{"id":"5ea79ba9-65f8-4426-8264-bf4f8e502cb0","arxiv_id":"1908.04050","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For dimensions 5 and 6, the Calderón inverse conductivity problem has uniqueness for conductivities in W^{1+(d-5)/(2p)+,p}, with d ≤ p < ∞, improving prior regularity thresholds.","lead":"This paper proves that the electrical conductivity inside a body in five or six dimensions is uniquely determined by boundary measurements under weaker smoothness assumptions than previously known. It does so by extending Tao's bilinear restriction estimate, a harmonic analysis tool that may be useful beyond the conductivity problem.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The bilinear theorem for the sphere/hemisphere rests on an unproved transversality claim in §4.5.1; the Kakeya estimate (69), and hence Theorem 3, depends on it.","rationale":"The paper's advertised improvement is Theorem 3, and the proof route is clear: Theorem 2 follows from Theorem 8, whose proof uses the bilinear restriction bound Theorem 7. The detailed proof in Section 4 covers only the paraboloid; the passage to the sphere/hemisphere is contained in the short 'Additional Remarks' of §4.5.1. The reader identified exactly the same weak point: the 'intuitively clear' transversality of the cone generated by S1 to tubes from S2. This is the step that makes the Kakeya estimate (69) true, and without (69) the induction on scales in §4.2 does not close. I agree with the reader's assessment. The gap is internal and is explicitly flagged by the phrasing 'intuitively clear,' so under the reviewing rule it must count against full verification. It is not a disagreement with an external consensus: even if the statement is true, the proof as written is incomplete at a load-bearing point. A single analytic check can decide whether the geometric claim is true: compute the relevant angle lower bound for the hemisphere. If the bound holds, the gap is likely fixable by writing out the perturbation argument. If it fails, the bilinear theorem as stated is false and the main theorem is unsupported. For this reason I keep the reader's CONDITIONAL verdict rather than moving to ACCEPT or REJECT.","tokens_in":28947,"tokens_out":14438,"duration_ms":169052,"concrete_test":"Prove (69) for the hemisphere S in (10), for n = 4,5 (the values used in d = 5,6). Place S1 and S2 symmetrically as ξ¹₂ = −a e₁ and ξ²₂ = a e₁, write the cone P̃ explicitly, and compute min over generators of P̃ and over tube directions from S2 of the absolute inner product. Show this minimum is ≥ c > 0 uniformly in a ∈ (0, 1/√2 + 1/10) and in cap radius ρ. If the minimum can be arbitrarily small, Theorem 7 fails and Theorem 3 does not follow. If it is uniformly positive, the transversality step is settled, though the full adaptation of §§4.1–4.4 still needs to be written out.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 2 and Theorem 3 depend on Theorem 7 through the bilinear reduction in §3.1, and Theorem 7 is only proved for the paraboloid (Theorem 7'). The extension to elliptic-type surfaces and to the hemisphere in §4.5.1 is a sketch rather than a proof. The critical place is the Kakeya-type estimate (69): the bound |T1 ∩ ...| ≲ |T2|/(μ²λ¹) uses that all relevant T1 tubes lie in an R^{1/2+δ}-neighborhood of a hyperplane, and that every T2 tube is uniformly transversal to that hyperplane. In the elliptic and hemisphere cases the hyperplane is replaced by a curved cone P̃. For the hemisphere the paper says only: 'It is intuitively clear that the portion of the cone generated by direction from S1 is uniformly transversal to tubes from S2.' No uniform angle lower bound, or estimate depending on the separation and cap parameters, is supplied. Since (69) is exactly what makes the induction-on-scales close and produces the ν^{1/n} factor, this is load-bearing: without this transversality step, the bilinear bound (11) is not established for the surfaces actually used (caps of the characteristic sphere, including antipodal configurations). This is an internal gap flagged by the paper itself, not a disagreement with consensus.