{"id":"a7b244f9-5597-44ce-9db4-99956e9dd6dd","arxiv_id":"2506.03783","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Orthonormal systems of inputs satisfy a global Mizohata-Takeuchi-type weighted inequality for the sphere and paraboloid, with an X-ray transform norm over the midpoint set K^diamond on the right-hand side.","lead":"This mathematics paper proves new weighted intensity estimates for wave fields built from many mutually orthogonal wave packets, improving on what is known for a single packet. The results show that a recently disproved conjecture of Mizohata and Takeuchi survives in this many-wave, orthonormal setting, and the paper supplies a new phase-space proof of fermionic Strichartz estimates.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proof of the key tomographic lemma (Lemma 2.1) relies on a false evenness assertion for g^diamond; the second hemisphere term is controlled only by a symmetry that is not written down, so Theorem 1.1's constant-one proof is incomplete as stated.","rationale":"Theorem 1.1 is the paper's flagship claim, and its proof is reduced through (2.4)-(2.6) to Lemma 2.1 with g=1_K. Thus any gap in Lemma 2.1 is load-bearing. Tracing the displayed computation in the lemma, the only non-routine step is the inequality applied to the sum of the two hemisphere terms; the second term corresponds to the antipodal midpoint -omega, so the stated justification that g^diamond is even is false for arbitrary nonnegative g. A two-line symmetry argument repairs the lemma, but it is absent. Since the reader's verdict was already CONDITIONAL, this does not change the verdict category, but it changes the requested condition: in addition to the companion-preprint lemmas, the authors should supply the corrected proof of Lemma 2.1. I do not see reason to believe Theorem 1.1 is false; the repair appears to work, and the rest of the Schatten deduction is clean. The companion-preprint dependence identified by the reader is real but secondary for the flagship theorem, since Lemma 2.1's only external input is the standard Jacobian formula for the reflection map.","tokens_in":31475,"tokens_out":34373,"duration_ms":382462,"concrete_test":"Perform the analytical repair in the proof of Lemma 2.1: set u_+(omega,xi)=a(xi)omega+xi/2 and u_-(omega,xi)=-a(xi)omega+xi/2, verify the identity I_-(omega,xi)=I_+(-omega,xi), and check that (1/2)(I_+(omega,xi)+I_+(-omega,xi)) integrates over dsigma(omega) to int I_+ dsigma. If this identity holds, (2.9) follows with constant 1 and Theorem 1.1 stands. Independently, numerically integrate both sides of (2.9) for n=3 with g=1_{C_1}+1_{C_2} for two small, antipodally asymmetric caps and phi=chi_{B(xi_0,r)} with xi_0 the difference of the cap centers and r much smaller than the cap diameter; if the left side exceeds the right, the lemma as stated is false.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the proof of Lemma 2.1, after the change of variables xi = omega' - R_omega omega', the two hemisphere terms are (1/2) int phi(xi) g(omega'(xi)) g(R_omega omega'(xi)) |omega·omega'|^{n-3} dxi and (1/2) int phi(xi) g(-omega'(xi)) g(-R_omega omega'(xi)) |omega·omega'|^{n-3} dxi. The first pair has great-circular midpoint omega; the second pair has midpoint -omega whenever omega·omega'>0. Hence the second term is bounded by g^diamond(-omega), not by g^diamond(omega), unless g^diamond is even. The paper states \"Here we have used that g^diamond is an even function,\" but the definition g^diamond(omega)=sup g(omega')g(omega'') over pairs whose geodesic midpoint is omega does not imply evenness; for g=1_C with C a single cap, g^diamond is supported near C, not near -C. This gap is load-bearing: Theorem 1.1 applies the lemma with g=1_K and needs (1_K)^diamond = 1_{K^diamond}; an unjustified symmetrization would introduce an extra factor and destroy the constant-one statement. The assertion is repairable by observing that the second term equals the first term with omega replaced by -omega and using R_0phi(-omega)=R_0phi(omega), so integrating over omega cancels the 1/2 and yields (2.9). But that repair is not present in the text; as written, the flagship lemma is not proved.