{"id":"dd9d84dd-d162-4768-8077-4c904011997b","arxiv_id":"2608.00234","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Viscous energy loss for 2D vortex sheets is at most O(1/|logν|), a square-root improvement over the previous bound, and fixed energy loss must wait at least a polynomial time in the Reynolds number.","lead":"This paper shows that viscous energy loss for two-dimensional vortex sheets is at most O(1/|log ν|), improving the previous O(1/√|log ν|) bound and refuting a recent conjecture. The gain comes from a sharp interpolation inequality that uses the total interaction energy of positive vorticity, not just its largest local mass.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified.","rationale":"The reader's weakest assumption was the propagated decomposition in Lemma 3.1. I examined that step in detail: the mollified components evolve by the same advection–diffusion operator, the maximum principle and L^p contraction give uniform bounds, weak-* limits yield f^ν∈L∞L^p and a measure m^ν with time marginal ||μ0||_M dt, and disintegration plus the distributional equality curlu^ν=μ^ν+f^ν give the required a.e. time-slice decomposition. The H^{-1} bound on μ^ν(t) follows from ||ω^ν(t)||_{H^{-1}}≤||u^ν(t)||_2 and L^p⊂H^{-1}; the Orlicz bound follows from L^p⊂L(logL)^{1/2}. Thus the key assumption is adequately supported. The only genuine mathematical error I found is Lemma 2.1's overbroad measure statement, which is false for atom masses on T^2. Since Theorem 1.3 applies it to H^1 densities and all applications at positive time involve smooth vorticities, the central claim remains valid. The sharpness examples and limitations are stated carefully and do not contradict the main theorem. Hence no verdict change.","tokens_in":22427,"tokens_out":35087,"duration_ms":347622,"concrete_test":"Verify Lemma 2.1 with ρ=δ on T^2: compute ||δ||_{H^{-1}}^2 = Σ_{k∈2πZ^2} 1/(1+|k|^2) < ∞ while P_δ(r)=1 for all r>0. If confirmed, restrict the lemma to non-atomic measures and re-check that every invocation in Theorem 1.3 and Sections 3–5 is on functions/smooth vorticities, so the central proof is unaffected.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim survives scrutiny. Theorem 1.1(ii) follows from Theorem 1.3 applied to the propagated decomposition; Lemma 3.1 is justified by the mollifier/linear-transport argument, the maximum principle, and the disintegration argument, which yields a.e. nonnegative μ^ν(t) with preserved mass and uniform H^{-1}/Orlicz bounds on the evolved components. The enstrophy integration and initial-layer argument (Lemma 4.1) are consistent. I find no load-bearing gap in (1.9)-(1.10) or the growing-time/waiting-time conclusions. One non-central overstatement should be flagged: Lemma 2.1 claims (2.3) for every finite nonnegative measure in H^{-1}(T^2), but an atom δ is in H^{-1}(T^2) with P_δ(r)=1, contradicting the bound. The proof of Theorem 1.3 uses (2.3) only for ρ∈H^1 (or for positive-time vorticities, which are smooth and non-atomic), so this does not threaten the main theorem; the lemma should be restated for atomless/absolutely continuous measures.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies viscous energy loss for two-dimensional Navier–Stokes solutions with Delort-type vortex-sheet initial data: vorticity split into a nonnegative measure μ_0^ν and an L^p background f_0^ν, with uniform energy and total variation bounds. The main theorem (Theorem 1.1) proves that for any fixed 0<δ<T, the accumulated dissipation D_ν(δ,T)=ν∫_δ^T∥ω^ν(t)∥_2^2 dt is bounded by CνT + C log( log(1/(νδ))/log(1/(νT)) ), hence O_δ,T,M,p(|logν|^{-1}) for fixed δ,T. This improves the previous O(|logν|^{-1/2}) bound and directly disproves Conjecture 1.6 of De Rosa–Marcotullio (Corollary 1.2). Under additional L^2 relative compactness of the initial velocities, the paper shows that D_ν(0,T_ν)→0 whenever log T_ν=o(|logν|), which includes all stretched-exponential and polynomial-in-ν^{-1} time scales, and gives a positive-power lower bound on the waiting time for any fixed energy loss. The paper also treats the Orlicz case L(logL)^α (Theorem 1.4), obtaining rates |logν|^{-min{2α,1}} and corresponding time scales, and constructs exact radial solutions on R^2 that attain the endpoint rates, together with a fixed-radial-datum theorem showing sub-logarithmic dissipation and a