{"id":"e0d59a59-2079-4496-a481-918b00450450","arxiv_id":"1908.00743","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For the 2D radial focusing INLS with 0 < b < 1, solutions below the ground state are shown to scatter, via a proof avoiding concentration compactness.","lead":"This paper proves a scattering theorem for a two-dimensional inhomogeneous nonlinear Schrödinger equation, extending known results to a wider range of parameters. It uses a new proof approach that avoids concentration compactness, based on virial estimates and a new scattering criterion.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reader's S(H^sp)-tail objection misreads the endpoint; the actual gap in Lemma 3.1 is that (3.15) is imposed at T>T0 but used to control I2=[T0-ε^{-θ},T0].","rationale":"The paper's central theorem is plausible and largely follows the Arora-Dodson-Murphy strategy, with the Morawetz estimate providing the required spacetime bound and the scattering criterion supplying the endpoint. The reader's weakest_assumption is not the real problem: the endpoint pair in S(Ḣ^{sp}) decays in time for the linear flow, so the linear tail can be made small on [T0,∞). The genuine concern is an indexing error in the proof of Lemma 3.1: smallness at T>T0 does not control the interval I2 ending at T0. This is a localized, fixable flaw rather than a fatal one; the intended argument is clear. Hence I would keep the CONDITIONAL verdict, but for a different reason than the one stated in the reader's report. The concrete test above settles whether the fix works.","tokens_in":9741,"tokens_out":32285,"duration_ms":294189,"concrete_test":"Rewrite the proof of Lemma 3.1 with T0 chosen as a time from the liminf in (3.10), large enough that ‖e^{itΔ}u0‖_{S(Ḣ^{sp},[T0,∞))}≪1, and set (3.15) with T=T0. Then repeat the argument: from |d/dt∫χ_R|u|^2|≲1/R and the initial smallness at T0, obtain ‖χ_Ru‖_{L∞_t L^2_x(I2)}≲ε for I2=[T0-ε^{-θ},T0]. Verify that this, combined with the radial Sobolev interpolation, gives ‖F2‖_{S(Ḣ^{sp})}≪1 as claimed. If the chain goes through, the scattering criterion is repaired; if not, the criterion lacks a proof.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The concern identified in the reader's verdict does not survive scrutiny. In Lemma 3.1, the term ‖e^{itΔ}u0‖_{S(Ḣ^{sp},[T0,∞))} includes the pair (q=∞, r=2/(1-sp)), but this is the L^∞_t L^r_x norm, not a conserved H^{sp} norm; for free evolution it decays like |t|^{-sp} on Schwartz data, and the H^{sp}→L^r embedding plus a Schwartz approximation makes the tail norm tend to 0 for any H^{sp} initial data. So the linear smallness step is not the gap. The real gap in the proof as written is in the F2 estimate. After fixing T0 for the linear tail, the proof invokes (3.10) to choose T>T0 satisfying (3.15): ∫χ_R|u(T)|^2<ε^2. But F2 is integrated over I2=[T0-ε^{-θ},T0], and the conclusion ‖χ_Ru‖_{L∞_t L^2_x(I2)}≤ε requires small mass at the right endpoint T0, not at a later time T. The time-derivative bound |d/dt∫χ_R|u|^2|≲1/R only propagates smallness from an endpoint of I2; with T outside I2 it does not control I2. This is fixable by choosing T0 itself from the liminf sequence in (3.10) and large enough for the linear-tail bound, then taking T=T0 in (3.15). Absent that correction, the proof of the scattering criterion has a genuine logical gap.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the 2D radial focusing inhomogeneous nonlinear Schrödinger equation i∂_t u + Δu + |x|^{-b}|u|^p u = 0 with 0 < b < 1 and 2 - b < p < ∞. The main result, Theorem 1.1, asserts that for radial initial data below the ground state threshold (in the sense of a mass-energy product), the solution is global and scatters in H^1. The authors propose a proof that avoids the concentration-compactness method, relying instead on a new scattering criterion (Lemma 3.1), radial Sobolev embedding, and a Virial/Morawetz estimate (Proposition 4.1). The paper is an extension of Farah and Guzmán's scattering result from 0 < b < 2/3 to the full range 