{"id":"0f143482-4fd3-45ae-ba30-004121d9e681","arxiv_id":"2501.07887","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Smooth solutions of u_tt - u_xx = (u_x)^2 admit stable, explicitly written blow-up profiles with logarithmic growth, and no smooth exact self-similar blow-up exists.","lead":"The authors construct explicit solutions of a one-dimensional wave equation that blow up in finite time with logarithmic growth, and they prove that these blow-ups are stable under small perturbations. The equation models nonlinear effects in cosmological effective field theories, so the result gives rigorous control of singularity formation in such models.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Uniform resolvent estimate in §4.5 is not established as written; Theorem 1.7 depends on it.","rationale":"The reader's named weakest assumption was the genericity of the self-similar ansatz, which is a limitation on the interpretation of the result but not a flaw in Theorem 1.7 itself. The more load-bearing issue is the uniform resolvent estimate, which the reader also flagged as under-derived in the rationale. My read agrees with the CONDITIONAL verdict: the overall strategy is coherent and the mode-stability argument is convincing, but the proof of Proposition 4.11 contains algebraic inconsistencies and missing justifications that prevent verification of the key linear decay. Since the nonlinear stability argument (Propositions 5.1, 6.3, and 6.6) relies directly on this estimate, the preprint should either supply the detailed computation or be revised. The proposed re-derivation would settle whether the concern is merely a presentation issue or a genuine gap; if the estimate holds, the central theorem is likely correct, and the current CONDITIONAL verdict remains appropriate.","tokens_in":41790,"tokens_out":34125,"duration_ms":311095,"concrete_test":"Re-derive Proposition 4.11 with correct sesquilinear forms: compute Re⟨(λ−Lα)q,q⟩_{H^{k+1}×H^k} and the analogous ⟨·,·⟩_0 term, and verify the lower bound (4.12) for all α∈B_ǫ(α0), k≥k_{α0}+3, and λ with Reλ≥−w0 and either |Imλ|≥n or Reλ≥m. Specifically, check that the coefficient k−√(1+α)−1/2−ǫ stays positive and that the real part of the potential term is absorbed uniformly in α. If the re-derivation yields (4.12) with α-independent constants, the concern is resolved; otherwise Theorem 1.7 is not proven.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1.7's nonlinear stability hinges on the exponential decay of the linearized semigroup on the stable subspace (Proposition 5.1), which is derived from the uniform resolvent estimate (4.11) of Proposition 4.11. The proof of (4.11) via the lower bound (4.12) is not checkable from the manuscript. In §4.5 the displayed expansion of ((λ−Lα)q,q) uses linear norms where quadratic norms are required (e.g., (1/2+k+Reλ)(‖∂^{k+1}q1‖+‖∂^k q2‖) and the later lower bound with the same linear factors), so (4.12) does not follow dimensionally. Additionally, the potential term ∫ 2α/(√(1+α)+y)∂^{k+1}q1∂^k \\bar q2 is treated as if it were controlled by the same positive coefficient as the free-wave dissipation, and the text asserts that the remaining terms are 'pure imaginary' although this potential term has a real part. A correct proof must absorb the real part of this term using k−√(1+α)−1/2−ǫ, uniformly for α∈B_ǫ(α0). Since the text does not provide this computation, the central resolvent estimate is unsupported.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the one-dimensional wave equation with quadratic spatial derivative nonlinearity, utt - uxx = (ux)^2. It proves (Theorem 1.1) that no smooth exact self-similar blow-up solutions exist, and it constructs an explicit five-parameter family of generalized self-similar solutions with logarithmic growth. The main stability result (Theorem 1.7) asserts that the beta = infinity subfamily, and by symmetry the beta = 0 subfamily, is asymptotically stable under small perturbations in H^{k+1} x H^k: the perturbed solution blows up at a nearby time T* with adjusted parameters, and the difference from the adjusted profile decays like (T* - t)^{1-delta} in all derivatives up to order k+1. The proof combines a Lorentz transformation in self-similar variables to reduce the mode-stability problem to a hypergeometric connection problem, a spectral analysis of the non-self-adjoint linearized operator via a finite-rank dissipative decomposition, a uniform resolvent estimate, and a Lyapunov-Perron fixed-point argument for the nonlinear evolution.","tokens_in":41977,"tokens_out":13174,"duration_ms":132220,"significance":"If completed, this