{"id":"4789e3b7-ea1d-4b79-8332-60756bc5f800","arxiv_id":"2607.09946","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"On manifolds without conjugate points the monochromatic nodal-volume variance of Riemannian random waves is O(ℓ^{1-n}/log ℓ), improving Canzani–Hanin by more than a square and establishing Berry cancellation.","lead":"The paper proves sharper high-frequency variance bounds for the nodal volume of Riemannian random waves on compact manifolds, including the monochromatic regime. On manifolds without conjugate points the monochromatic variance decays almost like the square of prior bounds, confirming Berry cancellation.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The reader correctly isolates the single external pillar (Keeler’s logarithmic gain) and correctly notes that every monochromatic claim collapses without it. That pillar is a published theorem under precisely the geometric hypothesis the authors assume; the paper’s own contributions (the chaos expansion of [44], the abstract criterion Theorem C, and the reduction Theorem B) are modular and self-contained once the scaling error is granted. No free parameters, no circularity, and no internal contradiction appear. The residual gap to sharpness is already flagged by the authors and does not undermine the stated O(ℓ^{1-n}/log ℓ) upper bound. Consequently the ACCEPT verdict stands without adjustment.","tokens_in":48587,"tokens_out":559,"duration_ms":5662,"concrete_test":"Verify that Keeler’s Theorem 1.1 (or the equivalent statement in [26, Thm. 1.3]) indeed yields a C^3 (or higher) remainder of size O(1/(ℓ log ℓ)) uniformly on the fixed-range set {dist(x,y)≤(1/2)inj(M)} that is used in Theorem B and Proposition 3.6.1; if the published statement only controls C^0 or C^1, the jet norms required by Theorem C would need an extra interpolation argument.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The monochromatic variance bound of Theorem A / Corollary 3.4.3 rests on Keeler’s logarithmic improvement of the pointwise Weyl law (Theorem 3.5.3(3)), which supplies the fixed-range scaling error δ_M_{[0,ℓ]}=O(1/(ℓ log ℓ)). Theorem B then converts this into δ_M_{[ℓ-1,ℓ]}=O(1/log ℓ) in fixed macroscopic range, which is exactly the hypothesis needed for the abstract variance machine (Theorem C) and the fourth-chaos integral estimate of Proposition 3.6.1. The paper states the geometric hypothesis (no conjugate points) cleanly, cites Keeler [26] for the only external analytic input, and derives every subsequent step (Berry cancellation, comparison with Canzani–Hanin) from that input. No internal inconsistency or hidden gap appears in the reduction. The residual factor ℓ/log ℓ to the conjectured Θ(ℓ^{-n}) rate is openly acknowledged and does not affect the claimed upper bound.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper studies the high-frequency variance of the normalized nodal volume Y_M(I_ℓ) of Riemannian random waves on a compact manifold, for spectral windows I_ℓ of various widths (full, weakly monochromatic, and monochromatic). Theorem A gives exact order Θ(ℓ^{-n}) for non-monochromatic windows (α<1) with no geometric assumptions, and the improved upper bound O(ℓ^{1-n}/log ℓ) for monochromatic windows on manifolds without conjugate points (Corollary 3.4.3). This improves the Canzani–Hanin O(ℓ^{-(n-1)/2}) bound by more than a square and implies Berry cancellation (Var = o(1/N)). The proofs combine Kac–Rice near the diagonal, the authors’ Wiener–Itô chaos expansion from the companion paper [44], and a new analysis of the scaling error in the pointwise Weyl law for arbitrary windows (Theorem B), all fed into an abstract variance machine (Theorem C) that requires only near- and off-diagonal covariance bounds.","tokens_in":48852,"tokens_out":667,"duration_ms":5857,"significance":"The monochromatic variance bound on manifolds without conjugate points is a genuine advance: it is the first result that reaches spectral windows of size O(1) on general chaotic geometries and yields a form of Berry cancellation that was previously known only for spheres, tori, and Euclidean space. The abstract machinery of Theorem C and the scaling-error reduction of Theorem B are reusable tools that cleanly separate geometry (Keeler’s logarithmic Weyl-law improvement) from probabilistic estimates. The