{"id":"d881d166-7729-456d-817a-920bf2153599","arxiv_id":"1908.11035","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For 2D Navier-Stokes near Couette, ν^{1/2}-small perturbations in H^log_x L^2_y undergo enhanced dissipation at rate ν^{1/3} and inviscid damping, with L^2 initial data sufficient for any threshold exponent above 1/2.","lead":"This paper proves that small perturbations of the 2D Couette flow, measured in an almost critical space with logarithmic regularity, decay exponentially fast through mixing-enhanced dissipation and then return to the Couette flow. The result identifies, up to a logarithm, the minimal regularity needed for the nonlinear stability threshold of order ν^{1/2}.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified","rationale":"The reader's conditional verdict is driven by typographical errors and the title mismatch. The stress-test pass examined the substance of Theorem 1.1: the nonlinear enhanced dissipation and inviscid damping estimates in the almost critical space H^log_x L^2_y. The proof is a standard bootstrap with explicit constants; I checked the linear enhanced dissipation and inviscid damping lemmas, the Duhamel formulation for the nonzero modes, the Bony decomposition bounds (4.2)-(4.4), and the N2/N3 controls. The H^log assumption is indeed the most delicate input: it supplies the L∞_{x,y} control of V^1≠ and V^2≠ needed for the nonlinear terms N1,2 and N1,3, and the direct Fourier summation ∑_{α≠0} |α|^{-1}(ln|α|)^{-2} < ∞ confirms that H^log is sufficient for the log-weighted L∞ estimates. No circularity was found: the log-weighted L∞ bounds are proved from the same H^log norm of the initial data. Corollary 1.2's time-weight argument is sketched rather than fully detailed, but the extra |lnν| factor is consistent with the β ≥ 1/2 threshold and can be closed by standard arguments. The constant mismatches are cosmetic because the bootstrap constants are all set to a common value X. Therefore no load-bearing objection emerges, and the reader's conditional verdict does not need to change.","tokens_in":18497,"tokens_out":56482,"duration_ms":539666,"concrete_test":"Reproduce the constant tracking in Proposition 3.1 with C_k = X for k ∈ {0,2,...,8} and C1 = 5 max{M,1}, recomputing (3.16) and (3.18) directly from Lemma 3.3. In particular, verify that the N2 contribution uses the constant C3 from (3.9) rather than C2, and check that the chosen ε0 = 10^{-2} X^{-2} M^{-2} c makes the right-hand side of (3.16) at t0 = (ln 4M)/(cν^{1/3}) strictly less than 1/2 for every ν < 1; if the C2/C3 mismatch is only typographical, the tracked constants still close with the stated factor 4.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is supported by the proof. The bootstrap in Proposition 3.1 closes with the constants chosen in §3, and each nonlinear term in Lemma 3.3 is paired with a linear estimate from Lemma 2.1 or Lemma 2.2. The H^log_x L^2_y assumption is the least secure hypothesis, but it is used exactly where the authors say it is: in the L∞_{x,y} bounds on V^1≠ and V^2≠ via Lemma 2.2, (2.7)-(2.10), and in the Bernstein-type inequalities (4.2)-(4.4). This is an explicit, non-optimal regularity condition, acknowledged in Remark 1.2, and Corollary 1.2 quantifies the cost of dropping it as a |lnν| factor in the smallness. I found no hidden assumption or circular step that the conclusion depends on. The constant mismatches flagged by the reader, such as C2 versus C3 in the N2 estimate of Lemma 3.3, are absorbed because all bootstrap constants C_k are later set to the common value X; they affect exposition, not the closing of the bootstrap.