{"id":"bef9fe90-6915-451e-a15a-a4afc4fdc760","arxiv_id":"2507.11259","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"For slightly mass-supercritical NLS in 1 to 10 dimensions, no unstable high-energy eigenmodes exist for the linearized self-similar profile operator, which settles its asymptotic stability.","lead":"This paper proves that certain \"mode\" oscillations around a self-similar blowup solution of the slightly supercritical nonlinear Schrödinger equation cannot grow, in dimensions 1 through 10. It completes a stability program for such blowup and also pins down the full spectrum of the linearized operator at the mass-critical ground state.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof of both main theorems hinges on Proposition 2.1(2), a coercivity statement 'proven with numerical help' in [23,11,30] and not reproduced; if that input fails, the rigidity conclusions in Theorems 1.1 and 1.2 would fail.","rationale":"I read the paper in good faith and followed the proof structure. Conditional on Proposition 2.1(2), the main argument is coherent: the linear Liouville reduction is clear, the modulation construction fixes the needed orthogonality, the energy and Virial identities provide monotonicity, and the a priori estimates in Appendix B supply the regularity and localization used later. I checked the principal algebraic steps, including the truncated Virial computation (3.15)-(3.17), the energy monotonicity (3.45)-(3.48), the weighted estimates (3.52)-(3.59), and the resolvent bounds in Appendix B; I did not find an internal algebraic inconsistency in those parts. The remaining risk is concentrated in the numerically verified coercivity statement. Proposition 2.1(2) is load-bearing for both theorems, and the manuscript does not ship the numerical verification or an independent proof. The reader's weakest assumption identifies exactly this point, and I agree. Because the identified concern does not move the verdict beyond the reader's conditional acceptance, I recommend no change to the reader's verdict. If the numerical coercivity is reproduced or made available with rigorous error bounds, and if the companion preprints [17,18] are validated, the central claim would be supported.","tokens_in":33086,"tokens_out":13754,"duration_ms":166892,"concrete_test":"Run a certified numerical computation of the minimal Rayleigh quotients for the quadratic forms (L1u,u) and (L2w,w) on the orthogonal complement (2.9), for each d = 1,...,10, using a high-order spectral or finite-element discretization with the stated weight exponent \\mu_d. If the quotients are bounded below by the claimed positive constants, ideally with rigorous interval arithmetic or by reproducing the numerical verification behind [23, Theorem 1.1], [11], and [30], then the concern is settled; if any dimension gives a negative or smaller quotient, Proposition 2.1(2) is false and Theorems 1.1 and 1.2 fail in that dimension.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The single load-bearing assumption is Proposition 2.1(2): (L1u,u)+(L2w,w) is bounded below by ||u||_{\\dot H^1}^2 + ||<r>^{-\\mu_d}u||^2 + ||w||_{\\dot H^1}^2 + ||<r>^{-\\mu_d}w||^2 under the orthogonality conditions (2.9), for 1 <= d <= 10. This is the only input carrying the full dimension range d <= 10 and the only numerically assisted step. In Theorem 1.1, Step 3(iii) uses (2.11) to obtain the lower bound on the time derivative of the truncated Virial identity, which forces \\tilde u = \\tilde w = 0; without this coercivity, nonzero eigenvalues in R \\ {±1,0} and endpoint resonances at ±1 are not excluded. In Theorem 1.2, the monotonicity argument for the energy (Step 2, estimate (3.48)) and the weighted coercivity (Step 3, estimate (3.53)) both invoke the same inequality; if (2.11) fails, the dissipative b-norm term can have the wrong sign and the linear Liouville argument yields no contradiction. The paper states that Proposition 2.1(2) is 'proven with numerical help' in [23,11,30], and Appendix A merely restates those cited results: no code, no rigorous interval bounds, and no verification script is included. Since Theorem 1.3 is assembled from Theorem 1.2 together with the companion papers [17,18], an unvalidated numerical input is a genuine correctness risk.