is a substantial contribution: it provides the first explicit blow-up profiles for (1.1), and it extends Donninger's spectral framework to a setting with a non-compact derivative perturbation. The explicit profiles are parameter-free and can be checked by direct substitution; the Lorentz transformation is an elegant device that reduces a Heun connection problem to a hypergeometric one; the nonlinear iteration is standard once the linear decay is available. The main caveat is that the uniform resolvent estimate in Section 4.5, which is the linchpin of the linear and nonlinear stability, is not established as written, so Theorem 1.7 is currently conditional. The paper also does not prove that the generalized self-similar ansatz captures generic blow-up dynamics; Remark 1.5 only conjectures this, so the stability result applies to a specially prepared family.","major_comments":[{"comment":"The uniform resolvent estimate (4.11) is load-bearing: it is used in Proposition 5.1 to obtain the exponential decay of the semigroup on the stable subspace, and hence it underlies Theorem 1.7. The proof of the lower bound (4.12) is not valid as written. In the expansion of ((lambda - L_alpha)q,q) on dot H^{k+1} x dot H^k, the principal terms are written with linear norms, e.g. (1/2 + k + Re lambda)(||partial^{k+1} q_1||_{L^2} + ||partial^k q_2||_{L^2}), while (4.12) is a quadratic lower bound in ||q||_{H^k}; similarly the term i Im lambda (||partial^{k+1} q_1|| + ||partial^k q_2||) should be i Im lambda (||partial^{k+1} q_1||^2 + ||partial^k q_2||^2). The potential term integral of 2 alpha/(sqrt(1+alpha)+y) partial^{k+1} q_1 overline{partial^k q_2} and the lower-order commutator sum are not pure imaginary, contrary to the assertion in the text; their real parts must be absorbed, for example by the large imaginary part after the norms are squared, or by the coefficient k - sqrt(1+alpha) - 1/2 - epsilon uniformly over alpha in B_epsilon(alpha_0). The manuscript provides neither computation. In addition, the displayed bracket with y partial^{2+k} q_1 partial^k q_1 appears twice with identical signs and is therefore identically zero, so it cannot represent the intended purely imaginary boundary term unless one factor is conjugated. A corrected proof must rewrite the expansion with squared norms and carry out the absorption of the real part of the potential term before (4.12) can be accepted.","section":"Section 4.5, Proposition 4.11 and the display following (4.12)"},{"comment":"Even after the homogeneity issue in the expansion is repaired, the treatment of the region S1 = {Re lambda >= -w0, |Im lambda| >= n} is incomplete: the final lower bound in the S1 case contains Im lambda rather than |Im lambda|, so negative imaginary parts with large modulus are not covered by the stated inequality. Because L_alpha has real coefficients, the resolvent satisfies R_{L_alpha}(conj(lambda)) = conj(R_{L_alpha}(lambda)), so the estimate for negative imaginary parts can be obtained by conjugation; this step must be stated explicitly and the lower bound should be written with |Im lambda|.","section":"Section 4.5, S1 region"}],"minor_comments":[{"comment":"The abstract says 'these blow-up solutions' are asymptotically stable, but Theorem 1.7 proves stability only for the beta = infinity and beta = 0 subfamilies; Remark 1.9 conjectures only co-dimensional stability for 0 < beta < infinity. The wording should be adjusted to avoid overstating the scope.","section":"Abstract and Theorem 1.7"},{"comment":"The displayed closed form for partial_y W_{alpha,beta,kappa} has a parenthesis mismatch and is difficult to read; the authors should rewrite it in a fully parenthesized form, since this formula is used in the explicit profile construction.","section":"Equation (2.5)"},{"comment":"In the definition of g^n_alpha, the term f_{0,beta} should presumably be f_{0,alpha}; as printed, an undeclared parameter beta appears.","section":"Section 6.2, dual basis definition"},{"comment":"The notation H^k is used in Section 4 for the product space H^{k+1}(-1,1) x H^k(-1,1), while Section 6 writes H^k(B); this should be harmonized and defined once in the notation section.","section":"Section 4.1 and Section 6"},{"comment":"The step from the coefficient ratio r_n(lambda) -> 1 to the conclusion that the radius of convergence is exactly 1 relies on the fact that a_n(lambda) is nonzero for all n when lambda is not in {0,1} and Re lambda > -1; this is true, but it should be stated explicitly, together with the non-termination of the hypergeometric series in the cases where c - a - b is a nonpositive integer.","section":"Proposition 