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper addresses the Calderón problem of determining a conductivity from boundary measurements. The main result, Theorem 3, asserts uniqueness for d=5,6 when the conductivity lies in W^{1+(d-5)/(2p)+,p}(Ω) for d≤p<∞, improving on earlier results by Haberman and by Ham, Kwon and Lee. The strategy follows the Bourgain-space method: prove that the average of the multiplication-operator norm vanishes (Theorem 2), after reducing to an estimate (Theorem 8) that is obtained via bilinear restriction theory. The novel input is Theorem 7, an extension of Tao's bilinear restriction estimate to surfaces of elliptic type and to the hemisphere. The paper contains a full proof of the paraboloid case (Theorem 7') and only a sketch of the extension to the other surfaces.","tokens_in":29158,"tokens_out":15083,"duration_ms":130863,"significance":"If established, the improvement in regularity is a real advance: for d=5 it reaches W^{1,p}, matching the critical W^{1,d} conjecture when p=d, and for d=6 it improves the exponent by 1/(2p). The paraboloid case of the bilinear theorem is proved in considerable detail, and the reduction from the bilinear estimate to the vanishing theorem is clearly structured. No circularity is apparent: the proof depends on external results (Tao's bilinear theorem, Haberman's Lemma 5.1, Ham-Kwon-Lee's Lemma 4.3) and does not introduce fitted parameters. However, the central bilinear bound for the hemisphere and elliptic surfaces is not fully proved; the load-bearing transversality step is left as an 'intuitively clear' statement, and the summation estimates in Section 3.2 are asserted without details.","major_comments":[{"comment":"The extension of Theorem 7' to the hemisphere is not proved. After reducing to the case where S1 and S2 are symmetric about e1, the paper states: 'It is intuitively clear that the portion of the cone generated by direction from S1 is uniformly transversal to tubes from S2.' This is exactly the statement needed to obtain the Kakeya-type estimate (69) via (71), which in turn closes the induction on scales in §4.2 and yields the ν^{1/n} factor in Theorem 18. Without a uniform lower bound on the angle between the cone P̃ and every T2 tube, the bilinear inequality (11) is not established for the sphere. Please provide a complete proof, with explicit estimates in terms of the separation parameters.","section":"§4.5.1, required for Theorem 7"},{"comment":"For surfaces of elliptic type, the transversality argument is also not rigorous: the sentence 'the inner product is basically equal to ⟨η1−η2, η1'−η2'⟩ for all the pairs ...' glosses over the dependence of the matrix A on the points. The paper needs to prove that A is uniformly close to the identity on the relevant compact ranges and that the lower bound |⟨η1−η2, η1'−η2'⟩| ≥ c > 0 holds uniformly, so that the constant Cδ in (71) does not degenerate.","section":"§4.5.1, elliptic-type surfaces"},{"comment":"The bounds for the terms I and II after (42) are asserted rather than derived. The text says 'We fix λ ≳ M^{-1/2}, and sum first in ν and then in μ. Since p ≥ d ≥ 5, then we get I ≤ c M^{(d-5)/(2p)+ε} ...' but no summation details are given. This is load-bearing: the power M^{(d-5)/(2p)} is exactly the improvement over Haberman's exponent, and the dyadic sums must converge for the range μ ≤ ν < μ^{1/2}, λ ∈ [ν^{1/2},1], with the constraints appearing in the definition of Q. Please provide the complete computation.","section":"§3.2, proof of Theorem 8"},{"comment":"The base case of the induction on scales is not justified. After stopping at R^{(1-δ)^N} ≈ ν^{-1}, the paper states 'If r ≤ ν^{-1}, then we can average over translations of the paraboloid and apply Tao's bilinear to get Kν(r) ≤ Cε r^{1 - (n+2)/(2p) + ε} ν^{1/2}' without proof. This estimate is used to close the recurrence and should be derived explicitly, including the role of the translation averaging and the precise dependence on ν.","section":"§4.2, base case of induction"}],"minor_comments":[{"comment":"The symbol p is used both for the Sobolev index (Theorems 2, 3, 8, 12) and, in Theorem 9 and Lemma 10, for the Hölder dual of p1. This makes the exponents in (25), (30), and (31) ambiguous; please define p in each theorem or use a different letter.","section":"Notation"},{"comment":"The statement reads 'For d ≤ p ≤ 8'; this should be 'd ≤ p < ∞'.","section":"Theorem 12"},{"comment":"The operator is denoted MB_i f in Theorem 2 and Theorem 8 but mB_i f in (42); please make the notation uniform.","section":"Throughout Section 3"},{"comment":"In the sentence 'By symmetry, we can assume that ξ2^2 = -a e1 and ξ1^2 = a e1', the variables ξ1^2 and ξ2^2 are not defined; they appear to be points on the two surfaces, but the notation conflicts with the second coordinate of ξ. Please clarify.","section":"§4.5.1"},{"comment":"Reference [13] is cited as an arXiv preprint (v2, 2019). If the published version differs, the paper should cite the final version and state which lemmas are used.