{"id":"76502ff2-f027-4c56-bf69-8a21ef8bb2bb","arxiv_id":"1909.01881","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For radial finite-energy solutions of the 3D defocusing energy-subcritical nonlinear wave equation, decay of only the inward part of the energy implies forward scattering, and slightly faster decay gives explicit rates.","lead":"This paper proves that radial solutions of a defocusing nonlinear wave equation scatter toward linear waves as time grows, provided only the inward-moving part of their initial energy decays. The result matters because it shows outward energy, even if large, cannot prevent scattering, and it gives explicit convergence rates.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The scattering contradiction in §4.1 depends on Theorem 1.2(c) from [12] in the positive time direction; that statement is only asserted as 'similar' and not proved here, so the central claim is conditional on an external result.","rationale":"I agree with the reader's weakest-assumption assessment: the central claim of Theorem 1.3(a) is conditional on the positive-time energy-location statement inherited from Theorem 1.2 of [12]. The internal estimates in Section 3 and the decomposition arguments in Section 4.2 appear coherent, and I found no independent contradiction within the present paper. The only serious soft spot is that the exact version of Theorem 1.2(c) used in §4.1 is not stated or proved in this manuscript, so the scattering conclusion has an unresolved external dependency. If [12] is correct and time-reversible, the dependency is benign; until then, CONDITIONAL is the appropriate verdict, which is why I recommend no change to the reader's verdict.","tokens_in":21738,"tokens_out":15249,"duration_ms":146477,"concrete_test":"Check the proof of Theorem 1.2 in arXiv:1808.08656 and verify that the negative-time result is proven for every finite-energy radial solution and that time reversal v(x,t)=u(x,−t) yields the positive-time localization with the same c=1/2 and β window. Concretely, derive the positive-time analogue of the third bullet of Theorem 1.2 directly from the energy flux formula (6) without citing Theorem 1.3; if the derivation succeeds, the §4.1 contradiction is sound, and if it fails, Theorem 1.3(a) is unproved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the opening of §4.1: if the solution did not scatter, equation (11) asserts lim_{t→∞} E(t;t/2,t−t^β)>0 for every β<2(p−2)/(p+1), 'according to part (c) of Theorem 1.2.' In this manuscript Theorem 1.2 states the negative-time energy-location result explicitly, but for the positive direction it only says 'the asymptotic behaviour ... is similar.' The proof of Theorem 1.3(a) therefore rests on a positive-time version of Theorem 1.2(c) that is neither stated precisely nor reproved. The subsequent weighted-Morawetz bounds and Proposition 4.1 are internally coherent; without the location theorem they do not contradict anything. Time reversal of the negative-time statement would supply the needed positive-time statement if [12] is correct for arbitrary finite-energy radial solutions, so the concern is not that Theorem 1.2 is false, but that the manuscript does not establish the exact input it invokes. This is a genuine external dependency, not a circularity, because [12] is a separate preprint.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the 3D defocusing energy-subcritical nonlinear wave equation ∂_t^2 u − Δu = −|u|^{p−1}u for 3≤p<5 in the radial finite-energy setting. The main result (Theorem 1.3) asserts that if the inward component of the initial energy satisfies a weighted decay condition with weight max{1,|x|^κ} and κ≥κ0(p)=(5−p)/(p+1), then the solution scatters forward in Ḣ^1×L^2; for κ>κ0(p), the solution also has finite L^pL^{2p}(R_+×R^3) norm, with explicit decay rates for the nonlinearity and for convergence to a free wave. The proof combines a new weighted Morawetz estimate (Section 3), giving decay of the inward energy and of the local outward energy, with a contradiction argument (Section 4.1) based on energy-distribution results imported from the author's earlier work [12]. The appendix constructs an example of a scattering solution with infinite L^{2(p−1)}L^{2(p−1)} norm, showing the phenomenon lies outside the critical-space scattering