{"id":"de7603a0-0426-4d56-8e10-fffe0a6d21e7","arxiv_id":"2507.23397","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Global well-posedness and scattering for the 2D modified Zakharov-Kuznetsov equation at a new critical regularity, achieved with an anisotropic two-parameter Sobolev space.","lead":"This paper proves that the 2D modified Zakharov-Kuznetsov equation has global solutions that scatter to linear waves for small data in a new weighted space. The result reaches a regularity threshold that earlier methods could not touch, and the new space is designed to handle the equation's weak dispersion near the frequency axes.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The endpoint result depends on the sharp weighted L4 Strichartz estimate (Lemma 2.8), which is imported with only a one-line determinant justification; if that estimate loses any power in the anisotropic frequency ratio, Theorem 1.1 fails.","rationale":"I read the paper in good faith: the central claim is Theorem 1.1, and the proof structure is a fixed-point argument over the atomic space Y, with the multilinear estimate (3.5) as the core. The reader identified the same load-bearing assumption I would: the weighted L4 Strichartz estimate Lemma 2.8 and the imported bilinear estimates. My independent scaling check for a single dyadic block with N≫M indicates that the L4 contribution with the |D_x|^{1/8}|D_y|^{1/8} weight is essentially O((M/N)^{3/2} log(N/M)) after the relevant time integration, so I did not find a concrete counterexample. The case analysis in Proposition 3.3 is intricate but internally consistent; the interpolation steps and the high-modulation argument in Case 2C2 work out, including the small-|ξ2+ξ3| regime where two factors in the ξ-resonant product are small. The sharpness results in Section 5 are standard. The only genuine risk is that Lemma 2.8 is cited rather than proved, and the determinant computation does not by itself establish the sharp exponent; if the true estimate required a stronger weight or a loss in one frequency variable, the endpoint s=0, a=1/4 would fail. This does not change my confidence in the result, but it does mean the paper's acceptance should carry the explicit condition that this cited estimate be independently verified. Since the reader already flagged this caveat and the overall verdict of ACCEPT seems appropriate, I recommend leaving the verdict unchanged.","tokens_in":1267,"tokens_out":1080,"duration_ms":631939,"concrete_test":"Independently verify Lemma 2.8 for the worst anisotropic dyadic block: take \\hat f_{N,M} = (NM)^{-1/2} on [N,2N]×[M,2M] with N≫M and compute ‖ |D_x|^{1/8}|D_y|^{1/8} e^{-t(∂x^3+∂y^3)} f_{N,M} ‖_{L^4_{t,x,y}}^4 by the separable product reduction of the phase. Check that the result is O(1) uniformly as N/M→∞ and, symmetrically, as M/N→∞; if it grows, the missing power cannot be absorbed by K in (3.6), and Theorem 1.1 fails. Also check Lemma 2.10(2) with the same block to confirm the (L1L2)^{1/2} factor is sharp, since Corollary 3.4 relies on it.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 1.1 reduces to the multilinear estimate (3.5), which in Proposition 3.3 is built on three imported inputs: Lemma 2.8, Lemma 2.10, and Lemma 2.11. Lemma 2.8 is the most load-bearing because it is used directly in the baseline bound (3.9) and again in Cases 1-3 via the U^4 embedding (2.15); a failure or even a logarithmic loss in the weight exponent would propagate through every interpolation step. The proof given (\"follows directly from [30], since det D^2(ξ^3+η^3)=36ξη\") does not display the argument, and the cited determinant merely identifies where curvature vanishes; it does not by itself explain why the compensating weight |D_x|^{1/8}|D_y|^{1/8} is exactly critical. In particular, one needs the estimate uniformly for all anisotropic dyadic blocks (N,M), including N≫M and N≪M, since Proposition 3.3 sums over all such blocks and the K-factor in (3.6) cannot absorb any N/M growth. The same concern applies to the bilinear estimates (2.10)-(2.18) imported from [33,42]: they carry the anisotropic factors that make the Case 1A and Case 2C bounds close. If any of these estimates requires a stronger weight or a nonzero regularity loss, the Y-space contraction in Theorem 2.6 and the endpoint s=0, a=1/4 would not follow. This is a dependency concern, not a demonstrated contradiction; the internal case analysis of Proposition 3.3 appears coherent, and the scaling heuristics indicate the stated exponent is plausible.