{"id":"de7952e3-59fa-47d4-b103-4b32bdb0db74","arxiv_id":"2412.13828","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A 2D inhomogeneous Navier-Stokes Leray-Hopf solution becomes immediately regular exactly when it satisfies the strong energy inequality, when Danchin's weighted derivative estimates hold, and when an associated BMO-regular pressure exists; the equivalence yields weak-strong uniqueness.","lead":"This paper proves that several different notions of a Leray-Hopf weak solution to the 2D density-dependent Navier-Stokes equations becoming smooth immediately after the initial time are actually the same: strong energy inequality, weighted derivative bounds, and a pressure bound in BMO. As an application, the authors get a weak-strong uniqueness theorem and a unified view of recent well-posedness results.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equivalence (ii)⇔(iii) in Theorem 1.1 imports the full weighted estimates of unpublished [Dan24, Thm 1.1] without verifying its hypotheses for arbitrary time slices; if [Dan24] needs extra Besov or smallness conditions on the density or velocity, the central characterization loses support.","rationale":"After reading the full manuscript, I find the internal arguments for the equivalences (i)⇔(ii) (via the semi-flow property) and (iii)⇔(iv) (via the anti-gradient operator and the BMO local-energy argument) to be coherent and largely self-contained. The one step that is not self-contained is Proposition 3.6: the direction (ii)⇒(iii) imports Danchin's dynamic-interpolation theorem. The paper states the conclusion and the dependence of the constant but not the hypotheses. Since Theorem 1.1's characterization of immediately strong solutions via the weighted quantities A0_i depends entirely on this step, an unverified hypothesis in [Dan24] would directly undermine the central claim. My proposed check (reading the precise statement of [Dan24] and testing it on time slices of a strong-energy Leray-Hopf solution) settles the matter. I do not see an internal contradiction or a flaw in the paper's own estimates; the issue is an external dependency. This matches the reader's weakest assumption. I therefore recommend that acceptance be made conditional on confirming that [Dan24, Theorem 1.1] applies under exactly the hypotheses (1.3) for arbitrary time slices, as the authors assert. If confirmed, the paper's main theorem stands.","tokens_in":1327,"tokens_out":968,"duration_ms":225217,"concrete_test":"Obtain the precise statement of [Dan24, Theorem 1.1] (from the author or an arXiv posting) and check three things: (a) does it cover all initial data with 0<c0≤ρ0≤C0<∞ and u0∈L²(R²) (or at least u0∈H¹)? (b) does it yield A0_i≤C with C=C(ν,c0,C0,||u0||_{L²}) as claimed in footnote 7? (c) if it imposes any additional condition (e.g., u0∈B^0_{2,1}, ρ0−1 small in a Besov space), verify that every time slice (ρ(ε),u(ε)) of a Leray-Hopf solution satisfying the strong energy inequality automatically satisfies that condition. If (a) or (b) fails, or if (c) cannot be verified, Proposition 3.6 has a genuine gap and the equivalence (ii)⇔(iii) is unsupported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Proposition 3.6 (proof of (ii)⇒(iii)) is the load-bearing step: it takes a Leray-Hopf solution satisfying the strong energy inequality, shifts by ε, and uses [Dan24, Theorem 1.1] to obtain A0_i(uε) ≤ C for a solution with initial data (ρ(ε),u(ε)). The paper asserts in footnote 7 that Danchin's constant depends only on c0, C0, and ||u0||_{L²}, but it never states the hypotheses of [Dan24, Theorem 1.1]. For the argument to work, that theorem must apply to arbitrary time slices where ρ(ε) is merely bounded away from zero and u(ε)∈H¹, with no additional Besov regularity or smallness of ρ(ε)−1. If [Dan24] instead requires, e.g., u0∈B^0_{2,1} with a density-dependent norm or a smallness condition on ρ0−1, then Proposition 3.6 fails for general Leray-Hopf solutions, and the equivalence (i)/(ii)⇔(iii) in Theorem 1.1 collapses. The same proof also leans on Theorem 2.7 (from [CŠV25]) to identify the Danchin solution with the shifted solution; Remark 2.10 asserts a weakened energy-inequality hypothesis is enough, but this is likewise inherited from the companion paper. This is not a circularity, but it is an unverified external dependency placed at the center of the main theorem.