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper's central gap is the extension of the bilinear theorem to the sphere, which is the advertised novelty. The 'intuitively clear' sentence in §4.5.1 is not acceptable in a proof of a central theorem. I would recommend requiring the author to complete the transversality argument and the summation estimates before publication. The rest of the structure is sound."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: the advertised improvement is real if the bilinear estimate holds, but the proof of that estimate for the hemisphere and elliptic surfaces is not complete. Theorem 3 lowers the Calderón uniqueness regularity in d=5,6 to W^{1+(d-5)/(2p)+,p}, which improves on Haberman and on Ham-Kwon-Lee. It is a meaningful step toward the conjectured W^{1,d} threshold, and the main novelty is Theorem 7, an extension of Tao's bilinear theorem to elliptic-type surfaces and the hemisphere for 1 ≤ p' ≤ n/(n-1).\n\nThe paraboloid case (Theorem 7') is proved in detail, with the induction-on-scales and the Kakeya estimate carried out. The reduction from Calderón to the bilinear estimate follows the framework of Haberman and Ham-Kwon-Lee and looks sound; I see no circularity and no fitted parameters. The paper is honest about what it imports from the literature.\n\nThe soft spot is exactly where the stress-test note points. Section 4.5.1 'Additional Remarks' is supposed to extend the proof to the surfaces actually used in the problem, and it is a sketch. The load-bearing point is the Kakeya-type estimate (69): it needs every T2 tube to be uniformly transversal to the cone generated by the T1 directions. For the hemisphere the paper says only that it is 'intuitively clear' that the cone is uniformly transversal to tubes from S2. No uniform angle lower bound is supplied, and the antipodal cap configurations are precisely where this matters. Without (69), the induction on scales does not close and Theorem 7' does not generalize. This is an internal gap flagged by the paper itself, not an artifact of the review.\n\nSecondary issues: Lemma 14 is quoted from an unreviewed preprint with only a sketch, and the summation in Section 3.2 ('sum first in ν and then in μ') is compressed. These are less serious; they are probably fixable with more detail. The primary gap is the missing proof of transversality for the hemisphere.\n\nBottom line: this is a serious paper with a genuine new theorem, but the central proof is not finished. It deserves a serious referee and a request for revision. I would not cite it yet; I would point a student to it as a good problem.","headline":"The paper's advertised regularity improvement depends on a bilinear extension whose load-bearing transversality step is asserted, not proved.","tokens_in":29785,"tokens_out":2932,"would_cite":false,"duration_ms":28106,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35R30","42B10","35J25"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that for dimensions 5 and 6, boundary measurements determine an electrical conductivity with just over one derivative in $L^p$ ($p\\ge d$), improving the previously known uniqueness threshold in Calderón's problem.","keywords":["Calderón problem","electrical impedance imaging","Dirichlet-to-Neumann map","bilinear restriction estimates","complex geometrical optics","Bourgain-type spaces","conductivity uniqueness","regularity threshold"],"falsifier":"Take the hemisphere case of Theorem 7 with two separated caps and compute explicitly the minimal angle between the cone over one cap and the tubes associated with the other: if any admissible pair of caps in the stated range yields angular separation tending to zero as $R\\to\\infty$, then estimate (69) is false and the theorem's proof chain breaks. More directly, the claimed bound $\\|f_\\mu g_\\nu\\|_{p'}\\le C_\\varepsilon\\mu^{n/(2p)-\\varepsilon}\\nu^{1/p-\\varepsilon}$ can be tested on explicit cap data; a