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the Cauchy problem for the two-dimensional modified Zakharov-Kuznetsov equation in the symmetrized form ∂t v + (∂x^3 + ∂y^3)v + (∂x + ∂y)(v^3) = 0. The authors introduce a two-parameter scale H^{s,a}(R^2) with anisotropic weights |D_x|^a|D_y|^{-a} and |D_y|^a|D_x|^{-a}, which scales like the classical H^s spaces. The main results are local well-posedness in H^{s,a} for s+a ≥ 1/4, 0 ≤ a < 1/4, and global well-posedness with scattering for small data in the critical space H^{0,1/4}(R^2). The proof reduces the critical case to a multilinear estimate (Proposition 3.3) proved by a detailed dyadic case analysis using weighted L^4 Strichartz estimates, bilinear Strichartz estimates, and U^2/V^2 atomic spaces. The paper also proves that the data-to-solution map is not C^3 below this threshold, and that no global fixed-point argument can work for a < 1/4.","tokens_in":23652,"tokens_out":26932,"duration_ms":285910,"significance":"If the proof is correct, Theorem 1.1 is the first scaling-critical global well-posedness and scattering result for the 2D modified Zakharov-Kuznetsov equation. The H^{s,a} scale is a natural device to compensate for the anisotropic loss of dispersion near ξη = 0, and the paper gives both positive results and sharp C^3 ill-posedness statements. The authors are transparent about the imported ingredients: the weighted L^4 Strichartz estimate is taken from Kenig-Ponce-Vega and the bilinear estimates from earlier work of Molinet-Pilod and the second author. The internal case analysis of Proposition 3.3 is extensive and the reduction from the multilinear estimate to the fixed-point argument is standard. These are genuine strengths that make the paper of high interest to the dispersive PDE community.","major_comments":[{"comment":"The definition of the weight ω in Section 2.2 is inconsistent with (3.1) and with the definition of H^{0,1/4}. As printed, ω(ξ,η) = |ξ|^{1/4}|η|^{1/4} + |η|^{1/4}|ξ|^{1/4} = 2|ξη|^{1/4}, whereas the Y norm must correspond to |ξ|^{1/4}|η|^{-1/4} + |η|^{1/4}|ξ|^{-1/4} to be equivalent to the H^{0,1/4} norm; the latter is exactly what is used in (3.1). In the same direction, the target (3.8) has the left factor M_4^{5/4}N_4^{1/4}, but after absorbing the derivative and ω_4 one expects M_4^{5/4}N_4^{-1/4} (as is used later in (3.12)). I believe these are typos, but they must be corrected and all subsequent exponents checked, because the Y norm is the central resolution space for Theorem 1.1.","section":"§2.2 and §3, Eq. (3.8)"},{"comment":"The weighted L^4 Strichartz estimate (2.7) is the endpoint input for the baseline bound (3.9) and, via (2.15), for every case of Proposition 3.3. The proof given is only the sentence \"This follows directly from [30], since det D^2(ξ^3+η^3) = 36ξη.\" The determinant identifies where the curvature vanishes, but it does not by itself explain why the compensating weights |D_x|^{1/8}|D_y|^{1/8} are exactly critical, nor does it display the uniform dependence on anisotropic dyadic rectangles with N ≫ M or N ≪ M. Since the K-factor in (3.6) cannot absorb any loss in the weight exponent, I ask the authors to provide a complete proof of (2.7) or a precise reference to the exact theorem/equation in [30] that implies this estimate in the required uniform form.","section":"§2.3, Lemma 2.8"},{"comment":"The dyadic summation estimates (3.14) and (3.15) are load-bearing: they are what convert the interpolated bounds into the K-factor (M_min/M_max)^{0+}(N_min/N_max)^{0+} times ω_1ω_2ω_3. As written, these inequalities are simply asserted, and the interpolation step in (3.12)–(3.13) suppresses a logarithmic factor that must be absorbed. This is the hardest part of the case analysis for a reader to verify. Please expand the derivation of (3.14)–(3.15), stating explicitly the summation lemma and the choice of the small parameters ε, δ and the 0+ exponents.","section":"§3, Eqs. (3.14)–(3.15)"}],"minor_comments":[{"comment":"The proof of Theorem 1.2 is announced as \"we prove Theorem 1.1\"; this should read Theorem 1.2 (or Theorem 2.7).","section":"§4, first line"},{"comment":"The proof begins with \"when 0 ≤ a < 1/4\", but the proposition states 0 ≤ a ≤ 1/4. Please clarify how the endpoint a = 1/4, s < 1/4 - a is handled, or adjust the statement if the proof really requires a < 1/4.","section":"§5, Proof of Proposition 1.5"},{"comment":"The fixed-point argument is written only for t ≥ 0. Since the equation is reversible under (t,x,y) ↦ (-t,-x,-y), global well-posedness on R follows, but this should be stated explicitly.","section":"§3, Proof of Theorem 1.1"},{"comment":"Item (2) is stated but the proof given only covers item (1). Please add a sentence explaining that (2) follows by the same argument after inserting the modulation projections Q_{L_j}.","section":"§2.3, Lemma 2.11"},{"comment":"There are several small typos: \"SCATTERING\" and \"EQUATION\" in the title, \"spces\" in Section 2.1, \"equativalent\" in Section 4, and the unused symbol δ in the K of Remark 3.2. These should be corrected in the final version.