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper is the second part of a study on mode stability for the linearized operator around self-similar blowup profiles of slightly supercritical NLS. It proves high-energy mode stability for the linearized operator H_b on (Ḣ^σ)^2 for 1 ≤ d ≤ 10 (Theorem 1.2) and, as a byproduct, a full spectral characterization of the linearized operator H_0 around the mass-critical ground state (Theorem 1.1). The proofs use a linear Liouville argument: an eigenfunction is viewed as a stationary solution of a linear evolution equation, and modulation plus truncated energy/Virial estimates produce rigidity. Theorem 1.3 then derives asymptotic stability of the self-similar blowup by combining the two spectral theorems with the author's companion paper [17] and previous preprint [18]. The main technical input is Proposition 2.1(2), a coercivity estimate for the Virial commutator that is stated to be 'proven with numerical help' in [23,11,30] and is not reproduced here.","tokens_in":33438,"tokens_out":15754,"duration_ms":183495,"significance":"If the main results are correct, they complete a long-standing program on the asymptotic stability of self-similar blowup for slightly mass-supercritical NLS in dimensions 1 ≤ d ≤ 10, answering a question raised in [1,24]. The spectral characterization of H_0 for 2 ≤ d ≤ 10 appears to be new. The linear Liouville strategy is an interesting adaptation of Martel–Merle's rigidity method to a non-self-adjoint spectral problem, and it may be transferable to other linearized operators. The significance is partly conditional, however: the full range d ≤ 10 and the rigidity conclusions in Theorems 1.1 and 1.2 rest on the numerically assisted coercivity of Proposition 2.1(2), and Theorem 1.3 depends on the author's unpublished companion works [17,18].","major_comments":[{"comment":"The proof of both Theorem 1.1 (Step 3(iii) of the proof in §3.1) and Theorem 1.2 (Step 2, estimate (3.48), and Step 3, estimate (3.53)) uses the coercivity (L1u,u)+(L2w,w) ≳ ‖u‖²_{Ḣ¹} + ‖⟨r⟩^{-µ_d}u‖² + ‖w‖²_{Ḣ¹} + ‖⟨r⟩^{-µ_d}w‖² under the orthogonality conditions (2.9). This is the only input that carries the full dimension range 1 ≤ d ≤ 10, and the paper states only that it is 'proven with numerical help' in [23,11,30], with Appendix A merely reformulating the cited results. No code, interval bounds, or verification script is included. If this inequality fails in any dimension, the rigidity conclusions that force the eigenfunction to vanish in Theorems 1.1 and 1.2 would collapse. This is a load-bearing correctness risk. Please either include a complete, reproducible verification of the coercivity (e.g., computer-assisted interval arithmetic), or state explicitly that Theorems 1.1 and 1.2 are conditional on an externally verified numerical–analytic property, with precise theorem numbers from [23,11,30].","section":"§2.1, Proposition 2.1(2), Eq. (2.11)"},{"comment":"The proof of Theorem 1.2 relies on Lemma 2.5 for regularity, smoothing, and delocalization of eigenfunctions; its proof in Appendix B invokes [18, Lemma 2.4, Proposition 2.2, Lemma 2.11] and [17, Proposition 2.4 and Proposition 4.13] as black boxes. Since [18] is an unpublished preprint and [17] is a companion paper not included in this submission, the a priori estimates used in the energy and weighted-energy steps (e.g., the bound ‖w̃‖_{Ḣ²} used around (3.55)) are not verifiable from the present manuscript. Please either reproduce the needed statements with proofs, or clearly delimit the argument as conditional on [17,18] and state the exact assumptions imported from those works.","section":"§3.2 and Appendix B, Lemma 2.5"},{"comment":"Theorem 1.3, the advertised asymptotic stability of the self-similar blowup, is not proved in this paper but assembled from Theorem 1.2, [17, Theorem 