3.11"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a genuine gap in the blow-up literature, and the explicit construction together with the Lorentz-transformation idea are attractive strengths. The serious issue is confined to the proof of the uniform resolvent estimate in Section 4.5, which is currently not checkable and is load-bearing for the stability theorem. I do not think this warrants rejection, because the defect appears repairable: the expansion must be rewritten with squared norms and the real part of the potential term must be absorbed, likely using the large imaginary part or the positive coefficient k - sqrt(1+alpha) - 1/2. The authors should be asked to provide a complete, self-contained proof of Proposition 4.11 before the paper is accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: the explicit generalized self-similar profiles (Theorem 1.1) are new, explicit, and checkable. The non-existence of smooth exact self-similar solutions is a simple and correct argument. The Lorentz-transform reduction of the Heun connection problem to a hypergeometric problem is genuinely elegant, and the mode stability result (only eigenvalues 0 and 1 with Re λ ≥ −1) appears right. The finite-rank decomposition idea for handling the non-compact potential is a good idea.\n\nThe problem is §4.5. The proof of the uniform resolvent estimate (4.11) does not work as written. The displayed expansion of ((λ − Lα)q, q) uses linear norms where quadratic norms are needed; the potential term ∫ 2α/(√(1+α)+y) ∂^{k+1}q1 ∂^k \\bar q2 dy is not pure imaginary, and the bound given for it is dimensionally wrong. Because of this, the lower bound (4.12) does not follow, and Proposition 4.11 is unsupported. That estimate is what drives the exponential decay of the linearized semigroup on the stable subspace (Proposition 5.1) and hence the nonlinear stability (Theorem 1.7). So the central stability theorem is not established by this preprint.\n\nI don't think the idea is fundamentally wrong. A correct proof should absorb the real part of that potential term using k − √(1+α) − 1/2 − ε, uniformly for α in a neighborhood of α0. That is plausibly doable, but the manuscript does not contain the computation. The rest of the spectral framework is coherent, and the Lyapunov–Perron setup in Section 6 would work once the resolvent estimate is fixed.\n\nMinor points: the abstract and introduction overstate higher-dimensional generality; the explicit profiles are 1D-specific, and no higher-dimensional analysis is supplied. Also, the stability theorem is stated for the family of profiles, not for generic blow-up dynamics; Remark 1.5 only conjectures genericity.\n\nWho is this for? Anyone working on blow-up for wave equations, especially derivative nonlinearities. The construction and mode stability are worth reading now. The stability proof needs work before it can be trusted. I would send it to peer review—the ideas are substantial and the gap is identifiable and probably fixable—but the referee should demand a complete, correct proof of §4.5 before acceptance.","headline":"Explicit profiles and mode stability are solid, but Theorem 1.7 rests on a resolvent estimate in §4.5 that has a norm error and a misclassified term; the stability proof needs a fix.","tokens_in":42523,"tokens_out":4213,"would_cite":true,"duration_ms":37934,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35L05","35B44","35B35"],"pacs":[],"model":"deepseek-v4-flash","headline":"The one-dimensional wave equation with quadratic spatial derivative nonlinearity admits no smooth exact self-similar blow-up, but a five-parameter family of logarithmically growing profiles is constructed and proved nonlinearly stable.","keywords":["wave equation","blow-up","quadratic derivative nonlinearity","generalized self-similar solutions","nonlinear stability","Lorentz transformation","logarithmic growth","effective field theory"],"falsifier":"Numerically solve the eigenvalue problem for the linearized operator at $u_{\\alpha,\\infty}$ on $[-1,1]$ for several $\\alpha>0$: mode stability asserts that no smooth eigenfunction exists with $\\mathrm{Re}\\,\\lambda > -1$ except $\\lambda=0,1$, with $\\lambda=0$ having exactly one generalized eigenfunction; finding any additional unstable or neutral mode, or a second generalized eigenfunction, would falsify the stability theorem.","tokens_in":41565,"feed_emoji":"🌊","tokens_out":11288,"duration_ms":110432,"temperature":0.7,"pith_summary":"The paper studies the one-dimensional wave equation $u_{tt}-u_{xx}=(u_x)^2$, a model that arises in effective field theory in cosmology. It