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies Leray–Hopf solutions of the 2D inhomogeneous incompressible Navier–Stokes system with bounded density bounded away from zero and L² divergence-free initial velocity. Its main result, Theorem 1.1, characterizes the class of 'immediately strong' solutions—those with ∂ₜu and ∇²u in L²((ε,∞);L²) for every ε>0—by four equivalent conditions: immediate strong regularity, the strong energy inequality, boundedness of the weighted estimates A₀₁,A₀₂,A₀₃, and existence of an associated pressure in L²_loc((ε,∞);BMO)∩L²_loc. As applications the paper proves a critical weak–strong uniqueness theorem and a uniqueness corollary covering several recent well-posedness frameworks. The proof strategy combines a semi-flow property derived from the strong energy inequality, weighted estimates imported from Danchin's dynamic-interpolation theory, and a pressure construction via an anti-gradient operator.","tokens_in":29736,"tokens_out":15783,"duration_ms":144891,"significance":"If the main theorem is correct, this is a substantial contribution: it gives a clean characterization of when Leray–Hopf solutions of a model where basic uniqueness remains open become regular for positive times, and it unifies several recent well-posedness and uniqueness results. The paper is careful in its structure, provides a detailed pressure construction via a Banach-space anti-gradient operator, and formulates a genuinely critical weak–strong uniqueness statement. The arguments are mostly self-contained except for two clearly identified external tools, and there are no fitted parameters or post-hoc assumptions. The main caveat is that one of the two central external results, Danchin's Theorem 1.1, is not stated in the manuscript, so the reader cannot independently verify that its hypotheses are met at arbitrary Leray–Hopf time slices.","major_comments":[{"comment":"The proof of (ii)⇒(iii) is the load-bearing step that imports the weighted estimates from [Dan24, Theorem 1.1]. After shifting by ε, the paper invokes that theorem to obtain (3.3) for a solution with initial data (ρ(ε),u(ε)), and then uses Theorem 2.7 to identify this solution with the shifted original solution. However, the manuscript never states the hypotheses of [Dan24, Theorem 1.1], and footnote 7 only asserts that the constant depends on c₀, C₀, and the L² norm of the initial data. For the central equivalence to hold, that theorem must apply to arbitrary time slices where ρ(ε) is merely bounded away from zero and u(ε)∈H¹, with no additional Besov regularity or smallness condition on ρ(ε)−1. If any hidden hypothesis is missing, the chain (i)/(ii)⇔(iii) collapses. Please state [Dan24, Theorem 1.1] in full and verify its assumptions for the shifted data, or replace the import by a direct proof of the weighted estimates in the current setting.","section":"Section 3.2, Proposition 3.6"},{"comment":"The displayed change of variables in the computation of A₀₁(u_ε) is incorrect: after setting τ=s+ε, the weight s becomes τ−ε, not τ. The expression should involve sup_{τ∈(ε,T+ε)} (τ−ε)∫|∇u(τ)|² and ∫_ε^{T+ε} (τ−ε)(|∂_τ u|²+|∇²u|²+|∇P|²). The subsequent assertion that A₀₁(u_ε)→A₀₁(u) as ε→0 is therefore not justified and is false in general. The intended ε-independent bound is recoverable by a Fatou/liminf argument for the integral term and by choosing ε<t/2 for the supremum term, but this argument is absent and the current text does not establish the conclusion as written.","section":"Proof of Proposition 3.6, shift-back computation"}],"minor_comments":[{"comment":"There is a typo in the opening sentence: 'tsatisfying' should be 'satisfying'.","section":"Proposition 3.6"},{"comment":"The notation 'u0 ∈ rB0,s ρ0,s' appears garbled; please provide the correct Besov-space notation and define the space, or cite the exact definition from [Dan24].","section":"Corollary 1.9, item (3)"},{"comment":"The claim that G := −ρ∂ₜu − ρ(u·∇)u + νΔu belongs to L⁴((ε,∞);L⁴) is stated without proof. A short interpolation argument using the A₀ᵢ bounds would make this step self-contained and easier to verify.","section":"Proposition 4.7"},{"comment":"Since [Dan24] is listed as 'To appear' and is central to the proof of Theorem 1.1, including the precise statement of its Theorem 1.1 in an appendix would substantially improve the verifiability of the paper.