residual factor ℓ/log ℓ to the conjectured Θ(ℓ^{-n}) rate is openly acknowledged, so the claimed upper bound is not overstated. The logical chain is fully written and the only external analytic input (Keeler) is correctly cited.","major_comments":[],"minor_comments":[{"comment":"The notation for the scaling error δ_M_I(α,C^k,[ρ',ρ]) is introduced in Definition 3.5.2 and then immediately abbreviated; a short reminder of the range and order of derivatives when the abbreviation is first used in Theorem B would help the reader.","section":null},{"comment":"In the statement of Theorem C the constant T_0 is described only as “big enough (e.g. 77)”; a brief indication of how it arises from the eccentricity and short-correlation thresholds would make the quantitative claim more transparent.","section":null},{"comment":"A few typographical slips remain (e.g. “Acknowlegdments”, “Hoermander”, “varaince”). They do not affect readability but should be corrected.","section":null},{"comment":"The comparison with the sphere (Section 3.8) is clear, yet a one-sentence pointer to the precise rate known for spherical harmonics (Eq. (3.2)) already in the introduction would orient the reader earlier.","section":null}],"recommendation":"accept","confidential_remarks":"The paper is a natural and substantial sequel to the authors’ own companion work [44]. The dependence is cleanly declared and does not create circularity. The result sits comfortably in the probabilistic spectral-geometry literature and is appropriate for a strong probability or analysis journal."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The headline result is real: on compact manifolds without conjugate points they get Var(Y_M([ℓ-1,ℓ])) = O(ℓ^{1-n}/log ℓ), more than a square better than Canzani–Hanin, and that is enough to see Berry cancellation (variance o(1/N)) for genuine monochromatic windows. For non-monochromatic windows they get the exact Θ(ℓ^{-n}) order with no geometric assumptions. That is the first time the monochromatic cancellation statement appears for general chaotic manifolds rather than spheres or tori.\n\nWhat works well is the modular structure. Theorem C is a clean abstract variance machine that only needs a near-diagonal C^3 bound and an off-diagonal short-correlation bound; Theorem B reduces the scaling error for arbitrary spectral windows to the Euclidean Bessel remainder plus the ordinary pointwise Weyl error. Once you feed Keeler’s logarithmic gain into that machine you recover the monochromatic rate. The companion chaos expansion is used as an independent analytic identity, not as a circular input. The proofs are written out, the geometric hypotheses are stated cleanly, and the residual factor ℓ/log ℓ to the conjectured Θ(ℓ^{-n}) is acknowledged rather than papered over.\n\nThe soft spot is exactly the one the stress-test flags: everything monochromatic rides on Keeler’s fixed-range O(1/(ℓ log ℓ)) improvement. If that gain is unavailable the bound collapses. That is not a hidden gap; the paper cites it and works under the stated hypothesis. The sphere is correctly excluded (conjugate points), so the method does not claim more than it can deliver. No free parameters, no data selection, citation pattern looks normal.\n\nThis is for people working on random waves, nodal statistics, or spectral geometry on chaotic manifolds. It is not a broad-audience paper, but the abstract machine is reusable. I would send it to a serious referee; the claims are sharp enough and the reductions clean enough to deserve that time. Worth engaging if you care about the monochromatic regime.","headline":"Solid monochromatic variance improvement on no-conjugate-point manifolds, with reusable abstract machinery; residual gap to sharp rate is openly left open.","tokens_in":49448,"tokens_out":509,"would_cite":true,"duration_ms":8291,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60G60","58J50","35P20"],"pacs":[],"model":"grok-4.5","headline":"On manifolds without conjugate points, monochromatic random waves have nodal-volume variance that decays like ℓ^{1-n}/log ℓ, more than squaring the previous bound and forcing Berry cancellation.","keywords":["Yau conjecture","nodal volume","Riemannian random waves","monochromatic regime","Berry cancellation","pointwise Weyl law","Wiener chaos","Kac-Rice