proves that no smooth exact self-similar blow-up solutions exist, and that the only stationary solutions are singular logarithmic functions. It then constructs a five-parameter family of smooth generalized self-similar solutions whose leading term grows like $-\\alpha\\log(1-t/T)$, and proves in Theorem 1.7 that the endpoint members of this family, $\\beta=\\infty$ (and by reflection $\\beta=0$), are nonlinearly stable. Concretely, any sufficiently small perturbation of the initial data in $H^{k+1}\\times H^k$ (with $k\\ge k_{\\alpha_0}+3$) still blows up, with adjusted parameters $\\alpha^*,T^*,\\kappa^*$ and a residual decay of $O((T^*-t)^{1-\\delta})$. If the theorem is right, small smooth perturbations of these explicit profiles remain in the same logarithmic blow-up regime; the only freedom is a slight shift of time, place, and amplitude.","feed_headline":"Logarithmic blow-up profiles proven stable","feed_subtitle":"For a quadratic wave equation, small perturbations still blow up at a shifted time and place, with the same logarithmic rate.","key_machinery":"The argument is carried by four devices. The generalized self-similar ansatz $U(s,y)=\\alpha s+\\tilde U(y)$ turns the PDE into a Riccati ODE that can be integrated in closed form, producing the explicit five-parameter family. The Lorentz transformation in self-similar variables, a space-time boost applied after passing to similarity coordinates, converts the Heun-type eigenvalue ODE with four singular points into a hypergeometric equation, whose connection theory proves that the only unstable eigenvalues are $0$ and $1$. Because the derivative nonlinearity produces only a bounded, non-compact perturbation of the free wave operator, the authors add a finite-rank projection to the linearized operator so that the remainder is dissipative; this restores compactness and locates the essential spectrum. A direct resolvent estimate then converts the spectral information into a uniform semigroup growth bound, and a contraction-mapping argument on an exponentially weighted space closes the nonlinear stability.","core_discovery":"The central claim is that blow-up for (1.1) is logarithmic rather than self-similar in the classical sense. Exact self-similar profiles fail: the stationary ODE forces a singularity inside the light cone, so no smooth exact self-similar blow-up exists. By allowing a term linear in the similarity variable, $U(s,y)=\\alpha s+\\tilde U(y)$, the authors solve the resulting Riccati ODE explicitly and obtain smooth profiles $u_{\\alpha,\\beta,\\kappa,T,x_0}$ for $\\alpha>0$ (with $\\beta=0,\\infty$ always smooth, and finite $\\beta$ smooth only when $\\sqrt{1+\\alpha}$ is a positive integer). The main stability theorem asserts that the profile $u_{\\alpha_0,\\infty,\\kappa_0,T_0,x_0}$ is nonlinearly stable: for every small real perturbation $(f,g)\\in H^{k+1}(\\mathbb{R})\\times H^k(\\mathbb{R})$, $k\\ge k_{\\alpha_0}+3$, there exist nearby parameters $(\\alpha^*,T^*,\\kappa^*)$ and a unique solution in the backward light cone whose difference from $u_{\\alpha^*,\\infty,\\kappa^*,T^*,x_0}$ decays like $(T^*-t)^{1-\\delta}$ in all derivatives up to order $k+1$; the reflected statement for $\\beta=0$ follows from the symmetry $x\\mapsto -x$.","pith_inferences":["Editorial inference: if the genericity conjecture in the paper is correct, the M-type singularities seen in numerical EFT simulations should carry a logarithmic prefactor; the paper suggests those simulations may have been capturing superpositions of profiles with $\\alpha$ approaching $0$ as $t\\to T$.","Editorial inference: the Lorentz reduction indicates that this derivative-quadratic nonlinearity becomes ODE-like in a boosted frame, so the same device may help classify the numerically observed V-type versus M-type singularities in related cosmological models.","Editorial inference: because the stability theorem is localized in the backward light cone and the equation has finite propagation speed, generic compactly supported perturbations of the stable profile should produce smooth global-in-space blow-up solutions, not merely cone-localized ones."],"forward_implications":["Any sufficiently small smooth perturbation of a stable profile $u_{\\alpha_0,\\infty,\\kappa_0,T_0,x_0}$ still blows up at a finite time, with blow-up time, center, and additive constant shifting by $O(\\epsilon)$ and the same logarithmic profile persisting.","No smooth exact self-similar or solitary-wave blow-up exists for (1.1); since the only stationary solutions are singular logarithmic functions, any smooth blow-up must carry logarithmic growth in similarity