purely diffusive loss clock.","tokens_in":22651,"tokens_out":23884,"duration_ms":217985,"significance":"This is a substantial quantitative advance in the vanishing-viscosity theory of vortex sheets. The improvement from |logν|^{-1/2} to |logν|^{-1} is obtained by a sharp interpolation inequality (Theorem 1.3) that exploits the positivity of the singular measure through the total log-interaction energy rather than the mass of the worst small ball. The proof is self-contained and the sharpness examples are explicit: Proposition 6.1 attains the Orlicz exponents, Proposition 6.2 shows that the endpoint rate cannot be improved without relative compactness, and Theorem 6.4 reveals a rigidity for fixed radial data. The paper also provides clean falsifiable predictions (e.g., the negation of Conjecture 1.6) and introduces a useful interpolative tool that may have applications beyond this setting. The computational steps, including the Gaussian calculations and the enstrophy integration, are transparent and internally consistent.","major_comments":[],"minor_comments":[{"comment":"The proof's phrase 'an atom would give infinite energy on the diagonal' is slightly confusing because atomic measures are not in H^{-1}(T^2) in two dimensions, so they are already excluded by the hypothesis. The lemma is correct, but the statement could explicitly note that the H^{-1} condition rules out atoms, avoiding any appearance of a gap.","section":"Section 2, Lemma 2.1"},{"comment":"The text refers to 'Theorem 4.1' but the result is Lemma 4.1. Please correct the cross-reference.","section":"Section 4, proof of Theorem 1.1(iii)"},{"comment":"The proof refers to 'Theorem 5.1', 'Theorem 5.3', and 'Theorem 5.4' but these are Lemmas 5.1, 5.3, and 5.4. Update the cross-references for consistency.","section":"Section 5, proof of Theorem 1.4"},{"comment":"Near the end of the proof, 'the periodic Theorem 2.1' should be 'Lemma 2.1'. The same correction applies in the α≥1/2 paragraph.","section":"Section 5, proof of Theorem 1.4"},{"comment":"The proof applies 'Theorem 6.3' but the statement is Lemma 6.3. Please fix the cross-reference.","section":"Section 6, proof of Theorem 6.4"}],"recommendation":"minor_revision","confidential_remarks":"I reviewed the stress-test concern about Lemma 2.1 and found that it does not land: atomic measures are not in H^{-1}(T^2) in two dimensions, so the lemma is not contradicted. The manuscript is carefully written and the central claims are well supported. The only issues are typographical cross-reference errors and a possible clarification in Lemma 2.1. I recommend acceptance after minor revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Kollege,\n\nYou should know two things about this one: the main theorem is real, and the one objection that got flagged in the stress test does not survive contact with the paper. Armegioiu proves the dissipation rate O(|log nu|^{-1}) for two-dimensional vortex sheets in the Delort class, improving the previous O(|log nu|^{-1/2}) from De Rosa and Marcotullio, and shows their Conjecture 1.6 is false. Under relative compactness of the initial velocities, the loss vanishes on any subpolynomial time scale, which gives a polynomial lower bound on the lifespan of the inviscid model.\n\nWhat is new is the sharp L^2-H^1-H^{-1} interpolation inequality (Theorem 1.3) under a one-sided sign condition. It retains the total interaction energy of the positive vorticity instead of just the worst ball, and that is exactly the square-root-log improvement that carries the whole paper. The proof of the inequality is transparent: the Bessel kernel is positive with a log singularity, the cellwise Poincare argument is elementary, and the sharpness examples in Section 2 match the exponent. The enstrophy integration is exact; the initial-layer lemma is standard Fourier truncation. The Orlicz extension and the exact radial solutions in Section 6 are convincing, and the paper is careful to state which conclusions are not sharp.\n\nI could not find a load-bearing gap. The most delicate part is Lemma 3.1, where the vorticity decomposition is propagated from the initial data through a mollifier argument; the uniform H^{-1} and L(logL)^{1/2} bounds on the evolved components are established with care. The stress-test note claims Lemma 2.1 fails for an atomic measure because an atom is in H^{-1}(T^2). That is wrong: the H^{-1} norm of delta on the torus is a sum of 1/(1+|k|^2) over all frequencies, which diverges in two dimensions, so delta is not a counterexample. The lemma may deserve a clarifying sentence about atomless measures, but the theorem does not rest on a false statement.