0 < b < 1.","tokens_in":10050,"tokens_out":15344,"duration_ms":143815,"significance":"If the proof is completed, the result is a meaningful extension of the known scattering theory for the focusing INLS in two dimensions, and it provides an independent route that bypasses concentration compactness. The explicit scattering criterion in Lemma 3.1 and the Morawetz estimate are potentially useful tools for related problems. The paper is self-contained relative to cited well-posedness and interpolation results, and the main argument has no fitted free parameters. However, the proof as written contains a logical gap in the scattering criterion and an omitted proof of a coercivity lemma, so the significance can only be assessed after these issues are resolved.","major_comments":[{"comment":"There is a logical gap in the treatment of the term F2. The proof chooses T > T0 satisfying (3.15), namely ∫ χ_R |u(T)|^2 < ε^2, but then uses this smallness to control the L∞_t L^2_x norm of χ_R u on the time interval I2 = [T0 - ε^{-θ}, T0]. Since I2 lies to the left of T0 and T is an arbitrary time larger than T0, smallness at T does not propagate to I2 from the derivative bound |d/dt ∫ χ_R |u|^2| ≲ 1/R unless T actually belongs to I2 or the distance from T to I2 is controlled by ε R. As written, the proof is invalid. The gap is fixable by choosing T0 itself from the liminf sequence in (3.10), sufficiently large for the linear-tail estimate, and then setting T = T0 in (3.15); with that choice the smallness at the right endpoint of I2 propagates over I2. The authors should make this choice explicit.","section":"Section 3, proof of Lemma 3.1, estimation of F2 (around (3.14)–(3.15))"},{"comment":"Lemma 4.2 (coercivity on balls) is stated without proof; the text says 'We refer to [1] for an analogous proof.' This lemma is load-bearing: it is used in the proof of Proposition 4.1 to obtain (4.24), which is essential for the Morawetz estimate. Since [1] is described as 'to appear' and the inhomogeneous nonlinearity |x|^{-b} is not treated in [1], the authors should either provide a complete proof or a detailed statement of the analogous argument and explain how it adapts to the present setting. Without this, the proof of Proposition 4.1 is incomplete.","section":"Section 4, Lemma 4.2"}],"minor_comments":[{"comment":"There are numerous typographical errors, including 'SCA TTERING' in the title, 'W e' and 'Key W ords' in the abstract, and 'Yangida' in place of 'Yanagida' in reference [19].","section":"Throughout"},{"comment":"The notation M(u0) = E(u0) = E reuses E both as the energy functional and as a generic positive constant; this is confusing. A different symbol, for example L, should be used for the a priori bound in (3.9).","section":"Section 3, Lemma 3.1 statement"},{"comment":"The expression '|∂t ∫_{I2} χ_R|u|^2 ds|' appears to be a typo; the intended quantity is the absolute value of the time derivative of ∫ χ_R |u|^2 dx. Please correct the notation.","section":"Section 3, proof of Lemma 3.1, estimation of F2"},{"comment":"The Hölder exponents in the display after the dispersive estimate, such as L^{4/(2-p^-)}, contain what appear to be OCR artifacts or typos, especially in the case p > 2 where the exponent 2 - p^- becomes negative. The authors should rewrite this estimate with a valid triple of exponents that covers the full range 2 - b < p < ∞.","section":"Section 3, estimation of F1"},{"comment":"The informal definitions of a^- and a^+ ('a− is a fixed number and slightly smaller than a') should be made precise, since the subsequent estimates rely on these quantities.","section":"Section 2, definition of admissible pairs"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a relevant open range of parameters and presents a promising alternative to concentration compactness. The main concern is that the proof of the scattering criterion, Lemma 3.1, contains a genuine gap in the F2 estimate, and Lemma 4.2 is delegated to an unpublished reference. Both issues appear fixable without changing the overall strategy. I recommend major revision and suggest that the editor verify the availability of reference [1] before final acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a worthy incremental paper with a fixable gap in Lemma 3.1. The main theorem, scattering for radial 2d focusing INLS below ground state in the full range 0<b<1, extends Farah–Guzman's b<2/3 and does so without concentration compactness. That's a real contribution.