{"id":"3c92375b-cd21-466f-8f68-bde6581b71d9","arxiv_id":"2412.13118","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A unique-continuation principle for sums of fractional Laplacians is proved on Euclidean space and applied to recover anisotropic coefficients and potentials in fractional polyharmonic equations.","lead":"The paper proves that if several functions all vanish on an open region and a weighted sum of their fractional Laplacians also vanishes there, then all of the functions must be identically zero. This entanglement principle yields new uniqueness theorems for inverse problems, including recovery of an anisotropic coefficient without the usual coordinate-change ambiguity.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: two fixable typos in the reduction steps; the essential super-exponential decay assumption is satisfied in all applications.","rationale":"The paper's central claim is the entanglement principle (Theorem 1.2), proved in three steps: analytic interpolation via Pila's theorem, singularity analysis of a meromorphic extension, and spherical mean support. I checked the transitions. The mollification argument from Theorem 1.2 to Theorem 3.1 is valid after deleting one redundant b_k; the singularity analysis correctly yields the moment identities (3.21)/(3.25) under condition (H), including the odd-dimensional half-integer exclusion; Proposition 3.8 is the only place where super-exponential decay is indispensable, and it is used exactly as the assumption states. The inverse-problem proofs apply the principle only to compactly supported functions or to functions that vanish outside a bounded set, so the decay hypothesis is available in all applications. The two typos, if read literally, break intermediate steps, but they are evident and do not affect the mathematical architecture. The cited tools are standard, and the growth bounds in Lemma 3.5 are consistent with the requirements of Pila's theorem. I therefore see no reason to change the conditional verdict; fixing the typos and stating the decay assumption explicitly is sufficient.","tokens_in":31458,"tokens_out":31037,"duration_ms":271734,"concrete_test":"Re-derive the reduction step in the proof of Theorem 1.2 with the corrected definitions \\tilde v_{k,\\epsilon}=u_k*\\psi_\\epsilon and v_{k,\\epsilon}=b_k(-\\Delta)^{\\lfloor s_k \\rfloor}\\tilde v_{k,\\epsilon}, and verify that the identity (3.3) follows from (1.4). Independently, rerun the second entanglement application in Theorem 1.4 with O=U\\setminus\\Omega and check that both conditions u_k|_O=0 and the nonlocal equation on O hold. If either check fails, the written proof is incomplete.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof is coherent apart from two localized, fixable typos. (1) In the proof of Theorem 1.2, \\tilde v_{k,\\epsilon} is defined with a factor b_k and then v_{k,\\epsilon}=b_k(-\\Delta)^{\\lfloor s_k \\rfloor}\\tilde v_{k,\\epsilon}; substituting into (3.3) would require \\sum_k b_k^2(-\\Delta)^{s_k}(u_k*\\psi_\\epsilon)=0, which does not follow from the hypothesis. Removing the b_k from either definition restores the argument. (2) In the proof of Theorem 1.4, Theorem 1.2 is applied with O=U, but \\tilde w is only known to vanish on U\\setminus\\Omega; the correct choice is O=U\\setminus\\Omega, where the nonlocal equation also vanishes. The genuinely load-bearing assumption is the super-exponential decay of Definition 1.1, used only in Proposition 3.8 to convert moment conditions into vanishing spherical means; this is explicitly acknowledged in Remark 1.3(ii), and every application uses compactly supported functions, so the assumption is satisfied. I find no threat to the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves an entanglement principle for the fractional Laplacian on Euclidean space: under a super-exponential decay assumption and an exponent gap condition (H), if finitely many distributions u_k all vanish on a nonempty bounded open set O and a nontrivial linear combination of distinct fractional Laplacians of the u_k vanishes on O, then every u_k is identically zero. The proof passes through a smooth mollified version, constructs a holomorphic function whose zero set is forced by Pila's interpolation theorem, meromorphically continues it, analyzes its poles to obtain vanishing moment identities, and finally uses support theorems for spherical means. The paper also applies the principle to an anisotropic Calderón-type problem