\n\nThe exposition is dense but honest. Section 1.2 explicitly lists the limits of the results. This paper is for researchers in vanishing viscosity and vortex-sheet limits; it deserves a serious referee. I would recommend acceptance after minor revision, mainly asking for a slightly more careful statement of Lemma 2.1 and maybe a pointer in the introduction to where the strict H^{-1} condition rules out atoms. I would take it for peer review and would cite it.","headline":"A sharp, credible improvement of the vortex-sheet dissipation rate; the one flagged objection does not hold up, and the paper deserves serious refereeing.","tokens_in":23191,"tokens_out":10272,"would_cite":true,"duration_ms":101937,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q30","35Q35","76D05","76B03"],"pacs":[],"model":"deepseek-v4-flash","headline":"Two-dimensional vortex sheets with one-signed vorticity dissipate energy at rate at most O(1/|log ν|), improving the previous O(|log ν|^{-1/2}) bound and disproving a conjecture.","keywords":["vortex sheets","vanishing viscosity","energy dissipation","one-signed measure vorticity","logarithmic interpolation","enstrophy","Navier-Stokes equations","waiting time"],"falsifier":"Compute Dν(δ,T)=ν∫_δ^T‖ω^ν(t)‖_2^2 dt for a family satisfying the initial bounds. For the explicit radial family in the paper, the result is exactly a positive constant times |logν|^{-2α}; for the noncompact L^p family it is log(T/δ)/(16πκ)|logν|^{-1}. Any family in the conjectured class with Dν(δ,1) decaying slower than 1/|logν| — for instance, numerically finding a viscosity-independent vortex-sheet datum with Dν(δ,1) ≳ 1/√|logν| — would refute the central claim.","tokens_in":22302,"feed_emoji":"🌀","tokens_out":7607,"duration_ms":70773,"temperature":0.7,"pith_summary":"The paper proves that for two-dimensional vortex sheets with a one-signed measure vorticity plus an L^p background, viscosity can remove kinetic energy only slowly as the Reynolds number grows: on any fixed time interval separated from zero, the viscous dissipation is at most order 1/|log ν|. This improves the previously known 1/√|log ν| bound under identical assumptions and rules out a recent conjecture that the slower rate is attained. The engine is a sharp L^2–H^1–H^{-1} interpolation inequality for nonnegative densities, which uses the total logarithmic interaction energy of the positive vorticity rather than the mass of a worst ball. With relative compactness of initial velocities, the paper also shows that any fixed energy loss must wait at least a positive power of the Reynolds number, so all subpolynomial observation times are nondissipative. The result extends to a logarithmic-Orlicz scale of backgrounds, with exponents that are attained by explicit radial solutions.","feed_headline":"Vortex-sheet energy loss capped at 1/|log ν|","feed_subtitle":"New bound kills the conjectured slower rate and shows any fixed loss waits a power of the Reynolds number.","key_machinery":"The central object is the sharp logarithmic interpolation inequality for nonnegative densities (Theorem 1.3): ‖ρ‖_2^4 log(e + ‖ρ‖_2^2/‖ρ‖_{H^{-1}}^2) ≤ C ‖ρ‖_{H^{-1}}^2(‖ρ‖_2^2 + ‖∇ρ‖_2^2), with a decomposition version for h = μ+f. It converts the finite H^{-1} interaction energy of the positive vorticity — measured by the positive Bessel kernel of (1-Δ)^{-1}, which counts every close like-signed pair — into a coercive L^2 versus H^1 control. Combined with the enstrophy identity Y' = -2νZ, it gives the uniform enstrophy bound Y(t) ≲ 1/(νt log(e+1/(νt))) and, after integration, the dissipation rate. The sign condition is essential: the logarithmic gain fails for general signed functions with","core_discovery":"Under the one-signed decomposition ω_0^ν = μ_0^ν + f_0^ν with μ_0^ν ≥ 0 and f_0^ν bounded in L^p (p>1), plus uniform kinetic energy and total vorticity variation, the author proves Dν(δ,T) = ν∫_δ^T ‖ω^ν(t)‖_2^2 dt ≤ C νT + C log( log(1/(νδ))/log(1/(νT)) ) ≲_{δ,T} 1/|log ν|. Because Dν is exactly the kinetic energy removed by viscosity, this is a quantitative