1.1], and [18, Theorem 1.4 and 1.5]. Because [17] is a companion preprint and [18] is the author's earlier preprint, the status of Theorem 1.3 as a theorem of the present paper should be clarified: either state it as a corollary conditional on the full mode stability assumption [18, Assumption 1.2] being verified in those works, or include the necessary parts of [17,18] so that the conclusion is self-contained.","section":"Theorem 1.3 and §1.2"}],"minor_comments":[{"comment":"The sentence 'we prove Theorem and 1.1 and Theorem 1.2' contains a typo and should read 'we prove Theorems 1.1 and 1.2'.","section":"§3, opening line"},{"comment":"In the display following item (a), the right-hand side should contain squares on the norms: the conclusion should be ∂ₜI(t) ≳ ‖ũ₀‖²_{Ḣ¹} + ‖⟨x⟩^{-μ_d}ũ₀‖²_{L²} + ‖w̃₀‖²_{Ḣ¹} + ‖⟨x⟩^{-μ_d}w̃₀‖²_{L²}; as written the inequality is dimensionally inconsistent with the use of Proposition 2.1(2).","section":"§3.1, Step 3(iii)"},{"comment":"In the definition of the auxiliary function w̃, the coefficient is given as β = (u,ρ)_{L²}/(Q,ρ)_{L²}, but this should be β = (w,ρ)_{L²}/(Q,ρ)_{L²}; the current formula appears to be a typo and makes the subsequent identity β = 2(u,ρ)/‖xQ‖² inconsistent with the orthogonality condition (w,ρ)=0.","section":"Appendix A, proof of Proposition 2.1(1)"},{"comment":"The equivalence symbol '∼' in (2.11) suggests both upper and lower bounds, but the proof and applications only use the lower bound. Please clarify, e.g., by writing '≳' for the coercivity statement and noting whether the matching upper bound is known.","section":"§2.1, Proposition 2.1(2)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a serious contribution to a well-established program, but its main theorems depend on a numerically assisted coercivity statement that is not reproduced and on the author's own unpublished companion works. I would recommend asking the authors to provide a fully documented verification of Proposition 2.1(2), or to state the main theorems as conditional on that external result, and to make the dependence on [17,18] explicit and verifiable. The mathematical strategy itself appears coherent and the proofs are detailed, but the current reliance on non-reproduced numerical input is a genuine correctness risk for the advertised dimension range."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is the high-energy half of the story, and if the inputs hold it completes the program: for 1 ≤ d ≤ 10 and sc small, Hb has no unstable eigenvalues away from a δ-neighborhood of the origin, and as a byproduct H0 has no embedded eigenvalues or endpoint resonances. That is a genuinely significant extension, and the first full H0 spectrum for 2 ≤ d ≤ 10. The 3D cubic case is not just replayed; the modulation step in the linear Liouville argument is new, and the d = 2 delocalization estimate is a real technical addition. The analytic structure is careful, and the appendices do real work—the resolvent smoothing lemma is solid.\n\nThe soft spots are the ones the stress test flagged, and they are genuine but not automatically fatal. Proposition 2.1(2)—the weighted coercivity of the Virial commutator—is the single input that carries d ≤ 10, and it is 'proven with numerical help' in [23, 11, 30] but not reproduced here. No code, no interval bounds, no verification script. Both Theorem 1.1 and Theorem 1.2 use it at the decisive step; without it, the rigidity conclusions do not go through. That makes the main theorems conditional in a way the abstract does not fully advertise. The reliance on the author's unpublished companion [17] and preprint [18] as black boxes is a second, lesser concern; those are the author's own papers, and the argument assumes them rather than re-proving them. That is normal for a Part II, but an editor should insist that [17] be available and, ideally, that the numerical verification be reproducible.\n\nMinor: the δ-dependence of sc,3(δ) is not tracked explicitly, which is acceptable but leaves the exact range vague. The d = 2 delocalization estimate in Lemma 2.5(2) is plausible but buried in Appendix B; a referee should check it carefully.