and to a zeroth-order coefficient recovery problem, and proves a Runge approximation property for fractional polyharmonic equations.","tokens_in":31659,"tokens_out":30643,"duration_ms":277488,"significance":"If the corrections below are made, this is a substantial contribution. The main theorem genuinely extends the compact-manifold entanglement principle of [FKU24] to the noncompact Euclidean setting, where the heat kernel does not decay exponentially and the singularity analysis has to be redone. The proof strategy is original and well structured: mollification reduces to smooth super-exponentially decaying functions; the holomorphic interpolation step is handled with Pila's theorem; the meromorphic continuation is explicit; and the spherical mean support theorem is used cleanly. The paper is also honest about the role of the super-exponential decay assumption, identifying it in Remark 1.3(ii) as needed only for the final spherical-mean step, and noting that all applications use compactly supported functions. The inverse problem applications are interesting and, modulo the self-adjointness issues discussed below, would be new. The main central derivation appears sound; the difficulties lie in a few localized but load-bearing spots in the reduction and in the application sections.","major_comments":[{"comment":"The reduction to Theorem 3.1 contains a factor mismatch. The proof defines \\tilde v_{k,\\epsilon}=b_k(u_k*\\psi_\\epsilon) and then v_{k,\\epsilon}=b_k(-\\Delta)^{\\lfloor s_k\\rfloor}\\tilde v_{k,\\epsilon}. With these definitions, \\sum_k(-\\Delta)^{\\alpha_k}v_{k,\\epsilon} equals \\sum_k b_k^2(-\\Delta)^{s_k}(u_k*\\psi_\\epsilon), whereas convolution of the hypothesis only gives \\sum_k b_k(-\\Delta)^{s_k}(u_k*\\psi_\\epsilon)=0 on \\tilde O. Thus the claimed verification of (3.3) is incorrect as written. Removing the factor b_k from either \\tilde v_{k,\\epsilon} or v_{k,\\epsilon} restores the argument. Because this step is the bridge from Theorem 1.2 to the smooth Theorem 3.1, it must be corrected.","section":"§3, proof of Theorem 1.2"},{"comment":"The second application of Theorem 1.2 in this proof is made with \"O = U\", but the function \\tilde w is only known to vanish on U\\setminus\\Omega, not on all of U. Since \\tilde w=0 in \\Omega_e and P_0\\tilde w=0 in \\Omega_e, the correct open set is O=U\\setminus\\Omega, where p_k is constant and the required nonlocal equation holds. The first application in the same proof already uses O=U\\setminus\\Omega, so the second occurrence appears to be a typo; as written, however, it invalidates the step.","section":"§4, proof of Theorem 1.4"},{"comment":"The Fredholm analysis for L_A is stated in terms of the kernel K_A=\\ker L_A, and the text immediately before Proposition 2.4 asserts that the adjoint of L_A is L_A. This is false when the multipliers p_k in (1.8) are complex-valued, as the paper explicitly allows: the formal adjoint of p_k(-\\Delta)^{s_k}(p_k\\cdot) is \\bar p_k(-\\Delta)^{s_k}(\\bar p_k\\cdot), not the original term. Consequently the solvability condition (2.10) should involve the kernel of the adjoint operator, not merely K_A, and Lemmas 2.7–2.8 inherit the same problem. The proof can be repaired either by restricting p_k (and hence the constants b_k) to be real-valued, or by reworking the Fredholm alternative and the lemmas with K_{A^*}. In addition, the proof of Lemma 2.7 applies the entanglement principle with \"N=m, u_k=a_k\\zeta, b_k=1\", but the equation available on an open subset of \\Omega_e is \\sum p_k(-\\Delta)^{s_k}(p_k\\zeta)=0, which does not reduce to \\sum(-\\Delta)^{s_k}(a_k\\zeta)=0 unless the p_k are handled explicitly. This step needs to be rewritten; the natural repair is to use u_k=\\zeta with coefficients b_k^2 on a set where p_k\\equiv b_k.","section":"§2.3, Proposition 2.4 and Lemmas 2.7–2.8"},{"comment":"The symmetry identity (2.20) and the integral identity (2.21) are used as the basis for the recovery proof of Theorem 1.6. These identities hold for real-valued potentials q with respect to the Hermitian pairing fixed in §2.1. The statement of Theorem 1.6 only assumes q_j\\in L^\\infty(\\Omega), which ordinarily allows complex-valued potentials; if q is complex, the Dirichlet-to-Neumann map is not symmetric in this pairing and the displayed derivation of (2.21) from (2.20) fails. The authors should either add a real-valuedness assumption on q, or replace the Hermitian pairing by a bilinear pairing in the DN-map