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies weighted L^2 inequalities of Mizohata--Takeuchi type for Fourier extension operators applied to orthonormal systems. The flagship result, Theorem 1.1, asserts that for n >= 3 and any orthonormal sequence (g_j) in L^2(S^{n-1}), one has sum_j (integral |g_j dsigma|^2 w)^2 <= ||Xw||^2_{L^2({(omega,v): omega in K^diamond, v in <omega>^bot})} for all signed weights w, where K is the union of the supports of the g_j and K^diamond is the great-circular midpoint set. The proof uses Schatten (Hilbert--Schmidt) duality to reduce the estimate to a tomographic bound, Lemma 2.1, for the Fourier transform of |g dsigma|^2. The paper also develops a direct approach based on spherical and S-carried Wigner distributions, yielding variants with additional orthogonality hypotheses (Theorems 1.2 and 2.8) and an extension to general convex hypersurfaces with bounded curvature quotient (Theorem 3.2). For the paraboloid, Theorem 1.4 proves an analogous orthonormal weighted Strichartz inequality with phase-space support M equal to the union of the supports of the classical Wigner transforms of the initial data. The final sections contain observations for p != 2: interpolation for 1 <= p <= 2, a co-positivity reformulation for even integers p, and reverse inequalities for p <= 1 under completeness assumptions.","tokens_in":31767,"tokens_out":17760,"duration_ms":185627,"significance":"If Theorem 1.1 holds as stated, it is a substantial new result in weighted extension theory for orthonormal systems: it is a global, scale-invariant estimate whose right-hand side is purely geometric, and it holds with constant one for signed weights. This is genuinely different from the single-input global Mizohata--Takeuchi inequality, which is known to fail in the form conjectured in (1.4). The Schatten proof is explicit and elementary in structure, and the direct Wigner approach provides a route to orthonormal Strichartz estimates that is different from the usual trace-ideal arguments. The paper is also careful to expose the connections to co-positivity, Sobolev smoothing, and reverse inequalities. The main caveats are the proof gap in Lemma 2.1 discussed below and the dependence of the general-hypersurface and direct-approach results on the companion preprint [8]; the spherical Schatten theorem does not inherit the latter dependence once Lemma 2.1 is repaired.","major_comments":[{"comment":"The line 'Here we have used that g^diamond is an even function' is false for the sup-autocorrelation definition in Lemma 2.1. For example, if g = 1_C with C a small cap, then g^diamond is supported near C, not near -C. In the displayed estimate after the change of variables xi = omega' - R_omega omega', the second hemisphere term is bounded by g^diamond(-omega) R_0 phi(omega), not by g^diamond(omega) R_0 phi(omega). The desired conclusion (2.9) is nevertheless recoverable: one should keep the two terms together, bound them pointwise by (1/2)(g^diamond(omega) + g^diamond(-omega)) R_0 phi(omega), integrate in omega, and then use R_0 phi(-omega) = R_0 phi(omega) together with the antipodal invariance of dsigma to obtain (2.9). This symmetrization step is not present in the manuscript; as written, the proof of the flagship lemma is incomplete. Please rewrite this step explicitly.","section":"Section 2.1, proof of Lemma 2.1"},{"comment":"The statement of Theorem 3.2 assumes only that S has finite curvature quotient Q(S), but the proof in Section 3.2 uses the additional structural hypothesis introduced in Section 3.1, namely that the normal set N(S) is geodesically convex. This hypothesis is needed both for the global definition of the map R_u and for the asserted surjectivity of u |-> R_u u'. If this hypothesis is intended to be part of the standing