violation for arbitrarily small $\\varepsilon$ would disprove the bilinear theorem.","tokens_in":28697,"feed_emoji":"⚡","tokens_out":13037,"duration_ms":124296,"temperature":0.7,"pith_summary":"The paper studies Calderón's inverse problem: can the full boundary measurement of voltage and current determine the electrical conductivity inside a body? It proves uniqueness for conductivities in $W^{1+(d-5)/(2p)+,p}(\\Omega)\\cap L^\\infty$ in dimensions $d=5,6$, with $d\\le p<\\infty$, improving on earlier regularity thresholds. The central new ingredient is a bilinear restriction estimate for surfaces of elliptic type and for the hemisphere, used to control the high-frequency part of the complex geometrical optics argument. On a sympathetic reading, the argument shows the limiting regularity for this method is just above $W^{1,p}$, and the restriction to $d\\le6$ is technical rather than essential.","feed_headline":"Boundary data fix conductivity just above W^{1,p} in d=5,6","feed_subtitle":"A new bilinear restriction estimate lowers the smoothness needed for unique conductivity recovery.","key_machinery":"The load-bearing object is the family of Bourgain-type spaces $X^b_\\zeta$, defined by weights built from the symbol $p_\\zeta(\\xi)=-|\\xi|^2+2i\\zeta\\cdot\\xi$, whose characteristic set $\\Sigma_\\zeta$ is a $(d-2)$-sphere; these spaces measure how concentrated a function's Fourier transform is near the surface where the complex geometrical optics equation degenerates. The proof's main new component is a bilinear restriction theorem for two separated surface patches of elliptic type or of the hemisphere: for functions $f_\\mu$ and $g_\\nu$ whose Fourier supports lie in $\\mu$- and $\\nu$-neighborhoods of such patches, it gives $\\|f_\\mu g_\\nu\\|_{p'}\\le C_\\varepsilon\\mu^{n/(2p)-\\varepsilon}\\nu^{1/p-\\varepsilon}\\|f_\\mu\\|_2\\|g_\\nu\\|_2$ for $1\\le p'\\le n/(n-1)$. A surface of elliptic type is, locally, the graph of a smooth function whose Hessian has all eigenvalues close to one. The estimate is proved through wave packet decompositions, an induction on scales, a decoupling step, and a Kakeya-type transversality estimate, following the bilinear-to-linear strategy.","core_discovery":"The paper's central claim is Theorem 3: for a bounded Lipschitz domain $\\Omega\\subset\\mathbb{R}^d$ with $d=5,6$, if $\\gamma_1,\\gamma_2\\in W^{1+(d-5)/(2p)+,p}(\\Omega)\\cap L^\\infty$ satisfy $\\gamma_j\\ge c>0$ and the Dirichlet-to-Neumann maps $\\Lambda_{\\gamma_1}$ and $\\Lambda_{\\gamma_2}$ agree on the boundary, then $\\gamma_1=\\gamma_2$ in $\\Omega$. The proof shows that the expected value of the norm of the multiplication operator by $B_jf$, acting between the adapted Bourgain-type spaces $X^b_\\zeta$, vanishes as $|\\zeta|\\to\\infty$; from that vanishing, the author extracts complex geometrical optics solutions with negligible error terms, so the products $w_1w_2$ are dense and uniqueness follows. The same mechanism yields Theorem 4 for all $d\\ge7$ once the two conductivities also have matching normal derivatives on the boundary.","pith_inferences":["If one pushes the argument further, the dimension restriction $d\\le6$ should be removable by a more careful localization of the trace condition; in that case Theorem 4 would become unconditional for all $d$.","The same wave-packet and induction-on-scales strategy could be adapted to other inverse boundary value problems whose complex geometrical optics characteristic sets are spheres, such as recovering a magnetic field or a potential.","A numerical verification of the Kakeya-type estimate (69) on the hemisphere, using randomly sampled separated caps, would independently confirm the one step the text leaves as 'intuitively clear' before the full proof is formalized."],"forward_implications":["If Theorem 3 is correct, uniqueness holds for conductivities in $W^{1+(d-5)/(2p)+,p}(\\Omega)\\cap L^\\infty$, $d=5,6$, $d\\le p<\\infty$.","Theorem 4 gives the same conclusion for all $d\\ge7$ when the normal derivatives of the two conductivities agree on the boundary, confirming that the $d\\le6$ restriction in Theorem 3 is a trace-extension issue rather than a failure of the bilinear estimate.","The expected-value vanishing in Theorem 2 holds for every $d\\ge3$, so the new bilinear restriction estimate is available for higher-dimensional inverse problems beyond the