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves nonlinear enhanced dissipation and inviscid damping for the 2D Navier-Stokes equations near Couette flow on T×R, when the initial vorticity perturbation lies in the almost critical space H^log_x L^2_y and has size at most ε0 ν^β with β ≥ 1/2. The main theorem (Theorem 1.1) gives an exponential decay rate ν^{1/3} for the nonzero Fourier modes in H^log_x L^2_y, an L^2 bound for the zero mode, and explicit inviscid damping estimates for the velocity perturbation. Corollary 1.2 extends the result to L^2 initial data with smallness ν^{1/2}/|ln ν|. The proof is self-contained: Section 2 derives linear estimates from an explicit Fourier solution, and Section 3 closes a bootstrap with seven nonlinear quantities using Bony paraproduct estimates, with the technical Littlewood-Paley facts and regularization lemmas collected in the appendix.","tokens_in":18662,"tokens_out":13861,"duration_ms":121258,"significance":"If the result is correct, it improves the known regularity threshold for nonlinear enhanced dissipation near Couette flow from high Sobolev or Gevrey spaces to a space only logarithmically weaker than L^2, which is the natural critical scale in this problem. The paper is fully self-contained, with all key lemmas proved in detail and explicit bootstrap constants. The proof is transparent and the statement is falsifiable through the precise estimates in Theorem 1.1 and Corollary 1.2. The main hypothesis, H^log_x L^2_y regularity, is used exactly where the authors indicate, and the non-optimality is explicitly acknowledged in Remark 1.2, with Corollary 1.2 quantifying the cost of dropping it.","major_comments":[],"minor_comments":[{"comment":"The inequality bounding ||V^1_0(s+τ)||_{L∞_y} by ||V^1_0(τ)||_{L²_y}^{1/2} ||ω_in||_{L²_{x,y}}^{1/2} is not justified in the text. It uses the one-dimensional Gagliardo-Nirenberg inequality together with the identity ∂_y V^1_0 = −ω_0 and the enstrophy bound ||ω_0(s)||_{L²_y} ≤ ||ω(s)||_{L²_{x,y}} ≤ ||ω_in||_{L²_{x,y}}. Please state these ingredients explicitly.","section":"3, proof of Lemma 3.3, estimate of N2"},{"comment":"In the split of the integral at t=1, the estimate on [0,1] is said to follow from the preceding gradient bound, but the reader must infer that ||α ln(|α|+e)ω||_{L²_t L²_y} is controlled by ||ln(|D_x|+e)∇ω||_{L²_t L²_{x,y}} since |α| ≤ |(α,η)|. Please add a sentence making this explicit.","section":"2, proof of Lemma 2.1, estimate (2.5)"},{"comment":"The abstract and introduction describe the smallness condition as being δ-close in H^log_x L^2_y to −1, while Theorem 1.1 states the condition as ||V_in||_{L²} + ||ω_in||_{H^log_x L^2_y} ≤ ε0 ν^β. The relation between δ and the norms appearing in the theorem should be clarified, since V_in is determined by ω_in through Biot-Savart.","section":"1, Theorem 1.1 and abstract"},{"comment":"In the proof of Lemma 4.2, the expression 'Cν^{−1/2}T^{1/2} ln((νT)^{−1}+e))||ω_in||²' appears to have an unbalanced parenthesis. The intended bound is clear, but the formula should be corrected.","section":"4, Lemma 4.2"},{"comment":"The text contains numerous typos and OCR-like artifacts, including 'Dissip A tion' and 'SP ACE' in the title, 'fist' for 'first', 'can cel' for 'cancel', and 'partical' for 'partial'. A careful proofreading pass is needed.","section":"Throughout"},{"comment":"The final display of Lemma 3.3 contains the factor C1(C2C5 + C6C2 + C2C0^{1/2} + C4C7 + C3C8); the terms C2C5 and C6C2 are identical in structure. This is not an error, but the notation could be simplified for readability.","section":"3, statement of Lemma 3.3"}],"recommendation":"minor_revision","confidential_remarks":"The paper is a competent and self-contained contribution to the stability threshold problem for Couette flow. The proof strategy follows the established Bedrossian–Masmoudi–Vicol framework, and the new element is the logarithmic regularity space. I found no load-bearing mathematical error in the main estimates; the flagged issues are local exposition gaps that do not affect the closing of the bootstrap. The paper is likely to be of interest to readers in mathematical fluid dynamics and should be published after minor revisions. No concerns about novelty or citation practices."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Main takeaway: this is a genuine advance. The theorem gets nonlinear enhanced dissipation and inviscid damping for 2D Couette with initial vorticity ν^{1/2}-small in H^log_x L^2_y, a space infinitesimally larger than L^2, and it gives an L^2 result with a log loss for β>1/2. Prior work stopped at Gevrey or H^s (s>1). So the central open threshold question is nearly settled.