharp bounds control solution separation","A priori bounds tame supercritical Navier-Stokes flows","Separation rates and time regularity for supercritical Navier-Stokes","Supercritical Navier-Stokes: how fast non-unique solutions diverge"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the perturbation $w=u-e^{t\\Delta}V$ obeys the same local energy inequality as the solution itself; the paper states this as an easy calculation, omits it, and defers to a cited lemma, so if that inequality fails for a solution in the modified class, the a priori bound and both applications collapse.","fun_headline_variants_meta":{"raw":{"variants":["Supercritical Navier-Stokes: sharp bounds control solution separation","A priori bounds tame supercritical Navier-Stokes flows","Separation rates and time regularity for supercritical Navier-Stokes","Supercritical Navier-Stokes: how fast non-unique solutions diverge"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000646,"raw_usage":{"total_tokens":2993,"prompt_tokens":995,"completion_tokens":1998,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":611,"completion_tokens_details":{"reasoning_tokens":1924}},"tokens_in":611,"tokens_out":1998,"duration_ms":15446,"temperature":1.0,"reasoning_tokens":1924,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T05:59:09.829806+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute, for one nontrivial pair $(u,V)$ with $V$ the caloric extension of a subcritical component of the initial data, whether the perturbation $w=u-V$ satisfies the claimed local energy inequality; a direct failure would remove Lemma 2.2 and with it the a priori bound. Alternatively, exhibit an $L^{p,\\infty}$-weak solution whose separation satisfies $\\liminf_{t\\to 0} t^{-\\sigma(p)}\\|u(t)-e^{t\\Delta}u_0\\|_{L^2}^2>0$, or two solutions with the same data whose local $L^\\infty$ separation exceeds $C t^{1+\\sigma}$ on a nested ball; either would contradict Theorems 1.4 and 1.8.","supporting_citations":[{"cited_title":"and ˇSver´ ak, V., On stability of weak Navier-Stokes solutions with large L3,∞ initial data","cited_arxiv_id":null,"evidence_quote":"supplies the critical $L^{3,\\infty}$ weak-solution framework, the existence and stability template, and the perturbation local energy inequality on which Lemma 2.2 relies."},{"cited_title":"P., Existence of weak solutions for the Navier-Stokes equations with initial data in Lp","cited_arxiv_id":null,"evidence_quote":"introduces the Calder\\'on splitting into subcritical and supercritical parts that Theorem 1.4's a priori bound is built on."},{"cited_title":"and ˇSver´ ak, V., Local-in-space estimates near initial time for weak solutions of the Navier-Stokes equations and forward self similar solutions","cited_arxiv_id":null,"evidence_quote":"provides the local smoothing theorem that anchors the short-time local asymptotic expansion in Theorem 1.8."},{"cited_title":"and Phelps, P., Estimation of non-uniqueness and short-time asymptotic expansions for Navier-Stokes flows, Ann","cited_arxiv_id":null,"evidence_quote":"gives the critical-case asymptotic expansion and separation-rate theorem that the supercritical Theorem 1.8 extends."},{"cited_title":"and Nirenberg, L., Partial regularity of suitable weak solutions of the Navier-Stokes equations","cited_arxiv_id":null,"evidence_quote":"defines the Caffarelli-Kohn-Nirenberg local energy inequality that membership in $L^{p,\\infty}$-weak solutions requires."},{"cited_title":"Acta Math","cited_arxiv_id":null,"evidence_quote":"supplies existence of finite-energy weak solutions used as approximating solutions in the proof of Theorem 1.3."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the $L^p$-$L^q$ estimates for the heat semigroup used in the mild formulation and the a priori bounds' proof."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the classical spatial-regularity estimates used for the local terms in the time-regularity application."}],"review_version":1}