theory.","tokens_in":21915,"tokens_out":26090,"duration_ms":228372,"significance":"If correct, the result is a substantial advance: it replaces the usual assumptions of full weighted energy decay or compact support by a one-sided (inward) energy decay condition, it covers the endpoint κ=κ0, and it supplies quantitative rates for the L^pL^{2p} norm and the approach to the free wave. The weighted Morawetz estimates in Section 3 are derived in detail and appear internally consistent, and the geometric decomposition in Section 4.2 is coherent. The main caveat is the paper's reliance on a large body of unproved results from the author's preprint [12], in particular on a forward-time version of the energy-location theorem that is not stated in the manuscript; the central claim is therefore conditional on those external inputs.","major_comments":[{"comment":"The contradiction step assumes, with reference to the third bullet of Theorem 1.2, that if the solution did not scatter forward then lim_{t→∞} E(t;t/2,t−t^β)>0 for every β<2(p−2)/(p+1). However, Theorem 1.2 as stated in this paper proves the energy-location statement only in the negative time direction; the positive time direction is dismissed with the sentence 'The asymptotic behaviour in the positive time direction is similar.' Since Theorem 1.3(a) rests on this forward-time dichotomy, the manuscript must state the forward-time version explicitly and either prove it (for instance by applying Theorem 1.2 to the time-reversed solution, which is legitimate because Theorem 1.2 concerns all radial finite-energy solutions) or supply a precise theorem number in [12]. Without this, the scattering proof is conditional on an unstated external result.","section":"§4.1, equation (11)"},{"comment":"The paper imports substantial machinery from the author's preprint [12] without proof: the general energy flux formula involving the function ξ (Proposition 2.5), the triangle law (Proposition 2.7), the identity for E_-(t) (Proposition 2.8), the monotonicity and flux limits (Propositions 2.9–2.10), and the convergence of the characteristic traces (Proposition 2.11). These results are used directly in the proof of Theorem 1.3, yet [12] is an arXiv preprint and no proofs are given here. The manuscript should either include proofs of these statements or provide precise theorem numbers in [12] so that the reader can verify them; in its present form the central theorem is not self-contained.","section":"§2.3, Propositions 2.5–2.11"}],"minor_comments":[{"comment":"In the displayed estimate for the endpoint case, the change of variables r=t−τ gives τ=t−r, so the interval r∈[t−ct^{1−κ0}, t−t^β] corresponds to τ∈[t^β, ct^{1−κ0}], not τ∈[ct^{1−κ0}, t^β]. As written the integrals on the right have reversed limits and are negative for large t; the intended argument is clear, but the formulas should be corrected.","section":"§4.1, endpoint κ=κ0"},{"comment":"The abstract states the condition as κ≥κ0 without the upper bound κ<1, while Theorem 1.3 requires κ∈[κ0,1). The abstract (and the informal description in Section 1.2) should state the restriction κ<1 explicitly.","section":"Abstract and Theorem 1.3"},{"comment":"In the definition of the one-dimensional energy E(w,w_t), the term 'r^{r−1}' is a typo; it should be 'r^{p−1}'.","section":"Section 1.3"}],"recommendation":"major_revision","confidential_remarks":"The main theorem depends on the author's preprint [12] for the energy flux machinery and, in particular, for the forward-time energy-location dichotomy used in §4.1. This dependence is not merely a citation of a published result; [12] is an arXiv preprint, and the needed forward-time statement is not even stated in the present paper. I recommend that the editor ask the author to make the external inputs precise and, if possible, include complete proofs of the imported propositions. The new weighted Morawetz estimates in Section 3 are the paper's clear contribution and appear sound."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Ruipeng has a genuinely new scattering criterion here: for radial, energy-subcritical defocusing NLW in 3D, you only need the inward energy to decay at rate |x|^kappa with kappa >= (5-p)/(p+1), and you get forward scattering, including the endpoint kappa0 that was previously open. The paper also produces explicit decay for the L^p L^{2p} norm and for the convergence rate when kappa > kappa0. That is a real step beyond the earlier work in [12], which assumed weighted decay of the total energy.