identities, as is customary in fractional Calderón problems with complex potentials.","section":"§4, Lemma 2.10 and Theorem 1.6"}],"minor_comments":[{"comment":"The notation \"supp(u^j_\\ell)\\subseteq\\Omega_j\" cannot refer to the full solution u^j_\\ell, because the solutions to the exterior value problem are not compactly supported; it should be the exterior Dirichlet data that are supported in the respective sets. This is a notational slip but should be clarified.","section":"§4, proof of Theorem 1.6"}],"recommendation":"major_revision","confidential_remarks":"The central entanglement principle appears sound and is likely correct after the local correction in the mollification step. The larger concern is that the applications section uses self-adjointness of L_A and of the DN map in places where the stated assumptions do not guarantee it, because complex-valued coefficients are allowed. These are fixable by adding real-valuedness assumptions or by changing the duality framework, but they are not merely cosmetic. The paper's reliance on [FKU24] is appropriate and the Euclidean proof is substantially new."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: read this. The paper gives the first extension of the FKU24 entanglement principle from closed manifolds to R^n, and the applications are stronger than the title suggests—anisotropic principal coefficient recovery with no gauge, which fails for the local Calderon problem. The proof is a serious piece of work: heat-semigroup representation, a Pila-type interpolation argument, meromorphic continuation to the left half-plane, and a spherical-mean support theorem to finish. I checked the structure of the argument rather than every estimate; it is coherent and the tools are standard enough that a good referee can verify it in reasonable time.\n\nWhere it is genuinely new: Theorem 1.2 is not in FKU24, which was compact manifolds only. The noncompact setting forces the meromorphic-continuation step and the decay assumption at infinity. The inverse results—Theorem 1.4 (global anisotropic uniqueness, no gauge) and Theorem 1.6 (potential recovery for fractional polyharmonic equations)—are new, and the gauge-free anisotropic result is the kind of thing people in the fractional Calderon community will sit up for.\n\nSoft spots, in proportion. The proof of Theorem 1.2 as written contains a small algebraic slip: the mollified function \\tilde v_{k,\\epsilon} is defined with a factor b_k and then v_{k,\\epsilon} multiplies by b_k again, which would give b_k^2 in the sum. Removing the b_k from one of the two places fixes it. Similarly, in the proof of Theorem 1.4, the entanglement principle is invoked with O=U but the function \\tilde w is only known to vanish on U\\setminus\\Omega; using O=U\\setminus\\Omega and checking the equation there fixes it. Both are local typos; they do not disturb the architecture.\n\nThe genuinely load-bearing assumption is super-exponential decay at infinity (Def 1.1). It is used exactly where the authors say it is—Step III, Proposition 3.8, converting vanishing moments into vanishing spherical means—and the discussion in Remark 1.3(ii) is honest about it. All applications use compactly supported functions, so the assumption is satisfied there. The odd-dimensional exclusion (sk-sj not an odd multiple of 1/2) looks like a proof artifact, and the authors say so themselves.\n\nBottom line: the paper deserves a serious referee. I would send it out. The typos should be caught in revision; nothing here suggests the central claims are wrong. If you work on fractional inverse problems, cite it; if you run a reading group, this is a good paper to go through—the proof has real ideas and the flaws are instructive.","headline":"Genuinely new noncompact entanglement principle for fractional Laplacians, with two fixable typos in the proof; the inverse problem applications are the main payoff.","tokens_in":32183,"tokens_out":2440,"would_cite":true,"duration_ms":23727,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35R30","26A33","42B37"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves an entanglement principle for fractional Laplacians on Euclidean space: if several functions vanish on an open set and a non-trivial linear combination of their fractional Laplacians of different orders also vanishes…","keywords":["fractional Laplacian","entanglement principle","unique continuation","inverse problems","spherical mean transform","Runge