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The main theorem is attractive and the overall architecture is sound, but the central equivalence depends on an unpublished theorem whose hypotheses are not stated, and the shift-back computation in Proposition 3.6 contains a concrete error. Both issues are fixable within the manuscript's scope. The reliance on [CŠV25] by three of the four authors is acceptable since that paper is published; however, the additional dependence on [Dan24] makes verification harder. Once the imported theorem is stated with hypotheses checked and the limiting argument is corrected, I would support acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Main point: Theorem 1.1 is a real structural result. The four-way equivalence—immediately strong, strong energy inequality, weighted A0 bounds, BMO pressure regularity—is new and gives a clean way to think about which Leray-Hopf solutions become regular. The pressure construction via the anti-gradient operator is elegant, and the BMO-pressure-to-suitability argument in Section 4 is a nice piece of analysis in its own right. Theorem 1.6, weak-strong uniqueness with regularity imposed on only one solution, is also a genuine improvement, and Corollary 1.9 widens the range of uniqueness for existing critical-space well-posedness results.\n\nMost of the proof is self-contained and coherent: semi-flow property from the strong energy inequality, energy equality for strong solutions, and careful handling of the local energy inequality with spatial cutoffs. I checked the steps I could verify mentally and found no hidden circularity. The self-citations to [CŠV25] are appropriate: that paper is published, and its weak-strong uniqueness theorem is used as a black box, which is normal.\n\nThe one soft spot is the step (ii)⇒(iii) in Proposition 3.6. The weighted A0 bounds are imported from Danchin's dynamic-interpolation theorem, which is listed as 'to appear.' The authors disclose via footnote 7 that the constant depends only on c0, C0, and the L² norm of the initial velocity, and they say the non-smooth case is handled in [Dan24, Section 2.2]. If that is accurate, then applying the theorem to time slices with u(ε)∈H¹ and bounded density away from vacuum is legitimate, and the stress-test worry about missing Besov hypotheses does not land. But the paper never states the actual hypotheses of Danchin's theorem, so a referee cannot verify this step without going to a separate unpublished manuscript. That is a real expositional gap, not a fatal flaw. The fix is straightforward: either state Danchin's theorem explicitly in the paper or give the few lines needed to show the L²-dependence.\n\nThe other external dependency, the weak-strong uniqueness theorem from the authors' own companion paper, is published, so it is less concerning. The same applies to Remark 2.10, which inherits a weakened-energy hypothesis from that paper; it would help to see the argument, but it is not a load-bearing issue.\n\nWho should read this: anyone working on inhomogeneous Navier-Stokes or on energy-inequality regularizing effects in fluid equations. It deserves a serious referee. My recommendation: send it to peer review, and ask the authors to make the hypotheses of [Dan24] explicit and confirm their applicability to all time slices. The central result likely holds, but the proof should be checkable within the paper.","headline":"A genuinely useful equivalence theorem for 2D inhomogeneous Navier-Stokes, with an honest but heavy dependency on Danchin's to-appear weighted estimates.","tokens_in":30293,"tokens_out":3160,"would_cite":true,"duration_ms":29418,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76D03","76D05"],"pacs":[],"model":"deepseek-v4-flash","headline":"A Leray–Hopf solution of the 2D inhomogeneous Navier–Stokes system is immediately strong if and only if it satisfies the strong energy inequality and admits a locally L², BMO pressure.","keywords":["inhomogeneous Navier–Stokes","Leray–Hopf solutions","immediately strong solutions","strong energy inequality","weak–strong uniqueness","BMO pressure","energy equality"],"falsifier":"Find a Leray–Hopf solution with data satisfying (1.3) that is strong on (ε,∞) for some ε>0 but fails the strong energy inequality; this would refute (i)⇒(ii) in Theorem 1.1. Equivalently, exhibit a Leray–Hopf solution whose associated pressure belongs to L²_loc((0,∞);BMO(R²))∩L²_loc and whose energy is strictly decreasing at a positive time, which would break the chain (iv)⇒(ii)⇒(i).","tokens_in":29226,"feed_emoji":"🌊","tokens_out":7393,"duration_ms":61327,"temperature":0.7,"pith_summary":"This paper characterizes, among all Leray–Hopf weak solutions of the 2D inhomogeneous incompressible Navier–Stokes system with bounded density bounded away from zero and square-integrable divergence-free velocity, exactly which ones become smooth for positive times. The main theorem states that four conditions are equivalent: the solution is immediately strong (time derivatives and second derivatives are square-integrable on every interval (ε,∞)); it satisfies the strong energy inequality; three weighted time-decay estimates on velocity and pressure are bounded; and there is an associated pressure lying locally in L² in both Lebesgue and BMO senses. This matters because in the inhomogeneous case the classical uniqueness arguments fail, and the equivalence converts the open question of uniqueness in the Leray–Hopf class into questions about energy behavior and pressure regularity. As applications the paper obtains a weak-strong uniqueness theorem at scaling-critical regularity for one of the two solutions, and uniqueness within the Leray–Hopf class for the strong solutions produced by several recent well-posedness frameworks.","feed_headline":"Instant smoothing equals energy control in 2D inhomogeneous fluids","feed_subtitle":"One theorem ties smoothness to energy, pressure, and decay; uniqueness then follows in the Leray–Hopf class.","key_machinery":"The argument is carried by three tools. First, the strong energy inequality for a Leray–Hopf solution, which turns every time translation into a Leray–Hopf solution and lets a weak-strong uniqueness theorem identify the translated solution with the smoother solution starting from an H¹ slice; this produces the immediately-strong regularity. Second, the weighted decay quantities A0₁, A0₂, A0₃ defined in (2.5), which measure time-weighted integrals of ∇u, ∂ₜu, ∇²u, ∇P, ∇Dₜu, and are bounded for strong solutions via a dynamic-interpolation well-posedness theorem that the paper imports and applies to time-shifted strong solutions. Third, an anti-gradient operator Φ that maps curl-free vector fields in L²∩L⁴ into $C₀^{{1/2}}$ ∩ BMO potentials, giving the pressure the BMO regularity that, together with commutator estimates, upgrades the local energy identity to the strong energy inequality.","core_discovery":"The central discovery, Theorem 1.1, is an equivalence for Leray–Hopf solutions of the 2D inhomogeneous Navier–Stokes system (density uniformly bounded above and below, initial velocity in L²σ): the solution is immediately strong if and only if it satisfies the strong energy inequality, if and only if the weighted quantities A0₁, A0₂, A0₃ are bounded by a constant depending only on the data, if and only if it admits an associated pressure in L²_loc((ε,∞);BMO(R²)) ∩ L²_loc((ε,∞)×R²) for every ε>0. The proof shows that the strong energy inequality gives the time-shifted solution a semi-flow property, so that an H¹ time slice can be identified, via weak-strong uniqueness, with the strong solution built from that slice, thereby producing instantaneous smoothing; conversely, an immediately strong solution satisfies the energy equality on positive times. The pressure implication builds an anti-gradient operator on curl-free fields in L²∩L⁴, obtains a BMO potential, and then uses commutator estimates to show that the BMO pressure bound forces the local energy inequality to globalize into the strong energy inequality.","pith_inferences":["If the equivalence is correct, then the open existence of Leray–Hopf solutions that never become strong is exactly the open existence of Leray–Hopf solutions violating the strong energy inequality; one can study the latter instead.","The BMO pressure condition may mark a scaling-critical threshold for energy equality in density-dependent fluids, playing the role that an L² pressure condition plays in the homogeneous case; this suggests testing anomalous dissipation in inhomogeneous flows at exactly this regularity.","The proof technique suggests a recipe for other systems with density coupling: any time-admissible solution class that grants a semi-flow property can be upgraded to immediate strongness whenever a weak-strong uniqueness theorem and time-weighted a priori estimates are available.","A computational check is possible: simulate smooth approximations of density