\n\nThe weighted Morawetz estimates in Section 3 looked sound to me; they are worked out in detail and the decay corollaries follow. The decomposition in Section 4.2 for the L^p L^{2p} bound is careful, and the pointwise estimate of Lemma 2.4 does a lot of work. The endpoint case uses the g+ approximation from [12] in a clean way.\n\nThe soft spot, and the one I would want addressed before betting on Theorem 1.3, is the contradiction argument in Section 4.1. It opens by invoking part (c) of Theorem 1.2 to say that if the solution does not scatter, energy remains in the annulus [t/2, t - t^beta] as t goes to infinity. But the manuscript's Theorem 1.2 states the energy-location result only for negative times; for positive times it just says 'the asymptotic behaviour is similar.' The proof needs that positive-time version, and it is neither stated precisely nor reproved. This is not circular--the inward-energy decay hypothesis is used earlier, in the weighted Morawetz--but it makes Theorem 1.3(a) conditional on a statement that lives in a separate preprint [12]. If [12] is solid, the rest of the scattering proof hangs together. If not, the whole thing falls. The author should fix this before publication, either by including the exact statement and proof of the positive-time analogue or by clearly citing a published version.\n\nMinor: the appendix example is interesting but a bit tangential; it demonstrates a limitation of critical-space scattering theory, not a step in the main proof. Not a problem, just a note.\n\nThis deserves a serious referee. The central result is new, the estimates are mostly self-contained, and the dependency on [12] is explicit enough to be checked. I would send it out and ask the author to resolve the positive-time Theorem 1.2(c) issue in revision. If I were working in scattering theory for NLW, I would cite it.","headline":"New scattering criterion for radial 3D NLW using only inward energy decay; proof is clean but leans at one load-bearing spot on an unproved positive-time statement from the author's earlier preprint.","tokens_in":22494,"tokens_out":2431,"would_cite":true,"duration_ms":23552,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35L71","35L05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that radial, finite-energy solutions of the defocusing nonlinear wave equation in R^3 scatter in positive time whenever the inward energy decays with weight max{1,|x|^kappa} for kappa at least (5-p)/(p+1), with an…","keywords":["defocusing nonlinear wave equation","radial solutions","scattering","inward energy","outward energy","weighted Morawetz estimate","LpL2p spacetime norm","energy distribution"],"falsifier":"For a fixed $p \\in [3,5)$ and $\\kappa \\geq \\kappa_0$, produce one radial finite-energy solution satisfying the inward decay condition whose $\\dot H^1 \\times L^2$ distance to every free wave fails to tend to zero; or, for the rate part, compute $\\|u\\|^p_{L^p L^{2p}([t,\\infty))}$ on a high-resolution radial simulation and find decay slower than $C t^{-\\frac{p+1}{p+3}(\\kappa-\\kappa_0)}$. A single such example would disprove Theorem 1.3.","tokens_in":21493,"feed_emoji":"🌊","tokens_out":8476,"duration_ms":79426,"temperature":0.7,"pith_summary":"This paper proves a new criterion for scattering of radial, finite-energy solutions of the three-dimensional defocusing semilinear wave equation $\\partial_t^2 u - \\Delta u = -|u|^{p-1}u$ with $3 \\leq p < 5$. The criterion is one-sided: only the inward part of the energy at time zero must decay, with weight $\\max\\{1,|x|^\\kappa\\}$ for $\\kappa \\geq (5-p)/(p+1)$, while the outward energy may be arbitrarily large and slowly decaying. Scattering means the solution approaches a free solution of the linear wave equation in $\\dot H^1 \\times L^2$ as $t \\to +\\infty$. When $\\kappa$ is strictly above the threshold, the paper additionally gives a power-law rate for the spacetime norm $\\|u\\|_{L^p L^{2p}(\\mathbb{R}_+ \\times \\mathbb{R}^3)}$ and for the convergence rate, tied directly to the excess decay $\\kappa - \\kappa_0(p)$. The significance is that finite energy