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"This is a strong paper and I believe the main theorem is very likely correct. My recommendation of major revision is driven by the need to make the proof of the imported weighted L^4 estimate and the dyadic summation bounds verifiable, and by the notational inconsistency in the definition of the Y space. If the authors can address these points in a revised version, I would support publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is the first L^2-critical global well-posedness and scattering result for the 2D modified ZK equation, and it looks correct to me. The new H^{s,a} space is a genuinely useful idea, and the main multilinear estimate is worked out in enough detail to be checked. The one thing I would ask the authors to tighten is the sourcing of Lemma 2.8, but that is a presentation issue rather than a mathematical gap.\n\nThe positive content: the paper defines H^{s,a} with weights |D_x|^a|D_y|^{-a} and |D_y|^a|D_x|^{-a}, which scales like H^s and penalizes concentration near \\xi\\eta=0 where dispersion degenerates. The main theorem, global well-posedness and scattering for small data in H^{0,1/4}, is a real advance; prior work was stuck at s>1/4 in H^s. The subcritical range s+a\\ge 1/4 follows by interpolation from the critical estimate and Kinoshita's earlier H^{1/4} bound, and the C^3 ill-posedness results show the threshold is sharp. The proof of the critical multilinear estimate (Prop 3.3) is a long case analysis, but the structure is coherent: it decomposes frequency interactions and uses bilinear Strichartz, the L^4 estimate, and high-modulation bounds. I could not spot a circular step, and the scaling checks out.\n\nThe soft spot, as your skeptic notes, is that Lemma 2.8 is load-bearing and imported with a one-line proof. The determinant computation identifies where curvature vanishes, but the exact critical weight exponent 1/8 deserves a reference to the precise theorem in Kenig-Ponce-Vega or Carbery-Kenig-Ziesler, or a short proof. That is not a credibility gap, but a referee should verify it. Same minor request for the bilinear estimates cited from [33,42]; they are standard, but only one is proved in the text. There are also the usual typos; nothing substantive.\n\nBottom line: this paper should go to peer review. It is the right kind of contribution, with a new space, an open endpoint, and a clean structural reduction, and the proof is detailed enough to be genuinely checkable. If the referee confirms the imported Strichartz inputs, I would accept.","headline":"First L^2-critical global well-posedness and scattering for 2D mZK via a new weighted space; proof looks sound, with a terse import of the key weighted Strichartz estimate.","tokens_in":24274,"tokens_out":5739,"would_cite":true,"duration_ms":54627,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35A01","35B30","35B40","35Q53"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that the 2D modified Zakharov-Kuznetsov equation is globally well-posed and scatters for small data in a new critical function space H^{0,1/4}(R^2), reaching the scaling-critical regularity that earlier H^s methods could…","keywords":["modified Zakharov-Kuznetsov equation","global well-posedness","scattering","scaling-critical regularity","anisotropic Sobolev spaces","atomic spaces","Strichartz estimates","ill-posedness"],"falsifier":"Compute the $L^{4}$ norm of |D_x|^{1/8}|D_y|^{1/8} $e^{{-t(∂_x^3+∂_y^3)}}$ f_N for a sequence f_N whose Fourier transform is concentrated in a unit ball around (ξ,η)=(N,$N^{{-1}}$), a point inside the weak-dispersion region near the axes. If the ratio of that $L^{4}$ norm to \\|f_N\\|_{$L^{2}$} is unbounded, Lemma 2.8 fails and Theorem 1.1 collapses; the proof requires this ratio to be uniformly bounded.","tokens_in":23108,"feed_emoji":"🌊","tokens_out":8127,"duration_ms":90000,"temperature":0.7,"pith_summary":"This paper tries to establish that the two-dimensional modified Zakharov-Kuznetsov (mZK) equation, a dispersive PDE describing nonlinear