\n\nWho is this for: anyone working on stability of NLS blowup or non-self-adjoint linearized operators. It deserves a serious referee—conditional acceptance, with the referee asked to verify the dependence on Proposition 2.1(2) and to request the numerics or a rigorous proof if possible.","headline":"A serious and likely important spectral result, but the main theorems are conditional on an externally verified coercivity that the paper does not reproduce.","tokens_in":33985,"tokens_out":2170,"would_cite":true,"duration_ms":27367,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q55","35B44","35P15","35P05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that in dimensions 1 through 10, the self-similar blowup linearized operator has no unstable discrete eigenvalues away from the origin.","keywords":["mode stability","self-similar blowup","slightly supercritical NLS","linearized operator","linear Liouville argument","Virial coercivity","spectrum of ground state","asymptotic stability"],"falsifier":"Compute the quadratic form $(L_1 u,u)+(L_2 w,w)$ for a pair $(u,w)$ satisfying the orthogonality conditions (2.9) in some dimension $d \\le 10$ and find a strictly negative value; or exhibit a nonzero solution of $(H_b-\\lambda)Y=0$ with $\\operatorname{Im}\\lambda < b(\\sigma-s_c)$ and $|\\lambda| \\ge \\delta$; or exhibit an $H_0$ eigenfunction at an embedded eigenvalue $\\lambda \\in \\mathbb{R}\\setminus\\{-1,0,1\\}$ or a resonance at $\\pm 1$. Any one of these would directly contradict Theorems 1.1 or 1.2.","tokens_in":32877,"feed_emoji":"🌀","tokens_out":10567,"duration_ms":104809,"temperature":0.7,"pith_summary":"The paper aims to close the last spectral gap in the asymptotic stability theory for self-similar blowup in slightly mass-supercritical nonlinear Schrödinger equations. It proves that, for dimensions $1 \\le d \\le 10$ and sufficiently small supercriticality, the linearized operator around the self-similar profile has no unstable eigenvalues outside a small neighborhood of the origin, and that the linearized operator around the mass-critical ground state has no embedded eigenvalues or endpoint resonances. Combined with the companion low-energy analysis, this yields full mode stability and hence the asymptotic stability of the self-similar blowup with explicit rates. A sympathetic reader should care because the proof converts a non-self-adjoint spectral problem into a coercivity estimate, and the method is designed to transfer to other non-self-adjoint linearized operators.","feed_headline":"No unstable high-energy modes for supercritical NLS blowup","feed_subtitle":"In d≤10 the linearized operator's spectrum is empty away from zero, completing the stability proof.","key_machinery":"The central mechanism is a linear Liouville argument: an eigenfunction $Y_0$ of $H_b$ is viewed as the initial datum of a globally defined linearized flow $Y(t)=e^{i\\lambda t}Y_0$, and the spectral question is converted into a rigidity question for that flow. The argument combines the algebraic structure of the generalized kernel of $H_0$, a modulation step that imposes almost orthogonality on the evolved pair $(u,w)$, and energy–Virial identities whose coercivity is the key quantitative input: the estimates $(L_1 u,u) \\sim \\|u\\|_{\\dot H^1}^2 + \\|\\langle r\\rangle^{-\\mu_d}u\\|^2$ and the analogous bound for $L_2 w$, valid under orthogonality conditions for $1 \\le d \\le 10$. For the bifurcated operator $H_b$, the proof adds a truncated and weight-adapted Virial functional to handle the near-neutral region $|\\operatorname{Im}\\lambda| \\le b^{4+d}$, and a delocalization estimate for the radial case $d=2$.","core_discovery":"On its own terms, the paper establishes Theorem 1.2: for $1 \\le d \\le 10$, any $\\delta > 0$, and sufficiently small $s_c$, the discrete spectrum of $H_b$ on $(\\dot H^\\sigma)^2$ satisfies $\\sigma_{\\mathrm{dist}}(H_b|_{(\\dot H^\\sigma)^2}) \\cap \\{z : \\operatorname{Im} z < b(\\sigma - s_c),\\ |z| \\ge \\delta\\} = \\emptyset$. In words, the only possible unstable modes of the linearized operator around the self-similar profile $Q_b$ are confined to a small neighborhood of the origin, where the known low-energy analysis applies. Together with that low-energy result, the paper concludes the full mode stability assumption needed for the asymptotic stability of $Q_b$, including sharp rates for the perturbation and logarithmic blowup of critical norms. A byproduct is Theorem 1.1, the complete spectrum of the ground-state linearized operator $H_0$: essential spectrum $(-\\infty,-1] \\cup [1,\\infty)$, discrete spectrum $\\{0\\}$ with the known generalized nullspace, no embedded eigenvalues, and no resonances at $\\pm 1$ for $1 \\le d \\le 10$.","pith_inferences":["I infer that the dimensional restriction $d \\le 10$ is an artifact of the numerically verified coercivity rather than of the structural argument; if the coercivity estimate is proved analytically in higher dimensions, both theorems should extend verbatim to those dimensions.","I infer that the same linear Liouville scheme could be turned into a numerical spectral oracle: running the coercivity check for a candidate operator in a new dimension would give a rigorous certificate of mode stability without computing Jost functions or resolvent kernels.","A testable extension suggested by the paper's closing comments is to push the allowed range of $\\sigma$ from $s_c + b^{d+4}$ up to $O(b)$; the obstruction identified is quantized eigenvalues, so a natural experiment is to search numerically for such eigenvalues in the interval $b \\ll |\\lambda| \\ll b^{d+4}$.","I infer that the absence of embedded eigenvalues for $H_0$ in $d \\le 10$ may be provable by the same weighted Virial identity without the numerical coercivity input, provided one only needs non-resonance instead of full mode stability."],"forward_implications":["For $1 \\le d \\le 10$ and $0 < s_c \\ll 1$, the self-similar blowup profile $Q_b$ is asymptotically stable in $\\dot H^\\sigma$ and in $H^1$ for suitable open sets of initial data, with explicit decay rates and logarithmic divergence of critical norms at the blowup time.","The complete spectrum of the ground-state linearized operator $H_0$ is now known for $1 \\le d \\le 10$: no eigenvalues embedded in the essential spectrum and no resonances at the endpoints $\\pm 1$.","With the low-energy analysis, Theorem 1.2 verifies the full spectral assumption of the stability theorem, so the self-similar blowup scenario is justified at the linear level.","The linear Liouville method replaces a non-self-adjoint eigenvalue problem by coercivity of self-adjoint quadratic forms, so the same strategy can be applied to other linearized operators around stationary states."],"supporting_citations":[{"why":"The companion paper establishing low-energy mode stability; the present theorem is its high-energy counterpart, and the two together yield full mode stability.","marker":"[17]"},{"why":"Constructs the self-similar profile $Q_b$ and its asymptotics; $H_b$ is the linearized operator around this profile.","marker":"[1]"},{"why":"Supplies the coercivity of the Virial commutator, the load-bearing input for the rigidity step, verified numerically for $1 \\le d \\le 10$.","marker":"[23]"},{"why":"Numerical verification of the same controlling quantities for 1D, cited as part of the coercivity input.","marker":"[11]"},{"why":"Numerical verification of the controlling quantities in high dimensions, cited as part of the coercivity input.","marker":"[30]"},{"why":"The stability result whose open question this paper answers, and whose spectral property relies on the same coercivity.","marker":"[24]"},{"why":"Characterizes the generalized kernel of $H_0$ for the mass-critical ground state, the starting point of the bifurcation analysis.","marker":"[29]"},{"why":"Origin of the linear Liouville argument for soliton stability, whose eigenproblem reformulation this paper adapts.","marker":"[21]"}],"fun_headline_variants":["Supercritical