{"id":"4d4ec48d-dda5-4f53-b0f5-ee0f80565665","arxiv_id":"2607.28917","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Scattering data from highly oscillatory null-form waves determines the q and p coefficient functions of a coupled wave system, with injectivity in the linear regime and in the time-independent weakly nonlinear regime.","lead":"This paper shows that the coefficient functions in a system of nonlinear wave equations built from \"null forms\" can be recovered from the trace of a carefully designed oscillatory wave on a later time slice. The result extends a known scalar inverse problem to multi-component systems and, in a time-independent weakly nonlinear regime, ties recovery to the non-abelian X-ray transform.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Kernel-injectivity for k>l depends on an unverified exterior-derivative computation (2.4) whose sign/index conventions could hide nonzero q,p.","rationale":"The paper's main novelty is the recovery of the full tensor family {q_{rkl}, p_{rijkl}} for systems. For k=l, the recovery follows from the scalar result [24] and the known light-ray transform injectivity; that branch is credible. For k>l, the proof hinges on the unproved computation (2.4). I examined the surrounding argument: the reduction to dG=0 is standard, the arbitrariness of θ does imply the algebraic system once (2.4) is accepted, and the weakly nonlinear regime's non-autonomous dependence can be handled by choosing φ with a plateau, so it is not the central risk. The genuinely unverified step is the exterior derivative. The reader's verdict CONDITIONAL is appropriate; I recommend no change. A precise symbolic or hand recomputation of dG would either confirm the step or reveal a sign/index error. This is a concrete, falsifiable check that the authors should be asked to provide.","tokens_in":14913,"tokens_out":24083,"duration_ms":218619,"concrete_test":"Independently re-derive dG from G= A_{1,0}^{(k)} G_r using the definition of G_r via (1.10)–(1.11), treating φ as an arbitrary compactly supported function of s=⟨x,V⟩_M. Collect the coefficient of dx0∧dxi as a linear combination of φ' and φ''. Verify that setting the φ'' coefficient to zero for all θ∈S^{d-1} gives 2θ_i A q + Σ_j θ_j A p_{ij}=0 for each i, and that this system has only the zero solution (e.g., test d=2,3 with explicit p-matrix). A symbolic computation (SymPy) with symbolic θ, A, q, p can settle this. If the resulting system admits nonzero (q,p), the conclusion of Theorem 1 for k>l fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The principal load-bearing step is the claim in §2.2 that the light-ray transform of G (defined in (1.10)–(1.11)) is injective for the class of fields arising from the k>l recovery. Because L1 is known to be injective modulo gradient fields, the proof must show that dG=0 implies q_{rkl}=p_{rijkl}=0. This is exactly what (2.4) asserts, but the displayed formula is not derived and the subsequent inference ('Since our choice of φ was arbitrary...') is compressed. The formula involves delicate index/sign conventions (e.g., p_{rijlk}=-p_{rijkl}, sums over i>j vs all j) and combines derivatives of A, q, φ. A single sign error in the dx0∧dxi coefficient would change the algebraic equations in θ and could admit nonzero solutions, invalidating the recovery of p_{rijkl}. This is not a peripheral point: the main theorem's recovery of all p_{rijkl} for k>l depends entirely on this step. The k=l case is independently supported by the scalar paper [24], but no independent check exists for (2.4).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper addresses the inverse problem of recovering the coefficients q_{rkl}(x,u) and p_{rijkl}(x,u) in a system of wave equations with null-form nonlinearities from scattering data. The author constructs highly oscillatory geometric optics solutions and shows that the trace of the solution at a fixed large time, over a family of incident waves parameterized by directions V,W, polarizations k,l and amplitudes, determines the coefficients. In the linear regime, the order h^3 term yields a light-ray transform of a vector field: for k=l this determines q_{rll}, and for k>l it determines q_{rkl} and p_{rijkl}. In the weakly nonlinear regime, the order h term yields a non-abelian light-ray/X-ray transform; when the coefficients are time-independent, injectivity of the non-abelian X-ray transform together with a gauge-fixing argument gives full recovery. The paper also constructs approximate solutions and invokes a result of Guès to pass to exact solutions.","tokens_in":15088,"tokens_out":23693,"duration_ms":227420,"significance":"If correct, the result is a meaningful extension of the scalar