{"id":"5e9616f1-5564-4189-bd10-209802e526f8","arxiv_id":"2412.00628","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Under Weyl-type eigenvalue asymptotics, the normalized truncated-trace functional equals the Dixmier-trace noncommutative integral after logarithmic averaging, yielding Szegő limit formulas, a density-of-states theorem, and a quantum-ergodicity-style theorem for spectral triples.","lead":"This mathematics paper shows that, for spectral triples whose eigenvalue counts obey a Weyl law, the average of an operator over a finite spectral truncation equals Connes' noncommutative integral in logarithmic mean, and it links this to quantum ergodicity, Szegő limit theorems, and density of states. It also proposes a notion of ergodicity for spectral triples, with an implication that the Dirac operator becomes quantum ergodic.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the central claim is sound conditional on the Weyl-law assumption, with the main proof steps delegated to [LSZ21] and [HMN24] as the reader noted.","rationale":"The reader's weakest-assumption analysis correctly identifies the Weyl law (Definition 2.3) as the engine of the paper: without it, the counting-function estimates, the simplification in Lemma 2.4, and the block-averaging in Lemma 2.6 all break down. My independent check of the main lines of the proofs found no contradiction or missing hypothesis beyond what the reader listed. The conditionality of the verdict is therefore appropriate, driven by the external citations [LSZ21] for the diagonal formula and [HMN24] for the commutator reduction, and by the fact that the paper does not reproduce the full proofs of these crucial inputs. I did not find a new load-bearing flaw that would change the verdict; the honest non-finding is that the paper is conditionally correct, not unconditionally established.","tokens_in":27778,"tokens_out":51382,"duration_ms":458688,"concrete_test":"Independently verify Lemma 2.4's diagonal formula on the circle spectral triple (D=-i∂_θ, eigenbasis e_n(θ)=e^{inθ}, d=1) for the noncommuting operator A=M_f with f(θ)=e^{iθ}. Compute the normalized Dixmier trace of f⟨D⟩^{-1} using the definition via eigenvalues of the compact operator and compare it with the logarithmic mean of the diagonal matrix elements ⟨e_n, f e_n⟩. If the two disagree, then Lemma 2.4 does not hold for arbitrary A∈B(H) and Theorem 2.7 would require a narrower operator class.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After reviewing the main arguments of Sections 2, 3, and 5, I found no internal inconsistency or fatal mathematical error. The central identification in Theorem 2.7 rests on the Weyl law (Definition 2.3) together with the diagonal formula cited from [LSZ21] (Lemma 2.4) and the block-averaging step (Lemma 2.6). The third equality in (3) is justified by Lemma 2.6 provided the Weyl law is understood in the strong sense of a heat-trace asymptotic, since that implies both N(λ_{n+1})/N(λ_n)→1 and N(λ_n)−N(λ_{n−1}) = o(N(λ_n)). Similarly, Proposition 2.1 requires the local Weyl laws for the operators involved; the Tauberian step for non-positive A can be handled by shifting by ||A||·1. Theorem 3.2 depends on Lemma 3.1, which is a translation of Widom's argument and relies on the cited reduction [D,A] bounded ⇒ [⟨D⟩^{1/2},A] bounded from [HMN24]. These are external dependencies that the reader already flagged; they are genuine sources of conditionality but not demonstrated errors. The quantum-ergodicity section, while suggestive, is not needed for the main noncommutative-integral identification. Overall, the paper's strongest claim is conditional on the Weyl law and on the correctness of the cited references; neither fails in the text itself.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes that the finite-rank normalized trace Tr(P_λ a P_λ)/Tr(P_λ) approximates Connes' noncommutative integral Tr_ω(a⟨D⟩^{-d})/Tr_ω(⟨D⟩^{-d}) for spectrally truncated spectral triples. Under a heat-kernel Weyl law and local Weyl laws it proves an equality after logarithmic averaging (Theorem 2.7), derives a Szegő limit theorem via Widom's commutator estimate (Theorem 3.2), relates the density of states to a Dixmier trace formula (Theorem 5.2), and proposes a notion of classical ergodicity for spectral triples (Definition 6.10) that yields a quantum-ergodicity theorem (Theorem 6.11). The main results are conditional on Weyl-law hypotheses and on several imported results, chiefly from [LSZ21], [Aza+22], and the authors' preprint [HMN24].","tokens_in":27953,"tokens_out":15360,"duration_ms":158730,"significance":"If the results are correct, this is a genuinely useful conceptual bridge: it identifies the truncated traces used in operator-system spectral triples with Connes' integral, and it gives a new interpretation of Szegő limit theorems as instances of noncommutative integration. I found no internal inconsistency in the central derivation of Theorem 2.7; the reader's strongest claim appears sound under the stated Weyl law, provided the imported lemmas are valid. Concrete strengths include the explicit and testable statement of Theorem 2.7, the sharp growth criterion in Proposition 5.1, the detailed descriptions of the noncommutative cotangent sphere in Example 6.9, and the corrections to earlier claims in [GL98, Lemma 2.2] and [Zel96, Corollary (3.1)]. The main weaknesses are that Proposition 2.1 as stated needs a positivity hypothesis for the Tauberian step, and that a load-bearing commutator estimate in Lemma 3.1 is delegated to an unpublished preprint.","major_comments":[{"comment":"The proof applies the Hardy–Littlewood Tauberian theorem from [Fel71, Theorem XII.5.2] to Tr(ae^{-tD²}) for arbitrary a∈B(H). In the standard form used there, the Laplace–Stieltjes transform must correspond to a positive or monotone measure; for non-positive a, the function λ↦Tr(P_λ a P_λ) is not monotone, so the second Tauberian conclusion is not justified as written. The proposition should be restricted to positive operators, or one should assume local Weyl laws for the positive and negative parts of a and decompose accordingly. This is load-bearing because Proposition 2.1 is used in the proof of Theorem 3.2(8) and in the proof of Theorem 6.11.","section":"Section 2, Proposition 2.1"},{"comment":"The first paragraph of the proof imports the key implication '[D,A] bounded ⇒ [⟨D⟩^{1/2},A] bounded' from [HMN24, Theorem 6 and Proposition 5.1]. The subsequent Widom argument is standard, but this imported estimate is exactly what connects the commutator hypothesis to the Hilbert–Schmidt bound needed later in the proof. The manuscript should either state and prove this commutator estimate, or formulate Lemma 3.1 with [⟨D⟩^{1/2},A] bounded as an explicit hypothesis. As written, Theorem 3.2 rests on an unproved external result from a preprint.","section":"Section 3, Lemma 3.1"},{"comment":"The hypotheses say only that the spectral triple has 'local Weyl laws', without specifying for which operators or with what uniformity. Proposition 2.1 gives full convergence only for those operators for which a local Weyl law is assumed. To obtain a single density-one subsequence that works for all A in the separable closure of A, one needs either pointwise convergence on a countable dense set together with a uniform approximation argument, or an explicit uniformity assumption. The statement should make this precise, since Theorem 6.11 is the main quantum-ergodicity conclusion.","section":"Section 6, Theorem 6.11"}],"minor_comments":[{"comment":"In the justification of the third equality in equation (3), the text says 'the Weyl law gives that N(λ_n)/N(λ_{n+1})→1'. Since Lemma 2.6 is applied with ϕ(n)=n+1, the displayed condition should be (N(λ_n)+1)/(N(λ_{n+1})+1)→1; the abbreviation is harmless but should be corrected for consistency.","section":"Section 2, proof of Theorem 2.7"},{"comment":"The final step, 'since we proved that λ_k∼k, it also follows that Tr_ω(M_w)=1', is too quick. One should display the ordered eigenvalues of M_w and note that they are asymptotic to 1/(k+1), so that the logarithmic partial sums converge to 1 for every extended limit.","section":"Section 5, proof of Theorem 5.2"},{"comment":"The argument that |D|-|D_M|⊗1 is a negative-order pseudodifferential operator uses the notation op^{-1+ε}(|D_M|⊗1) from [HMN24] without definition. A one-sentence explanation, or a precise statement of the compactness result being used, would make the example substantially easier to verify.","section":"Section 6, Example 6.9(2)"},{"comment":"The local Weyl law is stated for arbitrary operators a∈B(H) without any positivity condition. Given the fix needed in Proposition 2.1, it would be helpful to add a remark clarifying that in applications a is positive or decomposes into operators for which the local Weyl law is known.","section":"Section 2, Definition 2.3"}],"recommendation":"major_revision","confidential_remarks":"The paper leans heavily on the authors' own prior work: Lemma 2.4 from [LSZ21], Lemma 2.6 from [Aza+22], the commutator estimate in Lemma 3.1 from [HMN24], and Tr_ω(M_w)=1 from the first author's thesis. This is not disqualifying, but the editors should insist that the imported lemmas be stated explicitly and, where they come from a preprint, either published or included as an appendix. The paper's scope fits math.OA and the noncommutative geometry literature well; the main revision should focus on the positivity issue in Proposition 2.1 and on making the hypotheses of Theorem 6.11 precise."