no-anomalous-dissipation statement with an explicit rate. The same estimate disproves a published conjecture predicting that the rate 1/√|log ν| can be achieved by some viscosity-independent datum in this class. When initial velocities are relatively compact in L^2, the bound extends to growing observation times Tν with log Tν = o(|log ν","pith_inferences":["If the sharp interpolation inequality is the only obstruction, then the 1/|logν| rate should be universal for all fixed finite-energy one-signed data, not just radial ones; a good test is to run high-resolution vortex-sheet simulations with a single viscosity-independent initial condition and measure Dν(δ,1), which the paper's bound predicts decays at most like 1/|logν|.","Because Dν is literally the kinetic energy removed, the waiting-time theorem can be read as an observable prediction: at Reynolds number Re, no fixed-energy-loss threshold should be crossed before time Re^{a}; this is a concrete, falsifiable statement for shear-layer experiments or direct numerical simulations.","The mechanism — counting all close like-signed pairs via the positive Bessel kernel rather than a worst ball — is general enough that analogous interpolation inequalities might sharpen bounds in other settings where one-signed concentrations coexist with rough backgrounds, such as passive scalar filaments or aggregation models."],"forward_implications":["The previously conjectured lower bound Dν(δ,1) ≳ 1/√|log ν| is false for every finite-energy one-signed vortex-sheet datum with L^p background.","For subpolynomial observation times Tν = exp(o(|log ν|)), the accumulated dissipation Dν(0,Tν) vanishes as ν→0; in particular stretched-exponential times are nondissipative.","Any prescribed energy loss η appears first no earlier than time ν^{-aη} for some aη>0, giving a polynomial-in-Reynolds lower bound on the energetic lifetime.","If the background satisfies an L(log L)^α entropy bound, the rate becomes O(|log ν|^{-min{2α,1}}), with a waiting time that is stretched exponential below α=1/2 and polynomial at and above it.","On R^2, explicit radial heat-flow solutions attain the positive-time exponents for 0<α≤1/2, and a bounded-energy family attains the endpoint 1/|logν| rate only when relative L^2 compactness of initial velocities is dropped."],"fun_headline_variants":["Vortex-sheet dissipation improved to 1/|log ν|","Conjecture on vortex-sheet energy loss disproved","Energy loss in vortex sheets capped at logarithmic rate","Fixed vortex-sheet energy loss waits power of Reynolds number","Viscous loss in vortex sheets vanishes at log rate"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The argument relies on the propagated vorticity splitting ω^ν(t)=μ^ν(t)+f^ν(t) at almost every positive time, with μ^ν(t) nonnegative and the uniform bounds ‖μ^ν(t)‖_{H^{-1}} + ‖f^ν(t)‖_{L(logL)^{1/2}} staying bounded independently of ν; if the one-signedness of the evolved vorticity is lost, or if those component norms grow with time, the interpolation estimate and all dissipation bounds collapse.","fun_headline_variants_meta":{"raw":{"variants":["Vortex-sheet dissipation improved to 1/|log ν|","Conjecture on vortex-sheet energy loss disproved","Energy loss in vortex sheets capped at logarithmic rate","Fixed vortex-sheet energy loss waits power of Reynolds number","Viscous loss in vortex sheets vanishes at log rate"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000742,"raw_usage":{"total_tokens":3317,"prompt_tokens":1086,"completion_tokens":2231,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":830,"completion_tokens_details":{"reasoning_tokens":2149}},"tokens_in":830,"tokens_out":2231,"duration_ms":17081,"temperature":1.0,"reasoning_tokens":2149,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T00:57:57.215050+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute Dν(δ,T)=ν∫_δ^T‖ω^ν(t)‖_2^2 dt for a family satisfying the initial bounds. For the explicit radial family in the paper, the result is exactly a positive constant times |logν|^{-2α}; for the noncompact L^p family it is log(T/δ)/(16πκ)|logν|^{-1}. Any family in the conjectured class with Dν(δ,1) decaying slower than 1/|logν| — for instance, numerically finding a viscosity-independent vortex-sheet datum with Dν(δ,1) ≳ 1/√|logν| — would refute the central claim.","supporting_citations":[],"review_version":1}