null-form recovery [24] to systems and to the antisymmetric null forms Q_{ij}. The strategy is natural: it reduces recovery to known integral-geometric injectivity results (light-ray transform, non-abelian X-ray transform) and gives an explicit measurement procedure. The weakly nonlinear time-independent result is attractive because it avoids the instability of the light-ray transform. The main caveats are technical: the paper delegates a key injectivity lemma to an unpublished preprint and leaves a crucial exterior-derivative computation essentially undisplayed. These are fixable but currently stand in the way of a fully verifiable proof.","major_comments":[{"comment":"Equation (2.4) is load-bearing for the recovery of p_{rijkl}, but it is stated without derivation and the conclusion drawn from it is too compressed. Please provide a full computation of dG, state explicitly whether G is treated as a 1-form or vector field, and clarify the index conventions for p_{r j i k l} when j is not the larger spatial index. The isolation of the φ''_V coefficient by choosing the profile φ with prescribed φ' and φ'' should also be shown. Moreover, the algebraic step needs care: for a fixed θ the system 2θ_i A q + Σ_j θ_j A p_{r j i k l}=0 does not by itself force p=0 for d≥3; one must use that θ is arbitrary (or use the dx_m∧dx_i equations). As written, this step is not auditable.","section":"§2.2, Eq. (2.4)"},{"comment":"The k=l recovery depends on Proposition 1.3 of [24] for the statement that dF=0 forces F=0, and the proof of Theorem 4 is also cited only to [24]. Since [24] is an arXiv preprint and a self-citation, the manuscript is not self-contained on these points. Please either prove these results in the present paper or state them fully as lemmas with complete proofs. This is not a presentation issue: Proposition 1.3 supplies the kernel-injectivity for the entire k=l recovery.","section":"§2.1 and §3.1"},{"comment":"The gauge-fixing argument that removes the non-abelian X-ray transform ambiguity is written in three sentences. Please expand it: after applying the gauge transformation, the structural identity Φ + Σ_i θ_i Ψ_i = 0 for all θ gives Σ_i θ_i g^{-1}∂_i g = 0; choosing θ = e_j yields ∂_j g = 0, and the boundary condition gives g ≡ I. Also clarify the sign/gauge convention in the transformation formula for Ψ_i. Without this detail, the uniqueness claim in Theorem 2 is incomplete.","section":"§2.3"}],"minor_comments":[{"comment":"The displayed equation contains an index inconsistency: the Laplacian term should presumably be -□ A^{(r)}_m rather than -□ A^{(l)}_m. If this is intentional, please explain; otherwise correct it.","section":"§3.3, Eq. (3.7)"},{"comment":"The sum over j in the dx_0∧dx_i coefficient appears to run over the wrong index set: it should be j=1,...,d (spatial indices), not j=1,...,n (solution components). This makes the formula hard to parse.","section":"§2.2, Eq. (2.4)"},{"comment":"The notation '∑_{±m=0}^{∞}' in the expansions should be a finite sum over |m|≤N, or at least be defined consistently with the approximate solutions. As written, the infinite sum conflicts with the finite expansion in Theorem 3.","section":"Throughout"},{"comment":"There are many typographical and OCR-style errors (e.g., 'd’Alambertian', 'S` a Barreto', inconsistent use of n and d in indices, garbled display in (3.6)). A careful proofreading pass is needed.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is promising and the overall strategy is sound, but I cannot certify the central claim in its current form. The main issue is the compressed injectivity step in §2.2, which is load-bearing for the recovery of p_{rijkl}; I found no obvious fatal error, but the computation must be expanded. The heavy reliance on the authors' own unpublished preprint [24] for two key results should also be addressed before publication. I would be willing to look at a revised version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a real extension, not a repackaging. The paper moves the known scalar null-form recovery [24] to an n-component system and, importantly, includes the antisymmetric Q_ij terms, which produce the p_{rijkl} coefficients. The two solution regimes (linear and weakly nonlinear) are handled with stan-dard nonlinear geometric optics, and the weakly nonlinear part correctly lands on a non-abelian X-ray transform where injectivity is known up to gauge. The gauge ambiguity is then removed by the omega=theta observation, which looks sound. The k=l case is well supported by the earlier scalar result, and the paper is honest about the time-dependent weakly nonlinear case not having an injectivity result.