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Gordon, if you only read one thing about this paper, read Theorem 2.7. It gives the clean statement that, whenever D² satisfies a Weyl law, the logarithmic average of the normalized trace on spectral projections agrees with Connes' noncommutative integral. That is exactly the bridge needed for truncated or numerical NCG, and it is new as far as I know. Proposition 2.1 is also sound; the non-positive case in the Tauberian step is handled by a shift. The paper does more: it translates Widom's Szegő limit theorem into a statement about spectral triples (Theorem 3.2), it gives a Dixmier trace formula for a density-of-states with an ω-independence criterion (Theorem 5.2, Proposition 5.1), and it proposes a definition of classical ergodicity for spectral triples (Section 6). The examples, especially the noncommutative torus, are concrete and convincing.\n\nThe main soft spot is conditionality from delegation. Lemma 3.1's reduction from [D,A] bounded to the commutator estimate is taken from the authors' preprint [HMN24]; Theorem 3.2's continuity extension is lifted from Widom; Theorem 6.11 is a direct consequence of Zelditch. None of these are demonstrated errors—my tracing of Sections 2 and 5 found the logic sound—but they mean the paper as written is not self-contained at three load-bearing hinges. The dependence on the authors' own corpus ([LSZ21], [Aza+22], [Hek25]) is heavy but not circular in the definitional sense; those are established or sibling results, and the paper says so. The quantum-ergodicity part is the least essential; it is more a translation than a new theorem, though it does correct a false claim in Zelditch (Example 6.12.2), which I appreciate.\n\nOne caveat to keep in mind: the Weyl-law assumption is the engine, and the paper honestly flags that it fails for some spectral triples ([HMN24, Ex 5.7]). That does not undercut the results; it just means the identification between truncated traces and the noncommutative integral is conditional, not unconditional.\n\nWho should read this: anyone working in NCG with truncated spectral triples, and people in quantum ergodicity who wonder what Dixmier traces have to do with eigenfunction averages. It deserves a serious referee. I would accept it for review with the expectation that the referee checks the cited internals, especially [HMN24, Thm 6 & Prop 5.1] and [Zel96, Lemma 2.1]. My own verdict is conditional but leaning positive.","headline":"A genuinely useful bridge between truncated spectral projections and Connes' noncommutative integral, conditional on a Weyl law and on cited internals.","tokens_in":28711,"tokens_out":2982,"would_cite":true,"duration_ms":28306,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["46L87","58B34","35P20"],"pacs":[],"model":"deepseek-v4-flash","headline":"Finite spectral truncations reproduce Connes' noncommutative integral under a Weyl law.","keywords":["noncommutative integral","spectral triple","spectral truncation","Dixmier trace","Weyl law","Szegő limit theorem","quantum ergodicity","density of states"],"falsifier":"Construct an operator $D$ with $\\langle D\\rangle^{-d}\\in L^{1,\\infty}$ but eigenvalue counting $N(\\lambda)\\sim C\\lambda^d/\\log\\lambda$, and take $A$ with $\\langle e_k,Ae_k\\rangle=(-1)^k$. A direct calculation would settle the matter: if the logarithmic average of $\\operatorname{Tr}(P_{\\lambda_n}AP_{\\lambda_n})/\\operatorname{Tr}(P_{\\lambda_n})$ still equals the Dixmier integral of $A$, the Weyl law is not actually necessary, whereas if they differ, the paper's stated regime is genuine.","tokens_in":27375,"feed_emoji":"🎯","tokens_out":11501,"duration_ms":101267,"temperature":0.7,"pith_summary":"This paper proves that, on a spectral triple whose squared Dirac operator satisfies a Weyl law, the normalized finite-rank trace of a spectral truncation equals Connes' noncommutative