assumptions, it should be stated explicitly in Theorem 3.2 and Lemma 3.3; if not, the proof is incomplete as written. Please clarify the precise set of hypotheses under which Theorem 3.2 is claimed.","section":"Section 3.1--3.2, Theorem 3.2 and Lemma 3.3"},{"comment":"The direct-approach theorems (Theorem 1.2/2.8) and the general-hypersurface Theorem 3.2 depend on identities and bounds quoted from the companion preprint [8]: the phase-space representation (2.28) and the Jacobian-ratio estimate J(u,u')/J~ (u,u') <= c Q(S)^{(5n-8)/2} used in the proof of Lemma 3.3. Since [8] is a separate manuscript and not part of this paper, these results are conditional on the correctness of borrowed statements. Please either reproduce proofs of the needed identities and bounds in an appendix, or state explicitly that the theorems in question rely on [8]. The spherical Schatten proof of Theorem 1.1 does not share this problem once Lemma 2.1 is repaired.","section":"Sections 2.2 and 3.3"}],"minor_comments":[{"comment":"The qualifier 'suitable' in Lemma 2.1 and Proposition 2.9 is never defined. Please specify the regularity and support assumptions on g and on the test function phi under which the changes of variables and the sup-autocorrelation formulas are legitimate.","section":"Lemma 2.1 and Proposition 2.9"},{"comment":"Theorem 1.1 is stated for signed weight functions, but the proof in Section 2.1 initially treats real-valued w. Since the inequality is quadratic in w, complex weights can be handled by decomposing into real and imaginary parts; please state this explicitly in the proof or in a remark.","section":"Theorem 1.1 and Section 2.1"},{"comment":"The notation 'co-positive semi-definite' and the symbol <_{cpd} are used before a formal definition is given. Please add a precise definition of cpd-positivity for distributions or functions of the form (2.14).","section":"Remark 2.3"},{"comment":"The sentence 'Note that for n=2 the additional hypothesis means that the singularity from the jacobian factor is removed' is too terse. A compactness argument showing that the no-antipodal-points condition gives a uniform lower bound on |omega . omega'| over the relevant pairs would make the n=2 case clear.","section":"Lemma 2.1, n=2 case"}],"recommendation":"major_revision","confidential_remarks":"The false-evenness assertion in Lemma 2.1 is a genuine proof error, but it is local and has a clear repair via symmetrization after integration in omega; I do not see it as grounds for rejection. The main unresolved issue is the dependence of the general-hypersurface and Wigner-based results on the companion preprint [8]. If the authors add the missing proofs or clearly mark the dependency, and fix the Lemma 2.1 proof, the paper would be close to acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on arXiv:2506.03783. The main new thing is Theorem 1.1: a global, scale-invariant weighted L^2 inequality for orthonormal systems with signed weights, with a right-hand side that is a purely geometric X-ray norm over the midpoint set K^⋆. That is a real advance for weighted Fourier restriction, and it is refreshingly free of fitted parameters. The Schatten-duality reduction is clean and self-contained, and the passage from (2.5) to (2.6) via that tomographic bound is the right idea. The paraboloid theorem (1.4) is also a nice addition; the direct Wigner proof recovers known orthonormal Strichartz estimates and addresses a question from FLLS about a direct approach.\n\nNow the soft spots. The proof of Lemma 2.1 as printed contains an error. The paper asserts that g^⋆ is even, which is not true for general g (take g=1_C with C a cap: the second hemisphere term's midpoint is -ω, not ω). So the pointwise bound ≤ g^⋆(ω) R_0φ(ω) does not follow for the sum of the two hemisphere terms. The lemma is still true—you can repair the argument by noticing that the second term becomes the first under ω → -ω, and R_0φ(-ω)=R_0φ(ω); integrating over ω cancels the 1/2 and gives (2.9). But that repair is not in the text, so the flagship lemma's proof is incomplete as written.