\n\nThe proof is in the standard bootstrap shape and it is carefully done. The linear estimates in Section 2 come from the explicit Fourier solution (2.11); I checked the key decay factors and the inviscid damping bounds, and they are consistent. Nonlinear terms in Lemma 3.3 are paired with the right linear estimates; the log regularity is used exactly where the authors say, in bounding V^1_≠ and V^2_≠ in L^∞ via (2.7)-(2.10). The bootstrap closes with constants chosen explicitly at the end of Section 3. There is no circularity.\n\nSoft spots are cosmetic rather than mathematical. Lemma 3.3 has a C_2/C_3 mismatch in the N_2 estimate that is absorbed later because all constants are set equal, but it will confuse a careful reader. The appendix equation references are sloppy. And the title says 'critical space' while the results hold in an almost critical space; the authors acknowledge this in Remark 1.2, but it is the kind of overstatement that invites complaints.\n\nThe weakest hypothesis is the log regularity itself. For exactly β=1/2 and L^2 data, the bootstrap loses a |lnν| factor (Corollary 1.2), so the theorem does not literally reach L^2 at the ν^{1/2} threshold. That is an explicit, visible limitation, not a hidden one.\n\nWho is this for: anyone working on stability thresholds for shear flows, enhanced dissipation, or the inviscid limit. It deserves a serious referee. I would recommend conditional acceptance with the typos fixed and the title adjusted. The mathematics holds up.","headline":"Masmoudi and Zhao nearly settle the ν^{1/2} threshold question for 2D Couette by pushing the initial data space to H^log_x L^2_y; the proof is sound modulo typos and a slightly overstated title.","tokens_in":19256,"tokens_out":1614,"would_cite":true,"duration_ms":17128,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q30","76D05","76E05","35B40"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that a log-regular perturbation of the two-dimensional Couette flow, small of order $\\nu^{1/2}$ in $H^{\\log}_xL^2_y$, still decays at the enhanced rate $e^{-c\\nu^{1/3}t}$ and satisfies inviscid damping.","keywords":["Couette flow","Navier-Stokes equations","enhanced dissipation","inviscid damping","stability threshold","critical space","vorticity","two-dimensional channel"],"falsifier":"Simulate (1.3) on $\\mathbb{T}\\times\\mathbb{R}$ with initial vorticity whose $x$-Fourier coefficients decay like $(|\\alpha|\\ln(2+|\\alpha|))^{-1}$, so the $H^{\\log}_xL^2_y$ norm is finite while no positive Sobolev regularity is available, set the $L^2$ norm to exactly $\\varepsilon_0\\nu^{1/2}$, and measure whether each $\\nu^{-1/3}$ time window removes a fixed fraction of the $H^{\\log}$ norm; if the removal rate vanishes as $\\nu\\to0$, the theorem's $\\nu^{1/3}$ rate is false.","tokens_in":18287,"feed_emoji":"🌊","tokens_out":15917,"duration_ms":143384,"temperature":0.7,"pith_summary":"The paper asks how much regularity a perturbation of 2D Couette flow needs before viscosity's mixing-enhanced dissipation takes over. It proves that if the initial vorticity is $\\nu^{\\beta}$-close to the Couette vorticity $-1$ in the almost critical space $H^{\\log}_xL^2_y$, defined by $\\|\\ln(e+|D_x|)f\\|_{L^2_{x,y}}<\\infty$, with $\\beta\\ge 1/2$, then the nonlinear solution inherits the linear behavior: the nonzero-frequency vorticity decays like $e^{-c\\nu^{1/3}t}$, the zero-frequency mode stays bounded, and the velocity field obeys inviscid damping. This lowers the regularity bar from Gevrey class or $H^s$ with $s>1$ to a space only logarithmically more regular than $L^2$, the scale expected for a critical threshold. The same argument gives an $H^{\\varepsilon}_xL^2_y$ version for every $\\varepsilon>0$ and, after a short-time regularization step, a plain $L^2$ version for $\\beta>1/2$.","feed_headline":"Log-regular vorticity is enough for fast Couette decay","feed_subtitle":"At the square-root-of-viscosity threshold, log-regular data keep Couette's mixing-enhanced decay and inviscid damping.","key_machinery":"The central mechanism is the linearized semigroup $S(t,s)$ for $\\partial_t\\omega+y\\partial_x\\omega-\\nu\\Delta\\omega=0$, combined with a bootstrap that runs entirely in $H^{\\log}_xL^2_y$, the subspace of $L^2_{x,y}$ with finite $\\|\\ln(e+|D_x|)f\\|_{L^2}$. In the moving-frame