integral after logarithmic averaging. The equality makes the noncommutative integral numerically and physically accessible, since only finite truncations can be computed or measured. A translation of a classical eigenvalue-limit theorem yields a Szegő limit formula for noncommutative geometry: the noncommutative integral of $f(A)$ is a logarithmic average of truncated traces of $f(P_\\lambda A P_\\lambda)$. The paper then defines classical ergodicity for a compact spectral triple as uniqueness of the invariant vector for a noncommutative geodesic flow, and shows that under local Weyl laws this implies quantum ergodicity of the Dirac operator. If the results hold, Connes' integration formula is not an abstract residue but the logarithmic limit point of concrete finite-dimensional data.","feed_headline":"Spectral truncations match Connes' integral under Weyl's law","feed_subtitle":"Finite normalized traces reproduce the Dixmier integral in logarithmic average, linking noncommutative geometry to quantum ergodicity.","key_machinery":"The carrying object is the normalized Dixmier trace functional $a\\mapsto \\operatorname{Tr}_\\omega(a\\langle D\\rangle^{-d})/\\operatorname{Tr}_\\omega(\\langle D\\rangle^{-d})$, Connes' noncommutative integral, compared against the finite-rank functionals $a\\mapsto \\operatorname{Tr}(P_\\lambda aP_\\lambda)/\\operatorname{Tr}(P_\\lambda)$. Three tools carry the proof: the Weyl law, which supplies eigenvalue growth $N(\\lambda)\\sim C\\lambda^d$ and, through the Hardy–Littlewood Tauberian theorem, converts heat-trace asymptotics into Dixmier-trace coefficients; the logarithmic averaging operator $M$, which lets Cesàro means of diagonal matrix elements be exchanged with trace functionals and with subsequences of spectral projections; and the commutator estimate from [Wid79], which shows $\\operatorname{Tr}(P_\\lambda A(1-P_\\lambda)BP_\\lambda)/\\operatorname{Tr}(P_\\lambda)\\to 0$ and opens the way to the Szegő limit theorem. In the ergodicity half, the noncommutative cotangent sphere $S^*A$ with its flow $G_t$ replaces the geodesic flow, and uniqueness of the vacuum state is the mechanism by which classical ergodicity implies quantum ergodicity.","core_discovery":"Let $D$ be closed self-adjoint with $\\langle D\\rangle^{-d}\\in L^{1,\\infty}$, let $\\omega$ be an extended limit, and write $P_\\lambda=\\chi_{[-\\lambda,\\lambda]}(D)$. Assuming $D^2$ satisfies the Weyl law $\\operatorname{Tr}(e^{-tD^2})\\sim Ct^{-d/2}$, the paper proves for every $A\\in B(H)$ that $$\\frac{\\operatorname{Tr}_\\omega(A\\langle D\\$rangle^{{-d}}$)}{\\operatorname{Tr}_\\omega(\\langle D\\$rangle^{{-d}}$)} =(\\omega\\circ M)\\left(\\frac{\\operatorname{Tr}(P_{\\lambda_n}AP_{\\lambda_n})}{\\operatorname{Tr}(P_{\\lambda_n})}\\right),$$ where $M$ is logarithmic averaging over $n$. Thus the abstract Dixmier-trace integral is, in logarithmic average, the limit of concrete finite-dimensional normalized traces. The paper further proves the Szegő-type identity $$(\\omega\\circ M)\\left(\\frac{\\operatorname{Tr}(f(P_{\\lambda_n}AP_{\\lambda_n}))}{\\operatorname{Tr}(P_{\\lambda_n})}\\right) =\\frac{\\operatorname{Tr}_\\omega(f(A)\\langle D\\$rangle^{{-d}}$)}{\\operatorname{Tr}_\\omega(\\langle D\\$rangle^{{-d}}$)}$$ for self-adjoint $A$ with bounded $[D,A]$ and $f\\in C(\\mathbb{R})$, $f(0)=0$. It closes with a noncommutative definition of ergodicity: a spectral triple is classically ergodic when the flow $G_t$ on $L^2(S^*A)$ has a unique invariant vector, and with local Weyl laws this property forces a density-one subsequence of eigenvectors along which diagonal matrix elements converge to the noncommutative integral.","pith_inferences":["The equality in Theorem 3.2 suggests that classical Szegő-type determinant theorems can be read as computational recipes for Connes integrals: in any setting where a local Weyl law is verified, finite truncations give explicit numerical approximations to $\\operatorname{Tr}_\\omega(f(A)\\langle D\\rangle^{-d})$.","The definition of classical ergodicity opens a numerical route to quantum ergodicity in noncommutative examples: one can approximate $L^2(S^*A)$ and the $G_t$-invariant subspace at finite truncation and test whether its dimension collapses to one as the cutoff grows.","Where the Weyl law fails, the main identity is not claimed; the Fröhlich-functional section hints that finite-rank functionals may still track some thermodynamic limit through heat-kernel summability, but that would