variation with large gradients, bounded away from zero, and test whether the strong energy inequality holds along the approximation; a failure would produce the first example of a solution satisfying (iii) but not (ii), tightening the equivalence."],"forward_implications":["Every Leray–Hopf solution that is immediately strong automatically satisfies the strong energy inequality and the energy equality on positive times; so smoothing for positive times and energy behavior are inseparable.","If a Leray–Hopf solution admits a pressure that is locally L² in both space and BMO on the whole half-line [0,∞), then it conserves energy exactly (Corollary 1.5).","A weak-strong uniqueness theorem holds at scaling-critical regularity: any Leray–Hopf solution must coincide with an immediately strong solution that is continuous at time zero (Theorem 1.6).","The strong solutions constructed in several recent frameworks (critical Besov spaces, slightly supercritical data, bounded density without smallness) are unique in the entire Leray–Hopf class (Corollary 1.9).","For every admissible initial datum there exists at least one immediately strong Leray–Hopf solution (Proposition 1.3); whether any Leray–Hopf solution fails to be immediately strong remains open."],"supporting_citations":[{"why":"Supplies the dynamic-interpolation well-posedness theorem used to bound A0₁, A0₂, A0₃ for strong solutions, and the existence result behind Proposition 1.3.","marker":"[Dan24]"},{"why":"Provides the weak-strong uniqueness theorem (Theorem 2.7) that identifies the shifted Leray–Hopf solution with the strong solution from an H¹ time slice.","marker":"[CŠV25]"},{"why":"Builds the global strong solution from H¹ initial velocity with bounds (2.2), used in Theorem 2.8 and in the semi-flow argument.","marker":"[PZZ13]"},{"why":"Supplies the mollification and commutator strategy showing that an associated pressure in L²_loc forces suitability (local energy identity), adapted in Lemma 4.4.","marker":"[WB23]"},{"why":"Gives the commutator estimates and transport uniqueness used to handle the density difference in the weak-strong uniqueness proof.","marker":"[DL89]"},{"why":"Establishes the homogeneous-case energy equality under a pressure integrability condition, serving as the baseline that the BMO pressure assumption generalizes in Remark 4.6.","marker":"[Kuk06]"}],"fun_headline_variants":["Energy equality forces instant smoothing in 2D fluids","Pressure bounds decide strong solutions in inhomogeneous fluids","New equivalence: strong solutions meet energy inequality","2D Navier-Stokes: when weak becomes strong","Smoothing tied to energy and pressure in 2D"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The chain rests on importing two tools from outside the proof: a dynamic-interpolation theorem that bounds the weighted quantities A0₁, A0₂, A0₃ for strong solutions with H¹ initial data, and a prior weak-strong uniqueness theorem identifying a time-shifted Leray–Hopf solution with the strong solution built from its H¹ time slice; if either tool needs more regularity than the strong energy inequality alone guarantees for arbitrary time slices, the equivalence between (i)/(ii) and (iii) loses support.","fun_headline_variants_meta":{"raw":{"variants":["Energy equality forces instant smoothing in 2D fluids","Pressure bounds decide strong solutions in inhomogeneous fluids","New equivalence: strong solutions meet energy inequality","2D Navier-Stokes: when weak becomes strong","Smoothing tied to energy and pressure in 2D"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000139,"raw_usage":{"total_tokens":1113,"prompt_tokens":854,"completion_tokens":259,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":470,"completion_tokens_details":{"reasoning_tokens":183}},"tokens_in":470,"tokens_out":259,"duration_ms":2859,"temperature":1.0,"reasoning_tokens":183,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:45:18.053170+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a Leray–Hopf solution with data satisfying (1.3) that is strong on (ε,∞) for some ε>0 but fails the strong energy inequality; this would refute (i)⇒(ii) in Theorem 1.1. Equivalently, exhibit a Leray–Hopf solution whose associated pressure belongs to L²_loc((0,∞);BMO(R²))∩L²_loc and whose energy is strictly decreasing at a positive time, which would break the chain (iv)⇒(ii)⇒(i).","supporting_citations":[],"review_version":1}