two dimensions where the paper states its main uniqueness theorem.","The result narrows the gap to the conjectured $W^{1,d}$ threshold, with the remaining regularity excess measured by the small $(d-5)/(2p)+$ overload."],"supporting_citations":[{"why":"Establishes that the boundary values of the conductivities coincide from the Dirichlet-to-Neumann map, which justifies matching extensions before subtracting them.","marker":"[5]"},{"why":"Provides the rigorous reduction from density of products of CGO solutions to equality of the conductivities.","marker":"[6]"},{"why":"Supplies the expected-value method and the adapted Bourgain-space framework that Theorem 2 refines.","marker":"[11]"},{"why":"Introduces the adapted Bourgain-type spaces used to solve the CGO equation and to measure smallness of the error terms.","marker":"[12]"},{"why":"Provides the bilinear restriction estimates in the Calderón setting and the averaged projection lemma used in the expectation computation.","marker":"[13]"},{"why":"Reduces the uniqueness problem to density of products of complex geometrical optics solutions.","marker":"[24]"},{"why":"The bilinear restriction theorem for paraboloids that the paper extends to elliptic-type and hemispherical surfaces.","marker":"[25]"},{"why":"Supplies the bilinear-to-linear strategy of decomposing supports into caps and using transversality.","marker":"[26]"}],"fun_headline_variants":["Bilinear estimate lowers Calderon smoothness bar in d=5,6","Calderon uniqueness: just above W^{1,p} in 5 and 6 dimensions","Bilinear method delivers Calderon uniqueness just above W^{1,p}","Boundary data pin conductivity just above W^{1,p} in d=5,6"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof of the bilinear restriction theorem for elliptic surfaces and the hemisphere relies on the assertion, sketched rather than proved in Section 4.5.1, that the cone of directions generated by one surface is uniformly transversal to all tubes coming from the other surface; if that 'intuitively clear' transversality failed, the Kakeya-type estimate (69) and hence the vanishing expectation would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Bilinear estimate lowers Calderon smoothness bar in d=5,6","Calderon uniqueness: just above W^{1,p} in 5 and 6 dimensions","Bilinear method delivers Calderon uniqueness just above W^{1,p}","Boundary data pin conductivity just above W^{1,p} in d=5,6"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001141,"raw_usage":{"total_tokens":4698,"prompt_tokens":870,"completion_tokens":3828,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":486,"completion_tokens_details":{"reasoning_tokens":3739}},"tokens_in":486,"tokens_out":3828,"duration_ms":25893,"temperature":1.0,"reasoning_tokens":3739,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:55:03.067843+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the hemisphere case of Theorem 7 with two separated caps and compute explicitly the minimal angle between the cone over one cap and the tubes associated with the other: if any admissible pair of caps in the stated range yields angular separation tending to zero as $R\\to\\infty$, then estimate (69) is false and the theorem's proof chain breaks. More directly, the claimed bound $\\|f_\\mu g_\\nu\\|_{p'}\\le C_\\varepsilon\\mu^{n/(2p)-\\varepsilon}\\nu^{1/p-\\varepsilon}$ can be tested on explicit cap data; a violation for arbitrarily small $\\varepsilon$ would disprove the bilinear theorem.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that the boundary values of the conductivities coincide from the Dirichlet-to-Neumann map, which justifies matching extensions before subtracting them."},{"cited_title":"Haberman","cited_arxiv_id":null,"evidence_quote":"Supplies the expected-value method and the adapted Bourgain-space framework that Theorem 2 refines."},{"cited_title":"Haberman and D","cited_arxiv_id":null,"evidence_quote":"Introduces the adapted Bourgain-type spaces used to solve the CGO equation and to measure smallness of the error terms."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The bilinear restriction theorem for paraboloids that the paper extends to elliptic-type and hemispherical surfaces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the bilinear-to-linear strategy of decomposing supports into caps and using transversality."}],"review_version":1}