wave propagation, is globally well-posed and scatters for small initial data at the scaling-critical regularity. The authors' central move is a new two-parameter family of spaces $H^{{s,a}}$($R^{2}$) that scale exactly like the classical H^s spaces but assign extra cost to Fourier frequencies where the linear dispersion is weak, namely near the axes ξη=0. In the critical case s=0, a=1/4 the paper proves global well-posedness and scattering, and in the subcritical range s+a≥1/4, 0<a<1/4 it proves local well-posedness. It also shows these thresholds are sharp in the sense that no $C^{3}$ data-to-solution map exists below s+a=1/4. If correct, this is the first scaling-critical global result for the 2D mZK equation.","feed_headline":"Scaling-critical global result found for 2D modified ZK","feed_subtitle":"Anisotropic weights tame weak dispersion near the axes, yielding global solutions and scattering for small data.","key_machinery":"The central object is the two-parameter space $H^{{s,a}}$($R^{2}$) with anisotropic frequency weights |D_x|^a|D_y|^{-a} and |D_y|^a|D_x|^{-a}, chosen so that the space scales as H^s while penalizing concentration near the degenerate set ξη=0. The argument is carried by $U^{2}$ and $V^{2}$ atomic spaces adapted to the linear flow, the weighted $L^{4}$ Strichartz estimate \\||D_x|^{1/8}|D_y|^{1/8} $e^{{-t(∂_x^3+∂_y^3)}}$ f\\|_{$L^{4}$_{t,x,y}} ≲ \\|f\\|_{$L^{2}$}, the $L^{2}$ bilinear Strichartz estimates for nonresonant interactions, and the resulting multilinear estimate (3.5) that controls the Duhamel term in the Y norm. These estimates combine to make the fixed-point map a contraction for small data, and the Y-norm control gives the scattering limit.","core_discovery":"On the authors' own terms, the paper's discovery is that the failure of earlier H^s methods to reach the critical regularity s_c=0 is caused by insufficient control of frequency configurations with ξη≈0, where the linear group $e^{{-t(∂_x^3+∂_y^3)}}$ loses decay. The anisotropic spaces $H^{{s,a}}$, defined by the norms \\||D_x|^a|D_y|^{-a}u\\|_{H^s} + \\||D_y|^a|D_x|^{-a}u\\|_{H^s}, scale as H^s and penalize those configurations in a controlled way. Within these spaces, a contraction argument in atomic $U^{2}$-type resolution spaces yields a global solution for small data in $H^{{0,1/4}}$($R^{2}$) and the existence of a final state v_+ ∈ $H^{{0,1/4}}$($R^{2}$) with lim_{t→∞} $e^{{t(∂_x^3+∂_y^3)}}$v(t) = v_+, i.e., scattering. The paper further proves local well-posedness for s+a≥1/4 with a<1/4, and shows by explicit counterexamples that below s+a=1/4 the solution map is not $C^{3}$, and that no global-in-time fixed-point argument can work for a<1/4.","pith_inferences":["The same anisotropic-weighting strategy may transfer to the unmodified 2D Zakharov-Kuznetsov equation (k=1), where a gap between local well-posedness and the scaling-critical regularity s_c=-1/2 remains, or to other dispersive models whose symbols degenerate along submanifolds.","Because Theorem 1.1 gives scattering rather than only global existence, it may make it possible to construct wave operators at critical regularity, extending previous high-regularity wave-operator results.","A direct check of the weighted L^4 Strichartz estimate for data concentrated near the axes ξη=0 could serve as a test of the paper's critical lemma before adapting the method to neighboring equations.","The ill-posedness for a<1/4 indicates that any future proof reaching s=0 for related ZK-type equations will need to build in weights with exactly the critical homogeneity a=1/4, because the cubic interaction near the axes is the obstruction that must be controlled."],"forward_implications":["Small initial data in H^{0,1/4}(R^2) produce global solutions that converge, after multiplication by the linear flow, to a fixed final state as t→∞.","The local well-posedness range s+a≥1/4, 0<a<1/4 gives a family of subcritical well-posedness results that interpolate between the known endpoint a=0, s=1/4 and the new critical case.","The C^3-flow counterexamples show that no contraction-method well-posedness result below s+a=1/4 exists in the H^{s,a} scale, so the regularity threshold is sharp for this method.","The global-in-time fixed-point obstruction for a<1/4 explains why the anisotropic weight a=1/4 is necessary for scattering, not just sufficient."],"supporting_citations":[{"why":"supplies the weighted L4 Strichartz estimate (Lemma 2.8) on which the critical multilinear estimate depends","marker":"[30]"},{"why":"provides the L2 bilinear Strichartz estimates