NLS blowup: no high-energy unstable modes","Mode stability proven for slightly supercritical NLS","High-energy spectrum empty for NLS blowup profiles","Stability of self-similar blowup concluded in d≤10","Unstable modes only near zero for supercritical NLS"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on a coercivity estimate for the Virial commutator that the paper takes from earlier work with numerical verification and does not prove analytically; if that estimate fails in any dimension $1 \\le d \\le 10$, the rigidity conclusion that the eigenfunction is zero collapses.","fun_headline_variants_meta":{"raw":{"variants":["Supercritical NLS blowup: no high-energy unstable modes","Mode stability proven for slightly supercritical NLS","High-energy spectrum empty for NLS blowup profiles","Stability of self-similar blowup concluded in d≤10","Unstable modes only near zero for supercritical NLS"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000658,"raw_usage":{"total_tokens":3068,"prompt_tokens":1058,"completion_tokens":2010,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":674,"completion_tokens_details":{"reasoning_tokens":1929}},"tokens_in":674,"tokens_out":2010,"duration_ms":15423,"temperature":1.0,"reasoning_tokens":1929,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T17:12:10.556551+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the quadratic form $(L_1 u,u)+(L_2 w,w)$ for a pair $(u,w)$ satisfying the orthogonality conditions (2.9) in some dimension $d \\le 10$ and find a strictly negative value; or exhibit a nonzero solution of $(H_b-\\lambda)Y=0$ with $\\operatorname{Im}\\lambda < b(\\sigma-s_c)$ and $|\\lambda| \\ge \\delta$; or exhibit an $H_0$ eigenfunction at an embedded eigenvalue $\\lambda \\in \\mathbb{R}\\setminus\\{-1,0,1\\}$ or a resonance at $\\pm 1$. Any one of these would directly contradict Theorems 1.1 or 1.2.","supporting_citations":[{"cited_title":"Mode stability for self-similar blowup of slightly supercritical nls: I","cited_arxiv_id":null,"evidence_quote":"The companion paper establishing low-energy mode stability; the present theorem is its high-energy counterpart, and the two together yield full mode stability."},{"cited_title":"Self-similar blow-up profiles for slightly supercritical nonlinear Schrödinger equations.Ann","cited_arxiv_id":null,"evidence_quote":"Constructs the self-similar profile $Q_b$ and its asymptotics; $H_b$ is the linearized operator around this profile."},{"cited_title":"The blow-up dynamic and upper bound on the blow-up rate for critical nonlinear Schrödinger equation.Ann","cited_arxiv_id":null,"evidence_quote":"Supplies the coercivity of the Virial commutator, the load-bearing input for the rigidity step, verified numerically for $1 \\le d \\le 10$."},{"cited_title":"Proof of a spectral property related to the singularity formation for theL2 critical nonlinear Schrödinger equation.Phys","cited_arxiv_id":null,"evidence_quote":"Numerical verification of the same controlling quantities for 1D, cited as part of the coercivity input."},{"cited_title":"Blow-up dynamics and spectral property inthe L2-criticalnonlinearSchrödingerequationinhighdimensions","cited_arxiv_id":null,"evidence_quote":"Numerical verification of the controlling quantities in high dimensions, cited as part of the coercivity input."},{"cited_title":"Stable self-similar blow-up dynamics for slightly L2 super-critical NLS equations.Geom","cited_arxiv_id":null,"evidence_quote":"The stability result whose open question this paper answers, and whose spectral property relies on the same coercivity."},{"cited_title":"Weinstein","cited_arxiv_id":null,"evidence_quote":"Characterizes the generalized kernel of $H_0$ for the mass-critical ground state, the starting point of the bifurcation analysis."},{"cited_title":"Asymptotic stability of solitons for subcritical generalized KdV equations","cited_arxiv_id":null,"evidence_quote":"Origin of the linear Liouville argument for soliton stability, whose eigenproblem reformulation this paper adapts."}],"review_version":1}