alone has not been enough in this subcritical range, and the new assumption concerns only the energy moving inward, which matches the physical intuition that inward-moving energy is the only possible obstruction to forward scattering.","feed_headline":"Radial waves scatter as soon as their inward energy decays","feed_subtitle":"New proof covers the endpoint decay rate and gives explicit convergence rates for defocusing nonlinear wave equations.","key_machinery":"The argument runs through the radial reduction $w(r,t) = r u(|x|,t)$, which converts the 3D equation into a 1D wave equation with nonlinearity $-|w|^{p-1}w/r^{p-1}$ and preserves energy up to a constant. In this reduction, inward and outward energies are $E_\\mp(t) = \\pi \\int_0^\\infty (|w_r \\pm w_t|^2 + \\frac{2}{p+1}|w|^{p+1}/r^{p-1})\\, dr$. The load-bearing structural object is the energy flux formula, a boundary-integral identity with a distinguished $L^2$ function $\\xi(t)$ that records energy transferred from inward to outward motion through the origin; together with the triangle laws it expresses local energies as sums of flux integrals and spacetime integrals of $|w|^{p+1}/r^p$. Multiplying the flux identity by a weight $a(r+t)$ that grows like $(r+t)^\\kappa$ yields the weighted Morawetz estimates $\\int_0^\\infty t^\\kappa |\\xi(t)|^2\\, dt \\lesssim K$ and $\\int\\!\\int (r+t)^\\kappa |w|^{p+1}/r^p\\, dr\\, dt \\lesssim K$, which imply $E_-(t) \\lesssim K t^{-\\kappa}$ and the energy-in-cylinder bounds used later. The scattering proof then uses a prior energy-location theorem saying non-scattering energy must persist in an annulus just inside the light cone, and shows the weighted estimates make that annulus empty in the limit.","core_discovery":"The central claim, Theorem 1.3, is that radial initial data with finite energy scatter in positive time provided the inward energy satisfies $\\int_{\\mathbb{R}^3} \\max\\{1,|x|^\\kappa\\}(|\\nabla u_0\\cdot x/|x| + u_0/|x| + u_1|^2 + \\frac{2}{p+1}|u_0|^{p+1})\\, dx < \\infty$ with $\\kappa \\geq \\kappa_0(p) = (5-p)/(p+1)$. Here the integrand is the inward energy density, the combination of radial momentum and potential energy moving toward the origin. The proof shows that this decay of inward energy forces $E_-(t) \\leq C K t^{-\\kappa}$ for all later times via a weighted Morawetz estimate, and then combines this decay with a previously established energy-distribution theorem: if a solution failed to scatter, a fixed positive amount of energy would have to remain in the annulus $t/2 < |x| < t - t^\\beta$ for every $\\beta < 2(p-2)/(p+1)$. For $\\kappa > \\kappa_0$, the decay and annulus bounds contradict that persistence directly; at the endpoint $\\kappa = \\kappa_0$, an additional $L^2$ convergence statement along lightlike characteristics (Proposition 2.11) removes the remaining gap and still yields a contradiction. For $\\kappa > \\kappa_0$ the same estimates prove $\\|u\\|_{L^p L^{2p}}^p$ on $[t,\\infty)$ is bounded by $C t^{-\\frac{p+1}{p+3}(\\kappa-\\kappa_0)}$, and the distance in $\\dot H^1 \\times L^2$ from the evolved solution to the asymptotic free wave obeys the same bound.","pith_inferences":["If the energy-location theorem could be sharpened below $\\kappa_0$, the same weighted Morawetz estimates would likely push the scattering threshold lower: for $3 < p < 5$ the paper's own Remark 4.5 already shows the medium-radius contribution decays when $\\kappa$ is slightly below $\\kappa_0$.","The mechanism isolates the light-cone annulus $t/2 < |x| < t - t^\\beta$ as the only possible obstruction to scattering, so a numerical test of near-threshold radial data should look specifically at whether energy accumulates there.","The flux identity with a growing weight is not obviously specific to the power nonlinearity; similar inward-energy decay conditions may imply scattering for other defocusing nonlinearities or other dimensions where a radial 1D reduction and a $\\xi$-flux identity are available.","The explicit rate suggests an optimal-scattering-rate question: whether the exponent $\\frac{p+1}{p+3}(\\kappa-\\kappa_0)$ is the true worst-case decay for this class, or merely an upper bound."],"forward_implications":["If the inward energy decays at the endpoint rate $\\kappa_0 = (5-p)/(p+1)$, every radial finite-energy solution