for nonresonant interactions and a related L4 estimate","marker":"[42]"},{"why":"establishes the endpoint s=1/4 in H^s and supplies estimates and counterexample techniques adapted in the subcritical and ill-posedness arguments","marker":"[33]"},{"why":"introduces the atomic U^p/V^p spaces, duality, interpolation, and transference principles used for the resolution spaces","marker":"[24]"},{"why":"provides the Dysthe-equation global well-posedness and scattering proof whose fixed-point structure is adapted here","marker":"[45]"},{"why":"gives the optimal L4 restriction theorem underlying the weighted L4 Strichartz estimate","marker":"[8]"},{"why":"supplies the modulation-localized version of the L4 estimate used in the dyadic proof","marker":"[20]"},{"why":"supplies the high-modulation estimate used inside the multilinear proof","marker":"[37]"},{"why":"introduces the linear change of variables that decouples the dispersion into ∂_x^3+∂_y^3","marker":"[23]"}],"fun_headline_variants":["2D modified ZK: global well-posedness and scattering","Anisotropic weights give global solutions for 2D modified ZK","Sharp global result for 2D modified Zakharov-Kuznetsov","Scattering for small data in 2D modified ZK equation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole theorem rests on the imported weighted $L^{4}$ Strichartz estimate (Lemma 2.8) and on the similarly imported bilinear Strichartz estimates; if any of these is false or needs a stronger weight than 1/8, the critical multilinear estimate (3.5) and with it Theorem 1.1 would not follow.","fun_headline_variants_meta":{"raw":{"variants":["2D modified ZK: global well-posedness and scattering","Anisotropic weights give global solutions for 2D modified ZK","Sharp global result for 2D modified Zakharov-Kuznetsov","Scattering for small data in 2D modified ZK equation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000225,"raw_usage":{"total_tokens":1471,"prompt_tokens":957,"completion_tokens":514,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":573,"completion_tokens_details":{"reasoning_tokens":436}},"tokens_in":573,"tokens_out":514,"duration_ms":5877,"temperature":1.0,"reasoning_tokens":436,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T10:48:53.463320+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the $L^{4}$ norm of |D_x|^{1/8}|D_y|^{1/8} $e^{{-t(∂_x^3+∂_y^3)}}$ f_N for a sequence f_N whose Fourier transform is concentrated in a unit ball around (ξ,η)=(N,$N^{{-1}}$), a point inside the weak-dispersion region near the axes. If the ratio of that $L^{4}$ norm to \\|f_N\\|_{$L^{2}$} is unbounded, Lemma 2.8 fails and Theorem 1.1 collapses; the proof requires this ratio to be uniformly bounded.","supporting_citations":[{"cited_title":"Kenig, Gustavo Ponce, and Luis Vega","cited_arxiv_id":null,"evidence_quote":"supplies the weighted L4 Strichartz estimate (Lemma 2.8) on which the critical multilinear estimate depends"},{"cited_title":"Bilinear Strichartz estimates for the Zakharov-Kuznetsov equa- tion and applications.Ann","cited_arxiv_id":null,"evidence_quote":"provides the L2 bilinear Strichartz estimates for nonresonant interactions and a related L4 estimate"},{"cited_title":"Well-posedness for the Cauchy problem of the modified Zakharov- Kuznetsov equation.Funkcial","cited_arxiv_id":null,"evidence_quote":"establishes the endpoint s=1/4 in H^s and supplies estimates and counterexample techniques adapted in the subcritical and ill-posedness arguments"},{"cited_title":"Well-posedness and scattering for the KP- II equation in a critical space.Ann","cited_arxiv_id":null,"evidence_quote":"introduces the atomic U^p/V^p spaces, duality, interpolation, and transference principles used for the resolution spaces"},{"cited_title":"Global well-posedness and scattering for the Dysthe equation inL 2(R2).J","cited_arxiv_id":null,"evidence_quote":"provides the Dysthe-equation global well-posedness and scattering proof whose fixed-point structure is adapted here"},{"cited_title":"Carbery, C","cited_arxiv_id":null,"evidence_quote":"gives the optimal L4 restriction theorem underlying the weighted L4 Strichartz estimate"},{"cited_title":"Birkh¨ auser/Springer, Basel, 2014","cited_arxiv_id":null,"evidence_quote":"supplies the high-modulation estimate used inside the multilinear proof"},{"cited_title":"The Fourier restriction norm method for the Zakharov- Kuznetsov equation.Discrete Contin","cited_arxiv_id":null,"evidence_quote":"introduces the linear change of variables that decouples the dispersion into ∂_x^3+∂_y^3"}],"review_version":1}