\n\nThe proof architecture is standard and plausible: construct approximate oscillatory solutions, match powers of h, reduce to integral geometry. The calculations in Section 3 are detailed enough to follow. The main soft spot is exactly where the stress-test note points: equation (2.4). The claim that dG=0 implies q_{rkl}=p_{rijkl}=0 is load-bearing for all p-recovery in the k>l linear regime, but the displayed exterior derivative is stated without derivation and the 'arbitrary phi, theta, chi' argument is compressed. The index conventions here are delicate (p_{rijlk}=-p_{rijkl}, sums over i>j, the asymmetric matrix G), and a sign slip would change the algebraic system. This is not a fatal flaw, but it is a place where a referee needs to see the full computation. I also noticed a few minor typos in the displayed formulas, e.g., the cos/sin arguments in (2.1) appear to be missing a real factor, and the text of Theorem 1 is a bit terse about the support shift to x_0<0. These are minor.\n\nThe self-citation to [24] is legitimate: the scalar kernel-injectivity lemma is independent of the system case, and the other main tools (light-ray transform injectivity, non-abelian X-ray transform invertibility) are cited to standard references. No circularity problem. No invented entities, no hidden free parameters.\n\nWho is this for? Specialists in inverse problems for nonlinear hyperbolic equations, especially those working with null forms and geometric optics. It deserves a serious referee. I would recommend conditional acceptance: ask the author to expand the derivation of (2.4) and to double-check the sign/index conventions there. That is a concrete, addressable request, not a reason to reject.","headline":"A solid nonlinear-geometric-optics inverse problem paper that genuinely extends the scalar null-form case to systems with antisymmetric terms; the main soft spot is the compressed injectivity step for k>l, which should be expanded before publication.","tokens_in":15644,"tokens_out":1296,"would_cite":true,"duration_ms":16254,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35L05","35R30","44A12"],"pacs":[],"model":"deepseek-v4-flash","headline":"A specially constructed oscillatory wave, measured at one late time, completely determines all coefficients of null-form nonlinearities in a system of wave equations, via an injective light-ray transform in the linear regime and a non-abeli","keywords":["inverse problem","null forms","wave equation","nonlinear geometric optics","light-ray transform","non-abelian X-ray transform","scattering data","highly oscillatory solutions"],"falsifier":"Directly verify equation (2.4): compute the exterior derivative of G for arbitrary compactly supported q_{rkl} and p_{rijkl}, and check whether dG=0 truly forces all coefficients to vanish. If one can produce a nonzero compactly supported pair with dG=0 while the measured light-ray transform also vanishes, the claimed injectivity in Section 2.2 collapses.","tokens_in":14699,"feed_emoji":"📡","tokens_out":8464,"duration_ms":73474,"temperature":0.7,"pith_summary":"This paper solves an inverse problem: can you recover unknown nonlinear terms in a system of wave equations by observing special solutions at a distance? The answer is yes for nonlinearities built from null forms. The author constructs highly oscillatory incident waves and shows that the first new term they create — of order h^3 in the linear regime and order h in the weakly nonlinear regime — is exactly an integral transform of the unknown coefficients. For the symmetric null form, that transform is the classical light-ray transform, known to be injective. For the antisymmetric null forms, the same measurement gives a light-ray transform of a vector field built from both q and p coefficients, and the paper supplies the injectivity argument. In the weakly nonlinear, time-independent case, the data determine a non-abelian X-ray transform that is invertible up to a gauge, and the author shows the gauge is eliminated by a special choice of wave direction.","feed_headline":"Late-time wave data pins down every nonlinearity coefficient","feed_subtitle":"The h^3 term is an injective transform of the unknown coefficients — one late measurement fixes them all.","key_machinery":"The central object is the oscillatory geometric optics solution, a background plane-wave null solution C + h φ_V e_l plus higher-order oscillatory corrections. The identity that carries the argument is that null forms of plane waves vanish, so the first nonlinear interaction between the background and the h^2 perturbing wave appears at order h^3 in the linear regime and order h in the weakly nonlinear regime; its amplitude is governed by a transport equation T_W along the null direction