formula"],"falsifier":"Compute or rigorously bound the C^1 scaling error δ_M of the monochromatic spectral projector on a manifold without conjugate points in a fixed positive range; if the error remains only O(1) rather than O(1/log ℓ), the monochromatic variance upper bound collapses to the weaker Canzani–Hanin order.","tokens_in":49483,"feed_emoji":"📐","tokens_out":744,"duration_ms":10226,"temperature":0.7,"pith_summary":"Yau’s conjecture predicts that the nodal volume of a Laplace eigenfunction of frequency λ is of order λ. This paper studies the probabilistic version: the normalized nodal volume Y_M(I) of a Riemannian random wave built from eigenfunctions whose frequencies lie in a spectral window I of height ℓ. The authors prove that for any non-monochromatic window the variance of Y_M is exactly of order ℓ^{-n}, while on manifolds without conjugate points the pure monochromatic window of width 1 yields the sharper upper bound O(ℓ^{1-n}/log ℓ). That rate is more than the square of the earlier Canzani–Hanin bound and is already smaller than 1 over the dimension of the eigenspace, which is precisely Berry’s cancellation. The proof rests on a new abstract variance machine (Theorem C) that converts near-diagonal Kac–Rice estimates and off-diagonal chaos estimates into a global Law of Large Numbers once a scaling limit of the covariance holds in a fixed macroscopic range; the required scaling limit is supplied by Keeler’s logarithmic improvement of the pointwise Weyl law.","feed_headline":"Monochromatic nodal volume variance drops by more than a square","feed_subtitle":"On manifolds without conjugate points the bound improves past Canzani–Hanin and forces Berry cancellation","key_machinery":"Theorem C: an abstract quantitative Law of Large Numbers for the nodal volume of an arbitrary C^3 Gaussian field, controlled solely by a near-diagonal C^3 bound and an off-diagonal short-correlation bound on the covariance jet; once a macroscopic scaling limit of the random-wave covariance is available, the theorem yields the stated variance rates.","core_discovery":"On every compact manifold without conjugate points the monochromatic random wave ϕ_{[ℓ−1,ℓ]} satisfies Var(Y_M([ℓ−1,ℓ])) = O(ℓ^{1-n}/log ℓ). Consequently the variance is o(1/N_M), i.e., Berry cancellation occurs, while for every non-monochromatic window the variance is exactly Θ(ℓ^{-n}).","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Monochromatic nodal variance O(ℓ^{1-n}/log ℓ) without conjugate points","Berry cancellation for monochromatic waves on no-conjugate manifolds","Variance bound beats Canzani–Hanin by more than power 2","Monochromatic nodal volume variance is o(1/N_M) on chaotic manifolds","Non-monochromatic windows give exact Θ(ℓ^{-n}) nodal variance"],"cache_read_input_tokens":32896,"weakest_assumption_plain":"The monochromatic covariance must admit a scaling limit with error O(1/log ℓ) on balls of fixed positive radius, which is supplied only by Keeler’s logarithmic refinement of the pointwise Weyl law and fails as soon as that logarithmic gain is lost.","fun_headline_variants_meta":{"raw":{"variants":["Monochromatic nodal variance O(ℓ^{1-n}/log ℓ) without conjugate points","Berry cancellation for monochromatic waves on no-conjugate manifolds","Variance bound beats Canzani–Hanin by more than power 2","Monochromatic nodal volume variance is o(1/N_M) on chaotic manifolds","Non-monochromatic windows give exact Θ(ℓ^{-n}) nodal variance"]},"model":"grok-4.5","effort":"low","cost_usd":0.004458,"raw_usage":{"total_tokens":1311,"prompt_tokens":756,"num_sources_used":0,"completion_tokens":85,"cost_in_usd_ticks":44580000,"prompt_tokens_details":{"text_tokens":756,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":470,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":756,"tokens_out":85,"duration_ms":4503,"temperature":1.0,"reasoning_tokens":470,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T14:23:23.374482+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Compute or rigorously bound the C^1 scaling error δ_M of the monochromatic spectral projector on a manifold without conjugate points in a fixed positive range; if the error remains only O(1) rather than O(1/log ℓ), the monochromatic variance upper bound collapses to the weaker Canzani–Hanin order.","supporting_citations":[],"review_version":1}