\n\nThe cleanest part is the scattering criterion. It is the right kind of tool for this area, and the proof through the Duhamel split F1/F2 is mostly standard. The paper is also reasonably self-contained: the nonlinear estimates in Lemma 2.3 are stated with proof sketches, and the global well-posedness input is quoted properly.\n\nThe soft spots, in order of importance.\n\nFirst, the report I saw claims the linear tail ||e^{itΔ}u0||_{S(Ḣ^{sp},[T0,∞))} cannot be small because the S norm includes a conserved H^{sp} piece. I think that objection is off. The endpoint pair is (q=∞, r=2/(1−sp)), which is L∞_t L^r_x, not L∞_t H^{sp}_x. For fixed u0∈H^{sp}, that tail tends to zero by dispersive decay after a Schwartz approximation. So this is not a genuine gap.\n\nThe genuine gap is in the F2 estimate. After fixing T0, the proof chooses T>T0 satisfying (3.15). Then it uses the derivative bound |d/dt∫χ_R|u|^2|≲1/R to conclude smallness of ||χ_R u||_{L∞_t L^2_x(I2)} on I2=[T0−ε^{-θ},T0]. But the smallness is at T, which is outside I2 and whose distance from I2 is uncontrolled. The derivative bound only propagates from an endpoint of the interval. This is fixable by choosing T0 itself from the liminf sequence in (3.10) and setting T=T0 (or an equivalent rearrangement), but as written it is a real logical gap. It should be fixed before publication.\n\nSecond, the paper cites Campos [2] without saying how Theorem 1.1 relates to it. If Campos already proved the same range, the novelty is only the method; if not, the authors should say so explicitly. This needs clarification.\n\nThird, Lemma 4.2 is deferred to [1] with \"analogous proof.\" For the inhomogeneous nonlinearity the energy functional is different enough that a precise reference or a short proof would be better. Minor.\n\nOverall: the result is likely true and the method is worth having. The paper deserves a serious referee. My recommendation is to send it out, with the understanding that the F2/T0 issue must be addressed and the Campos comparison clarified.","headline":"Worth a serious referee: a real but incremental extension of the scattering range for 2d radial INLS, with a fixable gap in Lemma 3.1 and an unclear relation to Campos [2].","tokens_in":10659,"tokens_out":10591,"would_cite":true,"duration_ms":94361,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35P25","35Q55","47J35"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that radial solutions below the ground-state threshold to the 2d focusing inhomogeneous nonlinear Schrödinger equation scatter for all 0<b<1, extending the known range 0<b<2/3, and gives a proof that avoids concentration…","keywords":["scattering theory","inhomogeneous nonlinear Schrödinger equation","radial solutions","ground state threshold","concentration compactness","Morawetz estimates","Strichartz estimates","two-dimensional NLS"],"falsifier":"Take any nonzero radial $u_0\\in H^1$ and evaluate $\\|e^{it\\Delta}u_0\\|_{S(\\dot H^{s_p},[T_0,\\infty))}$ directly: the endpoint admissible pair $(q,r)=(\\infty,2/(1-s_p))$ is included in the definition, and at that pair the norm equals $\\|u_0\\|_{\\dot H^{s_p}}$, independent of $T_0$, so the asserted smallness as $T_0\\to\\infty$ fails exactly as written; a corrected non-endpoint Strichartz norm would be needed for the criterion's proof to succeed.","tokens_in":9485,"feed_emoji":"🌀","tokens_out":11531,"duration_ms":103432,"temperature":0.7,"pith_summary":"The paper establishes that for the two-dimensional focusing inhomogeneous nonlinear Schrödinger equation $i\\partial_t u+\\Delta u+|x|^{-b}|u|^p u=0$, every