\n\nThe other softness is structural: Theorem 3.2 and the general-hypersurface statements rely on identities and a Jacobian bound from the same authors' companion preprint [8]. These are transparently cited and parameter-free, but not proved here. If either were wrong, those theorems would fail—though Theorem 1.1's Schatten proof would still stand. This is a normal companion-paper issue; a revision should either include the needed statements or explicitly point to the preprint's sections.\n\nOverall, the central inequality is likely correct, the proof gap is small, and the dependence on [8] is clearly flagged. The paper deserves a serious referee; I'd ask the authors to fix Lemma 2.1's proof and to ensure the [8] results are accessible. Not a desk reject.","headline":"Strong paper with a genuinely new theorem, but the key tomographic lemma has a repairable evenness error and the general-hypersurface results lean on a companion preprint.","tokens_in":32406,"tokens_out":11063,"would_cite":true,"duration_ms":105448,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["42B10","44A12"],"pacs":[],"model":"deepseek-v4-flash","headline":"The global Mizohata–Takeuchi inequality, false for single functions, holds for orthonormal systems when the X-ray norm is taken over the midpoint set.","keywords":["orthonormal systems","Fourier extension operators","Mizohata–Takeuchi inequalities","Wigner distributions","Schatten duality","X-ray transform","restriction estimates","Strichartz estimates"],"falsifier":"Take a Cantor-type set $K\\subset S^{n-1}$ with very small surface measure and $K^\\diamond$ nearly the whole sphere, and check numerically whether the distribution $X_0^* 1_{K^\\diamond} - |\\widehat{1_K\\,d\\sigma}|^2$ has nonnegative expectation against every smooth compactly supported test weight $w$; a single negative expectation would disprove the tomographic estimate and with it Theorem 1.1.","tokens_in":31205,"feed_emoji":"🌐","tokens_out":12742,"duration_ms":128776,"temperature":0.7,"pith_summary":"The paper proves that a global, scale-invariant Mizohata–Takeuchi inequality — a weighted bound that is false for single functions — holds when the inputs are an orthonormal system. The flagship result, Theorem 1.1, says that for an orthonormal sequence $(g_j)$ in $L^2(S^{n-1})$ with $n\\ge3$, every signed weight $w$ satisfies $$\\sum_j \\Big(\\int_{\\mathbb{R}^n} |\\widehat{g_j\\,d\\$\\sigma$}|^2 w\\Big)^2 \\le \\|Xw\\|^2_{$L^{2}$(\\{(\\omega,v): \\omega\\in K^\\diamond,\\, v\\in\\langle\\omega\\rangle^\\perp\\})},$$ where $K=\\bigcup_j \\operatorname{supp}(g_j)$ and $K^\\diamond$ is the set of great-circular midpoints of pairs of points in $K$. The right-hand side is purely geometric, an $L^2$ X-ray norm of the weight over directions in $K^\\diamond$. The same mechanism gives a paraboloid/Schrödinger analogue, Theorem 1.4, via a direct Wigner-distribution argument, and for $p>2$ the proposed family is recast as co-positivity of explicit tensor forms.","feed_headline":"Orthonormal systems satisfy a global Mizohata–Takeuchi bound","feed_subtitle":"The weighted extension estimate, false for single functions, holds with a purely geometric X-ray norm and no scale loss","key_machinery":"The two load-bearing mechanisms are the Schatten-duality reduction and the spherical Wigner distribution. Schatten duality converts the desired inequality into the Hilbert–Schmidt norm bound $\\|E_K^* w E_K\\|_{C^2} \\le \\|Xw\\|_{L^2(\\omega\\in K^\\diamond)}$, which in turn is equivalent to the pointwise tomographic inequality $\\widehat{1_K\\,d\\sigma}*\\widetilde{\\widehat{1_K\\,d\\sigma}} \\le R_0^*(1_{K^\\diamond})$; here $R_0^*$ is the pullback of the radial X-ray transform and $K^\\diamond$ is the set of geodesic midpoints. The direct approach uses the spherical Wigner transform $W_{S^{n-1}}(g,g)(\\omega,v)$, together with the phase-space identity $|\\widehat{g\\,d\\sigma}|^2 = X^* W_{S^{n-1}}(g,g)$ and the spherical Moyal identity (2.31), which makes $\\langle W(f,f),W(g,g)\\rangle$ expressible as a sum of two weighted inner products. That identity is what turns orthonormality of the inputs into an $\\ell^2$ bound on the phase-space intensities.","core_discovery":"The central discovery is that the interference that breaks the global Mizohata–Takeuchi inequality for single functions is controlled by orthonormality, provided the line-integral norm is measured over the midpoint set $K^\\diamond$ rather than the support set $K$. The proof via Schatten duality reduces the inequality to the tomographic estimate of Lemma 2.1, $$\\widehat{1_K\\,d\\$\\sigma$}*\\widetilde{\\widehat{1_K\\,d\\$\\sigma$}} \\le R_0^*(1_{K^\\diamond}),$$ interpreted as positive semi-definite distributions; equivalently, the whole argument rests on a pointwise hyperplane-bundle bound for the Fourier transform of $|\\widehat{g\\,d\\sigma}|^2$. The direct Wigner approach proves a spherical Moyal identity, Proposition 2.9, showing that spherical Wigner transforms of orthonormal inputs inherit an almost-orthonormality property. The paper leaves open whether $K^\\diamond$ can be replaced by $K$ for nonnegative weights; that is the co-positivity question (2.14).","pith_inferences":["The midpoint set $K^\\diamond$ appearing on the right suggests that for orthonormal systems the effective direction set for X-ray control is governed by pairwise geodesic midpoints; a natural extension is to test whether nonnegative weights admit the smaller set $K$ by checking co-positivity of (2.14) on fractal examples.","Because the spherical Moyal identity is most explicit in $n=3$, the Wigner approach suggests a hierarchy of intermediate direction sets between $K^*$ and $K^\\diamond$ in higher dimensions; proving $\\ell^2$-boundedness of the kernel $L(j,k)$ in (2.24) would give the stronger estimate directly.","The tensor-form co-positivity reformulation connects the $p>2$ cases to a computational hardness question; if the kernels are special enough to avoid general NP-hardness, a proof for all even $p$ would likely require new structure beyond the $p=2$ mechanism."],"forward_implications":["Theorem 1.1 yields the refined smoothing estimate $\\big\\|\\sum_j \\lambda_j |\\widehat{g_j\\,d\\sigma}|^2\\big\\|_{\\dot H^{1/2}(\\mathbb{R}^n)} \\lesssim \\|(\\lambda_j)\\|_{\\ell^2}$, and hence the orthonormal Stein–Tomas restriction estimate (1.29) for $q\\in[2n/(n-1),\\infty]$ by interpolation.","In the paraboloid case, Theorem 1.4 gives, for $d=1$, the orthonormal Strichartz estimates previously proved by Schatten methods, and it provides a direct Wigner proof of those estimates.","Whenever $p=n+1$, the suggested family (1.12) together with the endpoint X-ray estimate of [19] implies the endpoint orthonormal Stein–Tomas inequality of [34].","For even $p$, the paper recasts the undirected form of (1.12) as the co-positivity of a specific $p$-tensor form whose kernel is built from autocorrelations of surface measure and an X-ray identity."],"supporting_citations":[{"why":"Supplies the Schatten duality principle that converts an orthonormal weighted estimate into a Hilbert–Schmidt norm bound and gives the orthonormal Stein–Tomas comparison.","marker":"[34]"},{"why":"Provides the counterexample showing the global single-function Mizohata–Takeuchi inequality is false, which motivates why the orthonormal-system statement is the new true global form.","marker":"[13]"},{"why":"Companion preprint quoted for the spherical phase-space identity, the spherical Wigner transform, and the Jacobian-ratio estimate used in the general-hypersurface and Wigner proofs.","marker":"[8]"},{"why":"Gives the classical Moyal identity for Wigner distributions used in the direct parabolic proof of Theorem 1.4.","marker":"[30]"},{"why":"Introduced the orthonormal Strichartz formalism and raised the open problem of a direct Wigner-type proof that this paper addresses.","marker":"[33]"},{"why":"Supplies