scatters in forward time; no assumption on the size or decay of outward energy is needed beyond finiteness of total energy.","For any $\\kappa > \\kappa_0$, the nonlinear forcing satisfies $\\| |u|^{p-1}u \\|_{L^1 L^2([t,\\infty)\\times \\mathbb{R}^3)} \\leq C t^{-\\frac{p+1}{p+3}(\\kappa-\\kappa_0)}$, so the solution converges to a free wave at that explicit polynomial rate.","By time reversal, the same statement holds backward in time if the outward energy decays with the same weight and the sign of the initial velocity is flipped.","The appendix constructs a scattering solution whose $L^{2(p-1)} L^{2(p-1)}$ norm in the critical-Sobolev sense is infinite, so the new criterion covers scattering phenomena invisible to the critical-space scattering theory."],"supporting_citations":[{"why":"Supplies the inward/outward energy flux formalism, the triangle laws, the L2 convergence statement along characteristics, and the non-scattering energy-location theorem that the contradiction step assumes.","marker":"[12]"},{"why":"Supplies the Strichartz estimates used in local well-posedness and in converting the finite LpL2p norm into an explicit convergence rate toward a free wave.","marker":"[5]"},{"why":"Supplies the radial pointwise estimate for H1 functions that is recalled as Lemma 2.3 and used in the estimates and in the appendix example.","marker":"[8]"}],"fun_headline_variants":["Scattering guaranteed by inward energy decay for radial waves","Inward energy decay suffices: radial NLW scattering proven","Radial wave scattering from inward decay, with rates","Inward decay forces scattering in defocusing NLW","Endpoint decay rate included: scattering proof for radial data"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The contradiction step assumes as a black box the earlier theorem that a non-scattering radial solution must keep positive energy in the annulus $t/2 < |x| < t - t^\\beta$ for every $\\beta < 2(p-2)/(p+1)$; if that energy-location result fails or does not apply forward in time, the scattering conclusion collapses.","fun_headline_variants_meta":{"raw":{"variants":["Scattering guaranteed by inward energy decay for radial waves","Inward energy decay suffices: radial NLW scattering proven","Radial wave scattering from inward decay, with rates","Inward decay forces scattering in defocusing NLW","Endpoint decay rate included: scattering proof for radial data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000598,"raw_usage":{"total_tokens":2923,"prompt_tokens":1196,"completion_tokens":1727,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":812,"completion_tokens_details":{"reasoning_tokens":1648}},"tokens_in":812,"tokens_out":1727,"duration_ms":11118,"temperature":1.0,"reasoning_tokens":1648,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:06:12.648958+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For a fixed $p \\in [3,5)$ and $\\kappa \\geq \\kappa_0$, produce one radial finite-energy solution satisfying the inward decay condition whose $\\dot H^1 \\times L^2$ distance to every free wave fails to tend to zero; or, for the rate part, compute $\\|u\\|^p_{L^p L^{2p}([t,\\infty))}$ on a high-resolution radial simulation and find decay slower than $C t^{-\\frac{p+1}{p+3}(\\kappa-\\kappa_0)}$. A single such example would disprove Theorem 1.3.","supporting_citations":[{"cited_title":"Energy Distribution of Radial Solutions to Energy Subcritical Wave Equation with an Application on Scattering Theory","cited_arxiv_id":"1808.08656","evidence_quote":"Supplies the inward/outward energy flux formalism, the triangle laws, the L2 convergence statement along characteristics, and the non-scattering energy-location theorem that the contradiction step assumes."},{"cited_title":"Generalized Strichartz inequality for the wave equation","cited_arxiv_id":null,"evidence_quote":"Supplies the Strichartz estimates used in local well-posedness and in converting the finite LpL2p norm into an explicit convergence rate toward a free wave."},{"cited_title":"Nondispersive radial solutions to energy supercritical non-linear wave equations, with applications","cited_arxiv_id":null,"evidence_quote":"Supplies the radial pointwise estimate for H1 functions that is recalled as Lemma 2.3 and used in the estimates and in the appendix example."}],"review_version":1}