radial solution whose mass-energy product and product norm lie below the ground-state values is globally well-posed and scatters, for the full parameter range $0<b<1$ and $2-b<p<\\infty$. This extends an earlier scattering theorem for the same equation that had been proved only for $0<b<2/3$. The new proof works through a scattering criterion together with a Virial/Morawetz estimate, and it avoids concentration compactness entirely. If correct, it completes the below-threshold scattering picture for radial solutions of this equation and demonstrates a more direct route to such results.","feed_headline":"Scattering below ground state for 2d focusing INLS extends to all b<1","feed_subtitle":"A Morawetz-based argument replaces concentration compactness and closes the range from b<2/3 to b<1 for radial data.","key_machinery":"The load-bearing object is the Strichartz norm $S(\\dot H^{s_p},I)$, defined as the supremum of $L^q_t L^r_x$ norms over all $\\dot H^{s_p}$-admissible pairs, and the scattering criterion (Lemma 3.1) is phrased in that norm. The proof follows a modified scheme from [1]: it splits the Duhamel term at a large time $T_0$ into a far past piece, controlled by dispersion and the assumed bound $\\int_0^T\\int |x|^{-b}|u|^{p+2}\\lesssim T^\\alpha$, and a recent piece, controlled by radial Sobolev embedding together with smallness of the mass inside a ball. The Morawetz estimate comes from the Virial/Morawetz identity (4.18)-(4.19) applied to a radial weight $a(x)$ that grows like $|x|^2$ inside a ball and like $|x|$ outside, giving $\\int_0^T\\int_{\\mathbb{R}^2}|x|^{-b}|u|^{p+2}\\lesssim R+T/R^\\alpha$; the choice $R=T^{1/(1+\\alpha)}$ then yields the spacetime bound required by the scattering criterion.","core_discovery":"The central claim is Theorem 1.1: under $0<b<1$ and $2-b<p<\\infty$, if $u_0$ is radial and satisfies $M(u_0)^{1-s_p}E(u_0)^{s_p}<M(Q)^{1-s_p}E(Q)^{s_p}$ and $\\|u_0\\|_{L^2}^{1-s_p}\\|\\nabla u_0\\|_{L^2}^{s_p}<\\|Q\\|_{L^2}^{1-s_p}\\|\\nabla Q\\|_{L^2}^{s_p}$, then the solution $u$ is globally well-posed and scatters, meaning there exist $u_\\pm\\in H^1$ with $\\|u(t)-e^{it\\Delta}u_\\pm\\|_{H^1}\\to 0$ as $t\\to\\pm\\infty$. The authors prove this by showing that such solutions satisfy a new scattering criterion: a uniform $H^1$ bound together with the weighted spacetime estimate $\\int_0^T\\int_{\\mathbb{R}^2}|x|^{-b}|u|^{p+2}dxdt\\lesssim T^\\beta$, for some $0<\\beta<1$, implies scattering forward in time. This provides the scattering conclusion for the full range $0<b<1$, reaching the boundary where the local well-posedness theory from [5] is available, and it removes the need for concentration compactness.","pith_inferences":["The endpoint issue in Lemma 3.1 suggests the intended norm may be a non-endpoint Strichartz norm; if so, the proof could likely be repaired by a standard limiting argument, but the written smallness assertion is not justified as it stands.","The same combination of a scattering criterion, radial Sobolev embedding, and a convex-weight Morawetz estimate may extend to other dimensions or to other inhomogeneous nonlinearities with decaying weights, wherever the relevant coercivity and radial decay estimates hold.","A numerical simulation of radial solutions near the threshold with $b$ close to $1$ could test the predicted power-law decay of $\\int_0^T\\int |x|^{-b}|u|^{p+2}$ independently of the proof, providing evidence on whether the Morawetz estimate is the right mechanism."],"forward_implications":["Scattering below the ground state for the 2d radial focusing INLS holds for every $b\\in(0,1)$ and $p\\in(2-b,\\infty)$, not just for the previously covered range $b<2/3$.","The Virial/Morawetz estimate $\\int_0^T\\int |x|^{-b}|u|^{p+2}\\lesssim T^\\beta$ supplies the long-time smallness of the nonlinear term directly, giving an alternate route to conclusions previously reached by concentration compactness.","The scattering criterion in Lemma 3.1 is a reusable tool: any radial solution with a uniform $H^1$ bound and the