the one-dimensional orthonormal Strichartz estimate recovered from Theorem 1.4 via the Plancherel identity for the space-time X-ray transform.","marker":"[17]"},{"why":"Gives the endpoint X-ray estimate that connects the proposed $p=n+1$ case to the orthonormal Stein–Tomas inequality.","marker":"[19]"},{"why":"Supplies the analytic interpolation theorem used in the proof of Proposition 5.1.","marker":"[29]"},{"why":"Provides the $L^p$ Radon-transform estimate (5.2) used in Proposition 5.1 to replace the X-ray norm by a Sobolev norm.","marker":"[55]"}],"fun_headline_variants":["Orthonormal systems satisfy global Mizohata–Takeuchi bound","Orthonormality controls interference in Mizohata–Takeuchi","Wigner transforms yield Mizohata–Takeuchi for orthonormal systems","Mizohata–Takeuchi bound uses geometric norm for orthonormal systems"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that two quoted results from the companion preprint [8] are correct — the spherical phase-space identity $|\\widehat{g\\,d\\sigma}|^2 = X^* W_{S^{n-1}}(g,g)$ and the Jacobian-ratio estimate $J(u,u')/\\widetilde J(u,u') \\le c\\,Q(S)^{(5n-8)/2}$ — since if either is wrong the Wigner-based and general-hypersurface theorems fail, even though Theorem 1.1 would survive.","fun_headline_variants_meta":{"raw":{"variants":["Orthonormal systems satisfy global Mizohata–Takeuchi bound","Orthonormality controls interference in Mizohata–Takeuchi","Wigner transforms yield Mizohata–Takeuchi for orthonormal systems","Mizohata–Takeuchi bound uses geometric norm for orthonormal systems"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000585,"raw_usage":{"total_tokens":2722,"prompt_tokens":887,"completion_tokens":1835,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":503,"completion_tokens_details":{"reasoning_tokens":1764}},"tokens_in":503,"tokens_out":1835,"duration_ms":14263,"temperature":1.0,"reasoning_tokens":1764,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:57:22.042813+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a Cantor-type set $K\\subset S^{n-1}$ with very small surface measure and $K^\\diamond$ nearly the whole sphere, and check numerically whether the distribution $X_0^* 1_{K^\\diamond} - |\\widehat{1_K\\,d\\sigma}|^2$ has nonnegative expectation against every smooth compactly supported test weight $w$; a single negative expectation would disprove the tomographic estimate and with it Theorem 1.1.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Schatten duality principle that converts an orthonormal weighted estimate into a Hilbert–Schmidt norm bound and gives the orthonormal Stein–Tomas comparison."},{"cited_title":"Bennett, S","cited_arxiv_id":null,"evidence_quote":"Companion preprint quoted for the spherical phase-space identity, the spherical Wigner transform, and the Jacobian-ratio estimate used in the general-hypersurface and Wigner proofs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the classical Moyal identity for Wigner distributions used in the direct parabolic proof of Theorem 1.4."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduced the orthonormal Strichartz formalism and raised the open problem of a direct Wigner-type proof that this paper addresses."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the one-dimensional orthonormal Strichartz estimate recovered from Theorem 1.4 via the Plancherel identity for the space-time X-ray transform."},{"cited_title":"Christ,Estimates for thek-plane Transform, Indiana U","cited_arxiv_id":null,"evidence_quote":"Gives the endpoint X-ray estimate that connects the proposed $p=n+1$ case to the orthonormal Stein–Tomas inequality."},{"cited_title":"Fefferman, E","cited_arxiv_id":null,"evidence_quote":"Supplies the analytic interpolation theorem used in the proof of Proposition 5.1."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the $L^p$ Radon-transform estimate (5.2) used in Proposition 5.1 to replace the X-ray norm by a Sobolev norm."}],"review_version":1}