W. The measured coefficient is an integral along the light ray t ↦ (t, x' + tω) — precisely the light-ray transform, an integral of a vector field over all lines parallel to a null direction — of a vector fie","core_discovery":"The central claim is that evaluating a specially designed oscillatory solution at one large time T completely determines the null-form coefficients: for k=l, the order-h^3 term of the r-th component is the future light-ray transform of A^{(l)}_{1,0} q_{rll} φ'_V e_V, and for k>l it is the light-ray transform of a vector field G built from q_{rkl} and p_{rijkl}. Since the light-ray transform is injective modulo potential fields and the paper shows the only potential field is zero, all coefficients q_{rkl} and p_{rijkl} are recovered. In the weakly nonlinear regime, the order-h term solves a matrix transport equation whose scattering data is the non-abelian X-ray transform of a connection plus","pith_inferences":["A natural next step would be to test numerically the injectivity of the non-abelian light-ray transform for genuinely time-dependent coefficients; the paper notes no injectivity result is known, and a counterexample would sharply delimit the method.","Because the coefficients may depend on the solution u, but the construction evaluates them at the constant C, varying C over a range of constants should trace out the full u-dependence from the same family of measurements — a likely but unstated extension.","The same template — engineering an oscillatory solution so that the leading interaction term is a Radon-type transform of the unknown — could apply to other hyperbolic systems with null structure, turning injectivity of the relevant integral geometry into a recovery theorem.","If the asserted exterior-derivative computation in Section 2.2 is made fully explicit, the method becomes a parameter-free reconstruction procedure: no regularization other than the oscillation frequency h and one measurement time."],"forward_implications":["All coefficients q_{rkl} and p_{rijkl} in the null-form system can be recovered from boundary values at a single time, without any knowledge of the initial data.","The recovery extends the previously treated scalar case to systems with both symmetric and antisymmetric null forms, covering equations such as wave maps and related geometric wave equations.","In the linear regime, the reconstruction reduces to inverting the light-ray transform, which is stable only when the coefficients are time-independent; time-dependent coefficients give an unstable inverse.","In the weakly nonlinear regime, time-independent coefficients are recovered through the non-abelian X-ray transform, which is injective only up to gauge; the paper shows the gauge is fixed by the ω=θ measurement.","The method does not require well-posedness of the Cauchy problem for arbitrary data; only the existence of these special oscillatory solutions is needed."],"fun_headline_variants":["Wave echoes reveal all nonlinear coupling strengths","Scattered waves unlock null-form coefficients","Single late measurement fixes all wave nonlinearities","Oscillatory wave data inverts null-form couplings","Wave scattering recovers all null-form coefficients"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof that a vanishing exterior derivative forces all q and p coefficients to vanish — used to show the light-ray transform of the vector field G is injective — is the load-bearing step; if that computation has a sign or index error, the recovery of the p coefficients fails.","fun_headline_variants_meta":{"raw":{"variants":["Wave echoes reveal all nonlinear coupling strengths","Scattered waves unlock null-form coefficients","Single late measurement fixes all wave nonlinearities","Oscillatory wave data inverts null-form couplings","Wave scattering recovers all null-form coefficients"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000608,"raw_usage":{"total_tokens":2724,"prompt_tokens":851,"completion_tokens":1873,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":595,"completion_tokens_details":{"reasoning_tokens":1806}},"tokens_in":595,"tokens_out":1873,"duration_ms":13911,"temperature":1.0,"reasoning_tokens":1806,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T17:13:59.883197+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly verify equation (2.4): compute the exterior derivative of G for arbitrary compactly supported q_{rkl} and p_{rijkl}, and check whether dG=0 truly forces all coefficients to vanish. If one can produce a nonzero compactly supported pair with dG=0 while the measured light-ray transform also vanishes, the claimed injectivity in Section 2.2 collapses.","supporting_citations":[],"review_version":1}