be a separate theorem rather than a corollary of this paper."],"forward_implications":["Whenever $D^2$ obeys the Weyl law and an operator $A$ has a convergent truncation sequence $\\operatorname{Tr}(P_\\lambda AP_\\lambda)/\\operatorname{Tr}(P_\\lambda)$, the limit is forced to equal the noncommutative integral of $A$; if $A\\langle D\\rangle^{-d}$ is Dixmier measurable, logarithmic averaging can be replaced by an ordinary limit.","The Szegő limit theorem gives computable approximations of $\\operatorname{Tr}_\\omega(f(A)\\langle D\\rangle^{-d})$ from finite truncations $f(P_\\lambda AP_\\lambda)$, and genuine $\\lambda\\to\\infty$ limits when the moments $\\operatorname{Tr}(A^k e^{-tD^2})$ obey the same power law.","For discrete metric spaces whose ball sizes grow slowly enough, the density of states has a Dixmier-trace formula $\\operatorname{Tr}_\\omega(T M_w)=\\omega\\circ M(\\operatorname{Tr}(T M_{\\chi_{B(x_0,r_k)}})/|B(x_0,r_k)|)$, making the DOS a noncommutative-integral quantity.","Classical ergodicity of a spectral triple, defined as uniqueness of the $G_t$-invariant vector, implies quantum ergodicity: on a density-one subsequence, $\\langle e_j,Ae_j\\rangle$ converges to the noncommutative integral for all $A$ in the algebra.","Among standard examples, compact manifolds with ergodic geodesic flow and the Toeplitz spectral triple are classically ergodic, while almost-commutative manifolds and noncommutative tori are not because their symmetry produces many invariant vectors."],"supporting_citations":[{"why":"Supplies the Dixmier trace diagonal formula and the heat-trace coefficient identity used in Proposition 2.1 and Lemma 2.4.","marker":"[LSZ21]"},{"why":"Supplies the Hardy–Littlewood Tauberian theorem converting heat-trace asymptotics into eigenvalue counting for Proposition 2.1 and Remark 2.2.","marker":"[Fel71]"},{"why":"Supplies the commutator estimate in Lemma 3.1 and the polynomial-to-continuous extension scheme used in Theorem 3.2.","marker":"[Wid79]"},{"why":"Supplies Lemma 2.1 linking uniqueness of the vacuum state to density-one convergence, the core of Theorem 6.11.","marker":"[Zel96]"},{"why":"Provides multiple-operator-integral bounds used in Lemma 3.1 and Example 6.9, and Example 5.7 showing where local Weyl laws can fail.","marker":"[HMN24]"},{"why":"Defines the noncommutative integral as the geometric analogue of integration, the object the truncation functionals approximate.","marker":"[Con94]"}],"fun_headline_variants":["Spectral truncations reproduce Dixmier trace in logarithmic average","Truncated spectra give concrete access to Connes' integral","Quantum ergodicity emerges from spectral truncation limits","Weyl law links finite traces to the noncommutative integral"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on the spectrum of $D^2$ growing like a power of the eigenvalue — the Weyl law $\\operatorname{Tr}(e^{-tD^2})\\sim Ct^{-d/2}$, plus the same power-law growth for the localized variants $\\operatorname{Tr}(ae^{-tD^2})$ — so that heat-trace coefficients and eigenvalue counting both align; without this, the equality between truncated traces and the noncommutative integral is not claimed.","fun_headline_variants_meta":{"raw":{"variants":["Spectral truncations reproduce Dixmier trace in logarithmic average","Truncated spectra give concrete access to Connes' integral","Quantum ergodicity emerges from spectral truncation limits","Weyl law links finite traces to the noncommutative integral"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000611,"raw_usage":{"total_tokens":2913,"prompt_tokens":1087,"completion_tokens":1826,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":703,"completion_tokens_details":{"reasoning_tokens":1756}},"tokens_in":703,"tokens_out":1826,"duration_ms":12216,"temperature":1.0,"reasoning_tokens":1756,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:14:54.750421+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct an operator $D$ with $\\langle D\\rangle^{-d}\\in L^{1,\\infty}$ but eigenvalue counting $N(\\lambda)\\sim C\\lambda^d/\\log\\lambda$, and take $A$ with $\\langle e_k,Ae_k\\rangle=(-1)^k$. A direct calculation would settle the matter: if the logarithmic average of $\\operatorname{Tr}(P_{\\lambda_n}AP_{\\lambda_n})/\\operatorname{Tr}(P_{\\lambda_n})$ still equals the Dixmier integral of $A$, the Weyl law is not actually necessary, whereas if they differ, the paper's stated regime is genuine.","supporting_citations":[],"review_version":1}