approximation","super-exponential decay","Bernstein functions"],"falsifier":"Construct two Schwartz-class (but not super-exponentially decaying) functions u_1, u_2 with u_1|_O = u_2|_O = 0 and b_1(−Δ)^{s_1}u_1 + b_2(−Δ)^{s_2}u_2 = 0 on some nonempty open set O while u_1, u_2 are not identically zero; such an example would disprove the necessity of super-exponential decay. Alternatively, in odd dimension n, exhibit such vanishing data with s_2 − s_1 = 1/2 (an odd multiple of 1/2) and nonzero u_k, which would show condition (H) is not necessary despite the resonance.","tokens_in":31247,"feed_emoji":"🔗","tokens_out":3522,"duration_ms":34330,"temperature":0.7,"pith_summary":"The paper establishes that distinct fractional powers of the Laplacian cannot be made to 'cancel' on an open set unless each contributing function is already zero. This extends a principle previously known only on compact Riemannian manifolds to the noncompact Euclidean setting, working with tempered distributions that decay super-exponentially at infinity. The result is then used to solve two inverse problems: recovering an anisotropic matrix coefficient in a fractional polyharmonic equation without any gauge ambiguity, and recovering a potential from exterior Dirichlet-to-Neumann measurements. A reader should care because it turns a strong unique-continuation statement into a tool for unique recovery of coefficients in nonlocal equations, a direction that has been driven by the fractional Calderón problem.","feed_headline":"Entanglement principle: linked fractional Laplacians force global zero","feed_subtitle":"Vanishing of several fractional powers on an open set now implies all functions vanish, unlocking new inverse-problem uniqueness.","key_machinery":"The central object is the function F(z) defined for each fixed x in the open set by a Mellin-type integral of the heat semigroup acting on the functions, with Gamma-function prefactors. F(z) is holomorphic on the right half-plane, vanishes at all positive integers because the differential equation holds on O, and admits an explicit meromorphic extension to all of C via a computation with the heat kernel. The poles of this extension are located so that taking residues forces, in even dimensions, all moments ∫ v(y)|x−y|^{2m}dy to vanish, and in odd dimensions all moments ∫ v(y)|x−y|^{2m+1}dy to vanish, for each x in O. These moment conditions, combined with super-exponential decay, imply that the spherical means of each function vanish on all spheres centered in O, and the Helgason-type support theorem for spherical mean transforms gives the global vanishing.","core_discovery":"Theorem 1.2 states the entanglement principle: for n ≥ 2, if a nonempty bounded open set O and finitely many functions u_k in $H^{{-r}}$(R^n) with super-exponential decay satisfy u_1|_O = ... = u_N|_O = 0 and ∑_k b_k (−Δ)^{s_k} u_k |_O = 0 for nonzero constants b_k and exponents s_k satisfying condition (H), then every u_k vanishes identically on R^n. The exponents must satisfy s_k − s_j ∉ Z for even dimensions and s_k − s_j ∉ (1/2)Z for odd dimensions. The proof reduces the statement to a smooth, rapidly decaying version, then uses the heat semigroup to build a holomorphic function whose zeros at positive integers force it to vanish everywhere; singularity analysis of its meromorphic extension yields vanishing moment integrals, and a spherical mean support theorem finishes the argument.","pith_inferences":["A natural testable extension is to replace the finite sum of fractional powers by a general Bernstein function of the Laplacian, for which the same Mellin-transform and pole-analysis strategy might still work.","The absence of a gauge in the anisotropic recovery suggests that adding nonlocal lower-order terms breaks the diffeomorphism invariance of the classical anisotropic Calderón problem; this may extend to other nonlocal perturbations and to partial-data settings.","The half-integer resonance in odd dimensions is flagged by the authors as a likely artifact of the proof; a concrete next step is to analyze the paired pole contributions directly, which could remove condition (H)'s odd-dimensional restriction entirely.","The super-exponential decay requirement is only used in the final spherical-mean step, so the principle might hold under weaker decay if the spherical mean support theorem can be replaced by a more robust argument."],"forward_implications":["The fractional polyharmonic operator ∑_k b_k(−Δ)^{s_k} satisfies the strong