time.","By finite propagation speed and a cutoff construction, the paper obtains smooth global-in-space blow-up solutions, and the stability theorem makes these solutions reachable from small localized perturbations of the explicit profile.","The proof strategy, combining a Lorentz reduction to a hypergeometric connection problem with a dissipative-plus-finite-rank decomposition of the linearized operator, is designed to extend to higher dimensions and to other equations with explicit self-similar solutions."],"supporting_citations":[{"why":"Supplies the spectral-theoretic framework and the local power-series toolkit used for mode stability.","marker":"[12]"},{"why":"Introduces the Lorentz transformation in self-similar variables and the blow-up profile methodology that the construction adapts.","marker":"[38]"},{"why":"Provides the functional framework used to restore compactness via finite-rank projections and subcoercivity estimates.","marker":"[35]"},{"why":"Shows how semigroup estimates plus a fixed-point argument yield nonlinear stability of ODE blow-up; this is the route extended here.","marker":"[16]"},{"why":"Developed the stable self-similar blow-up analysis for wave maps that underlies the linear stability setup.","marker":"[10]"},{"why":"Provides the hypergeometric connection formulas used after the Lorentz reduction to prove mode stability.","marker":"[44]"},{"why":"Provides spectral theory for non-self-adjoint operators, including essential-spectrum behavior under compact perturbations.","marker":"[29]"},{"why":"Supplies semigroup generation and resolvent-growth theorems used for the linearized flow.","marker":"[23]"}],"fun_headline_variants":["Logarithmic blow-up for wave equation proven stable","Stable logarithmic blow-up in quadratic wave equation","Wave blow-up is logarithmic, not self-similar","Quadratic wave blow-up: stable logarithmic profiles","Logarithmic blow-up profiles are asymptotically stable"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that generic blow-up of the equation can be represented by the generalized self-similar ansatz $U(s,y)=\\alpha s+\\tilde U(y)$ plus a decaying remainder; the stability theorem proves this for the constructed family, but the genericity of the ansatz is only conjectured in the paper.","fun_headline_variants_meta":{"raw":{"variants":["Logarithmic blow-up for wave equation proven stable","Stable logarithmic blow-up in quadratic wave equation","Wave blow-up is logarithmic, not self-similar","Quadratic wave blow-up: stable logarithmic profiles","Logarithmic blow-up profiles are asymptotically stable"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000782,"raw_usage":{"total_tokens":3512,"prompt_tokens":1063,"completion_tokens":2449,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":679,"completion_tokens_details":{"reasoning_tokens":2375}},"tokens_in":679,"tokens_out":2449,"duration_ms":17568,"temperature":1.0,"reasoning_tokens":2375,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:31:35.861558+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically solve the eigenvalue problem for the linearized operator at $u_{\\alpha,\\infty}$ on $[-1,1]$ for several $\\alpha>0$: mode stability asserts that no smooth eigenfunction exists with $\\mathrm{Re}\\,\\lambda > -1$ except $\\lambda=0,1$, with $\\lambda=0$ having exactly one generalized eigenfunction; finding any additional unstable or neutral mode, or a second generalized eigenfunction, would falsify the stability theorem.","supporting_citations":[{"cited_title":"Donninger","cited_arxiv_id":null,"evidence_quote":"Supplies the spectral-theoretic framework and the local power-series toolkit used for mode stability."},{"cited_title":"Merle and H","cited_arxiv_id":null,"evidence_quote":"Introduces the Lorentz transformation in self-similar variables and the blow-up profile methodology that the construction adapts."},{"cited_title":"Merle, P","cited_arxiv_id":null,"evidence_quote":"Provides the functional framework used to restore compactness via finite-rank projections and subcoercivity estimates."},{"cited_title":"Donninger and B","cited_arxiv_id":null,"evidence_quote":"Shows how semigroup estimates plus a fixed-point argument yield nonlinear stability of ODE blow-up; this is the route extended here."},{"cited_title":"Donninger","cited_arxiv_id":null,"evidence_quote":"Developed the stable self-similar blow-up analysis for wave maps that underlies the linear stability setup."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the hypergeometric connection formulas used after the Lorentz reduction to prove mode stability."}],"review_version":1}