{"id":"46ad38c5-a160-4a5a-a3fc-48f265b886d8","arxiv_id":"2411.14987","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A measure is twice Fourier transformable exactly when it is translation bounded and AG transformable, and this powers new Poisson summation and pure point diffraction formulas for weighted model sets.","lead":"This mathematics paper shows how Feichtinger's mild distributions give a single framework for diffraction theory, unifying the older tempered-distribution and Argabright-Gil de Lamadrid measure approaches. It proves when a translation bounded measure can be Fourier transformed twice and uses this to give pure point diffraction formulas for weighted model sets.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the cited W0 Poisson summation formula is standard and the Section 10 extension is justified.","rationale":"I read the paper as a review-style work whose central theorem is Theorem 8.8: a measure is twice AG transformable exactly when it is AG transformable and translation bounded, equivalently when it is translation bounded and has a translation bounded Fourier-pairing on S0. The proof of Theorem 8.8 is self-contained modulo the cited AG and mild-distribution facts, and the surrounding Sections 2-9 build the needed machinery coherently. Section 10 is the main extension and rests on the W0 Poisson summation formula. The reader flagged this as the weakest assumption; I agree it is the only external input that Section 10 depends on, but I do not see it as a load-bearing objection. Since S0 is dense in W0 and both sides of the Poisson summation identity define bounded linear functionals on W0, the formula extends from S0 to W0 by continuity; consequently Theorem 10.2 and the pure point diffraction intensities in Proposition 10.4 are justified. I also verified the projection-injectivity step turns on density of pi_H(L), which is present as an explicit hypothesis. The typo 'Theorem 6.2' in the proof of Proposition 10.4 is evidently meant to be Theorem 10.2 and does not affect correctness. Overall the central claims are well supported, and no verdict change is needed.","tokens_in":43133,"tokens_out":25579,"duration_ms":265649,"concrete_test":"Verify the citation in [33, p.1617]: locate the Poisson summation formula and determine whether it is stated for W0 or only for S0. If it is only stated for S0, complete the density extension: check that S0(G x H) is dense in W0(G x H) and that delta_L and dens(L) delta_{L^0} are bounded on W0, then Eq. (10.1) follows for all g in W0(G), h in W0(H). If the W0 statement is instead absent and the density extension fails, Theorem 10.2 would have a genuine gap.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The only candidate load-bearing input is the Poisson summation formula for W0(G x H), cited to [33, p.1617] in Theorem 10.2. I checked whether this is a real soft spot. W0(G) = {f in W(G) : f_hat in W(G_hat)} is contained in the Wiener algebra and contains S0 densely; both evaluation functionals delta_L and dens(L) delta_{L^0} are bounded on W0(G x H) because L and L^0 are lattices and W0 embeds continuously into W. Hence the W0-level PSF follows from the standard S0-level PSF by density and continuity, so Eq. (10.1) is justified for all g in W0(G), h in W0(H). The injectivity of the projection L^0 -> G_hat used in Proposition 10.4 also checks out: if (0,eta) is in L^0, then eta = 1 on the dense set pi_H(L), so eta = 1. The remaining key ingredients are the established characterization of twice AG transformability and the almost periodicity results, both cited and used consistently. I find no load-bearing flaw in the argument.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops mathematical diffraction theory for translation bounded measures on locally compact abelian groups using Wiener amalgams and Feichtinger's algebra S0(G) as test function spaces. It reviews the construction of bounded uniform partitions of unity, the Wiener algebra and Feichtinger's algebra, mild distributions, and the Fourier theory of measures in the sense of Argabright and Gil de Lamadrid. The central results are Theorem 8.8, characterizing twice AG transformable measures as precisely those that are AG transformable and translation bounded (equivalently, translation bounded and agreeing with a translation bounded measure on the dual group over S0(G)), and Theorem 10.2, a Poisson summation formula for weighted model sets with weights in W0(H), yielding pure point diffraction with intensities dens(L)^2 |\\hat{h}(\\eta)|^2.","tokens_in":43345,"tokens_out":26261,"duration_ms":233640,"significance":"If the proof issues flagged below are corrected, this is a valuable consolidation: it gives a self-contained treatment of BUPUs, identifies translation bounded measures with the dual of the Wiener algebra, subsumes the AG Fourier theory under mild distribution theory, and extends Poisson summation to W0 test functions in a way that directly yields diffraction results for weighted