variable $W(t,x,y)=\\omega(t,x+yt,y)$, the linearized equation becomes $\\partial_t\\widehat W+\\nu(\\alpha^2+(\\eta-\\alpha t)^2)\\widehat W=0$, so the Fourier multiplier $e^{-\\nu(\\alpha^2t^3/3+\\eta\\alpha t^2+\\eta^2t+\\alpha^2t)}$ produces the $\\nu^{1/3}$ enhanced dissipation for nonzero $x$-frequencies. The nonlinear part is controlled with Bony's paraproduct decomposition and Bernstein-type inequalities on the $x$-torus; the logarithmic weight is used precisely in the two terms where $V_{\\neq}$ has low $x$-frequencies and the two-dimensional embedding $H^1\\not\\subset L^{\\infty}$ would otherwise fail (Lemma 3.3 and inequalities (4.2)-(4.4)).","core_discovery":"On the paper's own terms, the discovery is that the nonlinear vorticity equation around Couette flow admits a bootstrap that closes in the almost critical norm $\\|\\omega\\|_{H^{\\log}_xL^2_y}=\\|\\ln(e+|D_x|)\\omega\\|_{L^2_{x,y}}$, provided the initial perturbation is no larger than $\\varepsilon_0\\nu^{\\beta}$ with $\\beta\\ge 1/2$. With this input, Theorem 1.1 gives the nonlinear enhanced dissipation estimate $\\|\\omega_{\\neq}(t)\\|_{H^{\\log}_xL^2_y}\\le Ce^{-c\\nu^{1/3}t}\\|\\omega_{\\mathrm{in}}\\|_{H^{\\log}_xL^2_y}$, the zero-mode bound $\\|\\omega_0(t)\\|_{L^2_y}\\le C\\|\\omega_{\\mathrm{in}}\\|_{L^2_{x,y}}$, and three inviscid damping bounds: on $V^2_{\\neq}$ in $L^{\\infty}_{x,y}$, on $|D_x|^{1/2}V^2_{\\neq}$ in $L^2_xL^{\\infty}_y$, and on $\\partial_xV^1_{\\neq}$ in $L^2_{x,y}$, all with constants independent of $\\nu$. The argument also yields an $H^{\\varepsilon}_xL^2_y$ version for any $\\varepsilon>0$ and, after a short-time regularization admitting a logarithmic loss, an $L^2$ statement for $\\beta>1/2$.","pith_inferences":["Editorial inference: since the logarithmic weight is used in only two nonlinear estimates, replacing $\\ln(e+|D_x|)$ by $(\\ln(e+|D_x|))^\\gamma$ with $\\gamma>1/2$ should still close the bootstrap, making Remark 1.2's non-optimality claim directly checkable by modifying Lemma 3.3.","Editorial inference: the result suggests that at the critical amplitude $\\nu^{1/2}$ the correct function-space threshold is logarithmic rather than any fixed Sobolev regularity; a numerical test with $L^2$ data of size $\\nu^{1/2}|\\ln\\nu|^{-1}$, as in Corollary 1.2, could check whether the $e^{-c\\nu^{1/3}t}$ decay is still visible.","Editorial inference: the separation of the zero mode $\\omega_0$ from the nonzero frequencies is what makes the $\\nu^{1/2}$ threshold natural, and the same decomposition should extend to other monotone shear flows whose linearized phase produces the same $\\nu t^3$ structure, with the mean-flow equation re-derived for that profile."],"forward_implications":["For $\\beta>1/2$, Corollary 1.2 makes plain $L^2_{x,y}$ an admissible initial space, since the short-time regularization argument absorbs the logarithmic loss and no extra derivative regularity is needed.","The theorem fixes the stability threshold for 2D Couette flow at $\\beta=1/2$ in the sense that $\\nu^{1/2}$-smallness plus a logarithmic factor in $x$ recovers the linear decay rate $\\nu^{1/3}$.","For times $t\\gg\\nu^{-1/3}$ the solution approaches a nearby shear flow and then converges back to Couette flow as $t\\to+\\infty$, matching the linearized prediction.","The inviscid damping estimates are independent of $\\nu$: the integrated $L^{\\infty}_{x,y}$ norm of $V^2_{\\neq}$, the half-derivative quantity $\\||D_x|^{1/2}V^2_{\\neq}\\|_{L^2_xL^{\\infty}_y}$, and $\\|\\partial_xV^1_{\\neq}\\|_{L^2_{x,y}}$ are all controlled by the initial $H^{\\log}_xL^2_y$ norm.","Because $c$ and $C$ do not depend on $\\nu$, the $\\nu^{1/3}$ dissipation rate is uniform as the viscosity tends to zero, so the mixing enhancement persists in the inviscid limit."],"supporting_citations":[{"why":"Prior Gevrey-class result for $\\beta=0$; supplies the baseline enhanced-dissipation and inviscid-damping estimates that the present work generalizes to lower regularity.","marker":"[5]"},{"why":"Prior $H^s$ result for $\\beta=1/2$ with $s>1$; the Sobolev threshold that the paper weakens to the almost critical $H^{\\log}_xL^2_y$ space.","marker":"[6]"},{"why":"Establishes the stability-threshold