stated power-law decay of the weighted nonlinear integral will scatter.","The theorem confirms that the below-ground-state threshold expressed by $M(u)^{1-s_p}E(u)^{s_p}<M(Q)^{1-s_p}E(Q)^{s_p}$ and the product-norm condition is sufficient for scattering for the full allowed range of $b$."],"supporting_citations":[{"why":"Supplies the modified approach that the proof adapts: the scattering-criterion structure and the radial Sobolev embedding used to control localized mass.","marker":"[1]"},{"why":"States the earlier scattering theorem for the same equation under the same threshold, valid only for $0<b<2/3$, which the present result extends.","marker":"[10]"},{"why":"Provides the local well-posedness theory in $H^1$ for $0<b<1$ and $2-b<p<\\infty$, used by the global well-posedness argument.","marker":"[5]"},{"why":"Provides the sharp Gagliardo-Nirenberg inequality with the ground state $Q$ in the constant, used to enforce coercivity below the threshold.","marker":"[8]"},{"why":"Supplies the Strichartz estimates used to control linear and Duhamel evolutions in the $S(\\dot H^{s_p})$ norm.","marker":"[17]"},{"why":"Supplies the radial Sobolev embedding used to convert small mass concentration in a ball into small $L^r_x$ norms in the scattering criterion.","marker":"[18]"}],"fun_headline_variants":["Scattering for 2D radial INLS extended to all b<1","Morawetz proof closes scattering range for radial INLS","Radial focusing INLS: scattering for every b<1, no compactness","New proof for 2D INLS scattering below ground state","Concentration compactness bypassed in 2D radial INLS scattering"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof assumes that for sufficiently large $T_0$, the linear evolution $e^{it\\Delta}u_0$ has small norm in the Strichartz space used by the scattering criterion; as written that space includes an endpoint admissible pair at which the norm is conserved rather than decaying, so the statement needs a non-endpoint version that is not supplied.","fun_headline_variants_meta":{"raw":{"variants":["Scattering for 2D radial INLS extended to all b<1","Morawetz proof closes scattering range for radial INLS","Radial focusing INLS: scattering for every b<1, no compactness","New proof for 2D INLS scattering below ground state","Concentration compactness bypassed in 2D radial INLS scattering"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000217,"raw_usage":{"total_tokens":1430,"prompt_tokens":935,"completion_tokens":495,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":551,"completion_tokens_details":{"reasoning_tokens":401}},"tokens_in":551,"tokens_out":495,"duration_ms":5314,"temperature":1.0,"reasoning_tokens":401,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:35:26.411294+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take any nonzero radial $u_0\\in H^1$ and evaluate $\\|e^{it\\Delta}u_0\\|_{S(\\dot H^{s_p},[T_0,\\infty))}$ directly: the endpoint admissible pair $(q,r)=(\\infty,2/(1-s_p))$ is included in the definition, and at that pair the norm equals $\\|u_0\\|_{\\dot H^{s_p}}$, independent of $T_0$, so the asserted smallness as $T_0\\to\\infty$ fails exactly as written; a corrected non-endpoint Strichartz norm would be needed for the criterion's proof to succeed.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the modified approach that the proof adapts: the scattering-criterion structure and the radial Sobolev embedding used to control localized mass."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"States the earlier scattering theorem for the same equation under the same threshold, valid only for $0<b<2/3$, which the present result extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the sharp Gagliardo-Nirenberg inequality with the ground state $Q$ in the constant, used to enforce coercivity below the threshold."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the radial Sobolev embedding used to convert small mass concentration in a ball into small $L^r_x$ norms in the scattering criterion."}],"review_version":1}