model sets. The paper is careful and largely expository, with detailed proofs of Theorem 8.3 and Theorem 8.8 and explicit statements of the new W0-level results. It contains no fitted parameters or empirically tuned assumptions; the main results are grounded in standard harmonic analysis and properly credited.","major_comments":[{"comment":"The final displayed chain \"ν(ϕ) = ν(pf) = μ(f) = μ~(pϕ)\" with f = pϕ is not valid under the paper's own conventions. By Theorem 6.5(a), p(pϕ) = ϕ~, so ν(pf) = ν(ϕ~), not ν(ϕ); and μ(f) = μ(pϕ) is not μ~(pϕ) in general. The argument can be repaired by taking f = p(ϕ~), because then p(p(ϕ~)) = ϕ and μ(p(ϕ~)) = μ((pϕ)~) = μ~(pϕ). This correction is necessary for the proof of the claimed equivalence.","section":"Theorem 8.8, proof of (c)⇒(a)"},{"comment":"With σ = (ι\\widehat{\\mu})^\\wedge, the double-Fourier identity p(pτ) = τ~ from Theorem 6.5(a) gives pσ = (ι\\widehat{\\mu})~, not pσ = ι\\widehat{\\mu} as stated. The conclusion that ι\\widehat{\\mu} belongs to ι(M^8(\\widehat{G})) still follows because reflection preserves translation boundedness, but the displayed equality \"pσ = ιpμ\" is false and the proof must be rewritten to use pσ = (ι\\widehat{\\mu})~.","section":"Theorem 8.3, proof of the forward direction"},{"comment":"The proof refers to \"Theorem 6.2\" when applying the Poisson summation formula for weighted model sets; this should be Theorem 10.2. While this is a citation slip rather than a mathematical error, it appears in the proof of the main diffraction result of the section and should be fixed.","section":"Section 10.2, Proposition 10.4"}],"minor_comments":[{"comment":"There are several typographical errors, including \"Skecth of proof\" (Section 5.2), \"bouded\" and \"translation bouded\" (Remark 6.2), and \"admissble\" (Section 10.3).","section":"Throughout"},{"comment":"The sentence \"recalling W^1(G) ⊆ S1_0(G)\" is imprecise: W^1(G) is the dual of the Wiener algebra, not literally a subspace of S1_0(G). What is meant is the canonical embedding of translation bounded measures into mild distributions, i.e., M^8(G) ⊆ S1_0(G) via restriction.","section":"Section 9.3"},{"comment":"The expression \"|1_W R L1ppHq\" is garbled; it should presumably read \"|\\widehat{1_W}| ∉ L^1(\\widehat{H})\".","section":"Section 10.3, last paragraph"},{"comment":"The remark refers to [60, Lem. 5.2] for the implication ω_h∈MT(G) ⇒ \\widehat{h}∈L^1(\\widehat{H}); adding one line explaining why this is consistent with the W0 condition would improve readability.","section":"Remark 8.7"}],"recommendation":"major_revision","confidential_remarks":"The reflection/sign issue in the proofs of Theorems 8.3 and 8.8 looks like a systematic notational slip about the Fourier transform of measures versus distributions, rather than a fatal flaw: the theorem statements appear correct, and the repairs are local (insert appropriate reflections). I therefore do not recommend rejection. The paper is largely a review but with the W0-level Poisson summation and diffraction applications it contains original material that fits the journal's scope. Once the proofs are corrected and the minor issues are addressed, it should be acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Plain take: this is a solid review-and-extension paper, not a breakthrough, and it knows it. The central characterization of twice AG transformable measures (Theorem 8.8) is correctly credited to Feichtinger's 1979–80 work, so don't sell it as new. The new content is Section 10: the Poisson summation formula and pure point diffraction for weighted model sets with weights in W0(H) on general LCA groups, extending the Euclidean results of Richard–Strungaru and Matusiak. That is a legitimate extension, and the arguments there are sound.\n\nWhat the paper does well: it gives a genuinely self-contained treatment of Wiener amalgams, Feichtinger's algebra, mild distributions, and the AG theory, and it shows how these fit together. Theorem 8.8 is proved in detail from definitions, and the weak* topology arguments in Section 9 are careful. The exposition is clear enough to serve as a reference for someone wanting to learn this material. The authors are honest about provenance — the self-citations are appropriate for a review of their own framework.\n\nSoft spots, in proportion: the load-bearing input for Section 10 is the W0 Poisson summation formula, cited from [33] rather than proved. That might look like a hole, but the stress-test note is right: W0 is contained in the Wiener algebra, the lattice evaluation functionals are bounded there, and the S0-level PSF extends by density and continuity. So it's a minor omission, not a flaw. The paper also restricts to second countable groups for the van Hove machinery and waves away the general LCA case as straightforward. It probably is, but that's a small unproven step. Finally, Proposition 10.4 leans on almost periodicity results from [44]; again, standard in this community.