unique continuation property: if a sufficiently decaying solution and the whole operator expression vanish on an open set, the solution is zero.","The anisotropic Calderón-type inverse problem for L_A u = 0 with nonlocal lower-order terms has a global uniqueness result in dimensions n ≥ 3, with no diffeomorphism gauge, provided the lower-order terms are nonzero constants near the domain.","The exterior Dirichlet-to-Neumann map for fractional polyharmonic equations uniquely determines the bounded potential q in the domain from partial exterior measurements.","A Runge approximation property holds: any L^2 function in the domain can be approximated arbitrarily well in L^2 by solutions of the fractional polyharmonic equation whose exterior data are supported in a prescribed open set.","The entanglement principle supplies the unique-continuation input needed to prove these results, replacing the single-term fractional Laplacian UCP used in earlier work."],"supporting_citations":[{"why":"The original entanglement principle on closed Riemannian manifolds, whose proof structure and counterexamples this paper extends and adapts to noncompact Euclidean space.","marker":"[FKU24]"},{"why":"Establishes the fractional Calderón problem and the unique continuation property for the single-term fractional Laplacian, which the new principle generalizes and which supplies the DN-map framework and well-posedness tools.","marker":"[GSU20]"},{"why":"Pila's version of Carlson's interpolation theorem is the key tool forcing the holomorphic function F that vanishes on positive integers with controlled growth to be identically zero.","marker":"[Pil05]"},{"why":"Provides the support theorem for spherical mean transforms that converts vanishing moment integrals into vanishing of the function in the final step of the entanglement proof.","marker":"[Qui08]"},{"why":"An injectivity result for the spherical means operator used alongside [Qui08] to conclude global vanishing from zero spherical means centered in an open set.","marker":"[Ram02]"},{"why":"Bartnik's elliptic regularity in weighted Sobolev spaces is used to set up the forward problem for the anisotropic operator and to prove the Fredholm properties of the Cauchy dataset.","marker":"[Bar86]"},{"why":"McOwen's isomorphism for the Laplacian on weighted Sobolev spaces is the basis for the solvability and regularity theory of the anisotropic Poisson equation in dimensions n ≥ 3.","marker":"[McO79]"}],"fun_headline_variants":["Entanglement principle: linked fractional Laplacians force global zero","Fractional Laplacian entanglement implies all functions vanish","If fractional powers link on an open set, all vanish globally","Entangled fractional Laplacians unlock new inverse problems","Zero combination of fractional Laplacians forces global vanishing"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The functions must decay super-exponentially at infinity; without that decay, the step that turns vanishing moment integrals into vanishing spherical averages collapses, and the authors note that some decay assumption appears unavoidable.","fun_headline_variants_meta":{"raw":{"variants":["Entanglement principle: linked fractional Laplacians force global zero","Fractional Laplacian entanglement implies all functions vanish","If fractional powers link on an open set, all vanish globally","Entangled fractional Laplacians unlock new inverse problems","Zero combination of fractional Laplacians forces global vanishing"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000596,"raw_usage":{"total_tokens":2789,"prompt_tokens":944,"completion_tokens":1845,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":560,"completion_tokens_details":{"reasoning_tokens":1765}},"tokens_in":560,"tokens_out":1845,"duration_ms":13917,"temperature":1.0,"reasoning_tokens":1765,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:26:04.948103+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct two Schwartz-class (but not super-exponentially decaying) functions u_1, u_2 with u_1|_O = u_2|_O = 0 and b_1(−Δ)^{s_1}u_1 + b_2(−Δ)^{s_2}u_2 = 0 on some nonempty open set O while u_1, u_2 are not identically zero; such an example would disprove the necessity of super-exponential decay. Alternatively, in odd dimension n, exhibit such vanishing data with s_2 − s_1 = 1/2 (an odd multiple of 1/2) and nonzero u_k, which would show condition (H) is not necessary despite the resonance.","supporting_citations":[],"review_version":1}