framework for Couette flow in Sobolev regularity and sets the threshold problem this paper addresses.","marker":"[4]"},{"why":"Provides the Littlewood-Paley theory and Bony's decomposition on $\\mathbb{T}\\times\\mathbb{R}$ used in the nonlinear estimates (4.2)-(4.4).","marker":"[1]"},{"why":"Source for the paraproduct and Bony-type inequalities needed in Lemma 3.3.","marker":"[8]"},{"why":"Gives the Gagliardo-Nirenberg inequality (4.5) used in the linear inviscid damping estimate.","marker":"[12]"},{"why":"Supplies the Minkowski integral inequality (4.6) used to bound the linear velocity in the inviscid damping proof.","marker":"[13]"}],"fun_headline_variants":["Log-regular data hit square-root viscosity threshold for Couette decay","Critical space Couette: log-regular vorticity triggers enhanced decay","Couette mixing wins with log-regular initial vorticity","At ν^1/2 perturbation, log-regular Couette flow decays fast","Nonlinear enhanced dissipation from log-regular Couette data"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof needs the initial vorticity to be small in a space with a full logarithmic derivative in $x$, namely $\\|\\ln(e+|D_x|)\\omega_{\\mathrm{in}}\\|_{L^2}<\\infty$, rather than merely in $L^2$; for $\\beta=1/2$ exactly, $L^2$ data require an extra $|\\ln\\nu|^{-1}$ factor in the smallness condition, and without that factor the bootstrap does not close.","fun_headline_variants_meta":{"raw":{"variants":["Log-regular data hit square-root viscosity threshold for Couette decay","Critical space Couette: log-regular vorticity triggers enhanced decay","Couette mixing wins with log-regular initial vorticity","At ν^1/2 perturbation, log-regular Couette flow decays fast","Nonlinear enhanced dissipation from log-regular Couette data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000583,"raw_usage":{"total_tokens":2782,"prompt_tokens":1023,"completion_tokens":1759,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":639,"completion_tokens_details":{"reasoning_tokens":1669}},"tokens_in":639,"tokens_out":1759,"duration_ms":12287,"temperature":1.0,"reasoning_tokens":1669,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:27:33.701478+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate (1.3) on $\\mathbb{T}\\times\\mathbb{R}$ with initial vorticity whose $x$-Fourier coefficients decay like $(|\\alpha|\\ln(2+|\\alpha|))^{-1}$, so the $H^{\\log}_xL^2_y$ norm is finite while no positive Sobolev regularity is available, set the $L^2$ norm to exactly $\\varepsilon_0\\nu^{1/2}$, and measure whether each $\\nu^{-1/3}$ time window removes a fixed fraction of the $H^{\\log}$ norm; if the removal rate vanishes as $\\nu\\to0$, the theorem's $\\nu^{1/3}$ rate is false.","supporting_citations":[{"cited_title":"Bedrossian, N","cited_arxiv_id":null,"evidence_quote":"Prior Gevrey-class result for $\\beta=0$; supplies the baseline enhanced-dissipation and inviscid-damping estimates that the present work generalizes to lower regularity."},{"cited_title":"Bedrossian, V","cited_arxiv_id":null,"evidence_quote":"Prior $H^s$ result for $\\beta=1/2$ with $s>1$; the Sobolev threshold that the paper weakens to the almost critical $H^{\\log}_xL^2_y$ space."},{"cited_title":"Bedrossian, P","cited_arxiv_id":null,"evidence_quote":"Establishes the stability-threshold framework for Couette flow in Sobolev regularity and sets the threshold problem this paper addresses."},{"cited_title":"Bahouri, J","cited_arxiv_id":null,"evidence_quote":"Provides the Littlewood-Paley theory and Bony's decomposition on $\\mathbb{T}\\times\\mathbb{R}$ used in the nonlinear estimates (4.2)-(4.4)."},{"cited_title":"Chemin, Perfect Incompressible Fluids, Oxford Un iversity Press, New York, 1998","cited_arxiv_id":null,"evidence_quote":"Source for the paraproduct and Bony-type inequalities needed in Lemma 3.3."},{"cited_title":"Nirenberg, On elliptic partical differential equations, Ann","cited_arxiv_id":null,"evidence_quote":"Gives the Gagliardo-Nirenberg inequality (4.5) used in the linear inviscid damping estimate."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Minkowski integral inequality (4.6) used to bound the linear velocity in the inviscid damping proof."}],"review_version":1}