\n\nWho is this for? Diffraction theorists and harmonic analysts working with model sets and almost periodic measures. They will get a clean unified framework and a useful extension to W0 weights. I'd cite it if I was working on weighted model sets.\n\nRecommendation: send it to peer review. It deserves a serious referee. With a request to either prove or explicitly defer the PSF input — and to say a bit more about the non-second-countable extension — it should be acceptable. This is honest, careful mathematics.","headline":"A careful, honest review that unifies existing diffraction frameworks via mild distributions; the genuinely new piece is the W0-weighted model set extension in Section 10, which holds up on inspection.","tokens_in":43886,"tokens_out":1728,"would_cite":true,"duration_ms":19402,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["43A25","52C23","42B10"],"pacs":[],"model":"deepseek-v4-flash","headline":"For measures on locally compact abelian groups, double Fourier transformability is equivalent to translation boundedness, unifying diffraction theory.","keywords":["twice Fourier transformable measures","mild distributions","Segal algebra S0(G)","translation bounded measures","diffraction theory","weighted model sets","Poisson summation formula","Wiener amalgam spaces"],"falsifier":"Compute the Poisson summation identity $\\omega_h(g) = \\operatorname{dens}(L)\\,\\omega_{\\hat h}(\\hat g)$ for a cut-and-project scheme with non-euclidean groups, such as a $p$-adic internal space, and a weight $h$ in $W_0(H)$ but not in the Schwartz–Bruhat class; a single pair $(h,g)$ where the two sides differ would disprove Theorem 10.2 and the pure point diffraction conclusion. Independently, a translation bounded measure $\\mu$ whose mild Fourier transform is a Radon measure that is not translation bounded would disprove Theorem 8.8.","tokens_in":42936,"feed_emoji":"💎","tokens_out":10903,"duration_ms":101293,"temperature":0.7,"pith_summary":"The paper shows that the class of measures that can be Fourier transformed twice in the AG Fourier theory—where a measure's Fourier transform is required to be a measure—is exactly the class of AG-transformable measures that are translation bounded. Equivalently, a translation bounded measure on a locally compact abelian group is twice transformable precisely when its Fourier transform as a mild distribution is again a translation bounded measure. Because every translation bounded measure is automatically a mild distribution, the diffraction theory of translation bounded measures can be developed in one test-function framework, subsuming both the tempered-distribution approach and the AG measure approach. For weighted model sets with weights in the Segal algebra $W_0(H)$, the paper derives pure point diffraction with explicit Bragg intensities, using a Poisson summation formula for the underlying lattice.","feed_headline":"Twice Fourier transformable measures are exactly translation bounded","feed_subtitle":"A single test-function framework unifies measure Fourier theory and gives weighted model sets pure point diffraction.","key_machinery":"The machinery is built from bounded uniform partitions of unity (BUPUs), which control local support, uniform boundedness, and finite overlap of a partition of unity on arbitrary locally compact abelian groups. The Wiener algebra $W(G)$ is defined by absolute summability of local pieces, and its dual is identified with translation bounded measures. The smaller Segal algebra $S_0(G)$ is defined from the Fourier algebra in the same way; its dual, the mild distributions, carries a bijective Fourier transform and contains translation bounded measures. The load-bearing identity is the Poisson summation formula for the lattice $L$ in $G\\times H$, valid on the larger Segal algebra $W_0(G\\times H)$, which turns the lattice comb into $\\operatorname{dens}(L)$ times the dual lattice comb and drives the pure point diffraction computation.","core_discovery":"On the paper's own terms, the central discovery is Theorem 8.8: for a measure $\\mu$ on an LCA group $G$, double AG transformability is equivalent to AG transformability plus translation boundedness, and also equivalent to the existence of a translation bounded measure $\\nu$ on the dual group with $\\mu(f)=\\nu(\\hat f)$ for all $f$ in $S_0(G)$. The second transform therefore requires no separate existence argument once a translation bounded measure has a measure-valued Fourier transform on the dense test space. The same mild-distribution machinery turns the diffraction measure of a translation bounded measure into the Fourier transform of its autocorrelation. For weighted model sets with weight $h$ in $W_0(H)$, the Fourier transform is $\\operatorname{dens}(L)$ times the dual weighted comb, so the diffraction measure is pure point with intensities $\\operatorname{dens}(L)^2 |\\hat h(\\eta)|^2$ at the contributing dual-lattice points.","pith_inferences":["Beyond the paper: if Theorem 8.8 holds, checking double transformability of a translation bounded measure reduces to computing its Fourier transform as a mild distribution and testing whether that distribution is a translation bounded measure.","Beyond the paper: the mild-measure framework invites a diffraction theory for Radon measures that are mild but not translation bounded, where the autocorrelation step would have to be reformulated.","Beyond the paper: the authors call extension to arbitrary LCA groups via van Hove nets straightforward but do not prove it; testing the Poisson summation formula and the diffraction statements on non-second-countable groups is a concrete open task.","Beyond the paper: the weight class $W_0(H)$ is larger than the Schwartz–Bruhat class, so the weighted-model-set results are testable for non-smooth weights whose Fourier transforms are integrable enough to lie in the Wiener algebra."],"forward_implications":["Every twice AG transformable measure is translation bounded, and every AG-transformable measure that is translation bounded is automatically twice transformable.","For translation bounded measures, the diffraction measure exists as a translation bounded measure on the dual group and is the mild Fourier transform of the autocorrelation.","The Fourier–Bohr coefficients of a twice transformable measure equal the point masses of its Fourier transform, so pure point diffraction can be read directly from the transform.","Weighted model sets with weights in $W_0(H)$ are twice AG transformable and strongly almost periodic, and their diffraction is pure point with intensities $\\operatorname{dens}(L)^2 |\\hat h(\\eta)|^2$.","The Poisson summation formula on $W_0(G\\times H)$ extends pure point diffraction computations for weighted model sets beyond euclidean cut-and-project schemes."],"supporting_citations":[{"why":"Defines AG transformable and twice transformable measures and poses double transformability as an open problem.","marker":"[1]"},{"why":"Supplies the theorem that twice AG transformable measures are exactly AG transformable and translation bounded, reproduced as Theorem 8.8.","marker":"[15, Thm. C1(ii)]"},{"why":"Identifies the dual of the Wiener algebra with translation bounded measures, the bridge between $W'(G)$ and $M_\\infty(G)$.","marker":"[45, Thm. 6.1]"},{"why":"Establishes the minimality property of the Segal algebra $S_0(G)$ used to set up mild distributions.","marker":"[17, Thm. 1]"},{"why":"Gives the Poisson summation formula on $W_0(G\\times H)$ used in Theorem 10.2 for weighted model sets.","marker":"[33, p. 1617]"},{"why":"Provides the almost-periodicity argument that yields a unique autocorrelation and pure point diffraction for strongly almost periodic measures.","marker":"[44, Thm. 7.6]"},{"why":"Earlier pure point diffraction results for model sets that the present proof extends to $W_0$ weights.","marker":"[59, Thm. 4.10]"}],"fun_headline_variants":["Double Fourier transformability equals translation boundedness","Wiener amalgams unify measure Fourier transforms","Weighted model sets yield pure point diffraction","Single test space simplifies diffraction theory"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything in the model-set section rests on the unproved Poisson summation formula for the Segal algebra $W_0(G\\times H)$, cited from the literature; if that formula fails at the stated level of generality, the conclusion that weighted model set combs are twice transformable with pure point diffraction loses its justification.","fun_headline_variants_meta":{"raw":{"variants":["Double Fourier transformability equals translation boundedness","Wiener amalgams unify measure Fourier transforms","Weighted model sets yield pure point diffraction","Single test space simplifies diffraction theory"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000243,"raw_usage":{"total_tokens":1459,"prompt_tokens":803,"completion_tokens":656,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":419,"completion_tokens_details":{"reasoning_tokens":602}},"tokens_in":419,"tokens_out":656,"duration_ms":7046,"temperature":1.0,"reasoning_tokens":602,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:38:53.527138+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the Poisson summation identity $\\omega_h(g) = \\operatorname{dens}(L)\\,\\omega_{\\hat h}(\\hat g)$ for a cut-and-project scheme with non-euclidean groups, such as a $p$-adic internal space, and a weight $h$ in $W_0(H)$ but not in the Schwartz–Bruhat class; a single pair $(h,g)$ where the two sides differ would disprove Theorem 10.2 and the pure point diffraction conclusion. Independently, a translation bounded measure $\\mu$ whose mild Fourier transform is a Radon measure that is not translation bounded would disprove Theorem 8.8.","supporting_citations":[{"cited_title":"Argabright and J","cited_arxiv_id":null,"evidence_quote":"Defines AG transformable and twice transformable measures and poses double transformability as an open problem."}],"review_version":1}