{"id":"88f89e82-5c67-491a-99e0-ff9663c6f9e6","arxiv_id":"2607.25808","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For almost every realization of a whole-line ergodic block Jacobi operator, any maximal spectral measure's mass in the positive-Lyapunov region is supported on a Borel set of zero logarithmic capacity.","lead":"This paper proves a sharper version of a classical result about where the spectrum of random Schrödinger-type operators can live, extending it from half-line to whole-line and from scalar to matrix-valued (block) operators. The result resolves an open problem from the Damanik–Fillman book and may change how mathematicians bound the possible support of spectral measures of such operators.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 4.4's Craig–Simon bound is stated for every E for P-a.e. ω, but the cited upper-envelope argument only gives a capacity-zero exceptional set; the main estimate needs this gap resolved.","rationale":"I read the paper in good faith. The flipping construction (Section 3) is clever and the finite-volume identifications are sound; Lemma 3.1 and Lemma 3.2 check out. The determinant asymptotics (Lemma 4.1) follow from standard upper-envelope and Birkhoff arguments. The exterior-power bound (Lemma 4.3) is a correct linear-algebra estimate. The Schnol step at the end is appropriate for the bounded half-line block Jacobi operator J_ω. The single load-bearing weak point is Lemma 4.4: it states a simultaneous-in-E bound for P-a.e. ω that is stronger than what the cited theorems are known to provide. If the correct statement only gives a capacity-zero exceptional set, the proof still goes through after a small modification, so the central claim is plausible and the reader's CONDITIONAL verdict is appropriate. No reason to change the verdict; the concern is exactly the reader's weakest assumption, and the proposed check would settle whether the gap is merely expository or substantive.","tokens_in":12130,"tokens_out":32925,"duration_ms":286014,"concrete_test":"Retrieve the exact statements in Craig–Simon [2] and Goldsheid–Sodin [8] and verify whether (14) is proved for every E∈R for P-a.e. ω, or only for all E outside a capacity-zero set. If the latter, re-run the proof of Proposition 4.2 by enlarging Q_ω to the union of the capacity-zero exceptional sets from Lemma 4.1 and the exterior-power bounds, and confirm that the final conclusion μ_ω(S^+ \\ Q_ω)=0 still follows.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Proposition 4.2, and hence Theorem 1.2, rests on Lemma 4.4: for P-a.e. ω and every E∈R, limsup (1/n) log ‖∧ⁱ T_±(n;E,ω)‖ ≤ L_i(E). This is stronger than what the cited potential-theoretic/Craig–Simon results directly provide. For a fixed E, the subadditive ergodic theorem gives a full-measure Ω_E, but the intersection over all E need not have full measure. The upper-envelope theorem (Theorem 2.2) gives, for each fixed ω in a full-measure set, a capacity-zero exceptional set of E; it does not yield 'every E' unless the exceptional set is empty. The paper's own introduction emphasizes the failure of full limits for non-regular cocycles and the possibility of positive-capacity exceptional sets, which makes the unqualified 'every E' in Lemma 4.4 especially suspect. If Lemma 4.4 is false as stated, Lemma 4.2 is unproved. The gap is repairable: replacing 'every E' by 'all E outside a capacity-zero set Q_{ω,i}' and taking the union of those sets (plus the set from Lemma 4.1) still yields Proposition 4.2 with a capacity-zero exceptional set. But as written, the proof relies on a lemma that is likely only true in the weaker quasi-everywhere form, and the authors do not pinpoint where in [2,8] the strong form is proved.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves an 'ultimate Ishii–Pastur theorem' for whole-line ergodic block Jacobi operators: for P-almost every realization, the restriction of any maximal spectral measure to the set S^+ = {E : L_m(E) > 0} is carried by a Borel set of zero logarithmic capacity. The proof introduces a unitary flip of the whole line into a half-line block Jacobi operator of block size 2m, identifies the finite-volume counting measures with those of the original operator, and obtains a quasi-everywhere lower bound on the smallest singular value of the flipped Dirichlet fundamental matrix via determinant asymptotics and exterior-power growth bounds. In the scalar case A ≡ 1 this settles Problem 4.7.18 of Damanik and Fillman.","tokens_in":12358,"tokens_out":34535,"duration_ms":313701,"significance":"If the proof is correct, this is a significant contribution: it extends Simon's half-line zero-capacity theorem to the whole-line setting and to matrix-valued coefficients, and resolves a published open problem. The flipping construction is an elegant and apparently new mechanism that directly addresses the genuine whole-line difficulty explained in Section 1.1. The argument is largely self-contained, uses standard tools (Thouless formula, upper envelope theorem, Craig–Simon bounds, Shnol theorem) without ad-hoc assumptions, and the main theorem is falsifiable in the usual spectral-theoretic sense. The paper is clearly written and the proof structure is coherent.","major_comments":[{"comment":"Lemma 4.4 is load-bearing: Lemma 4.2 and hence Proposition 4.2 use (14) for every E∈R for a fixed ω. This is stronger than the quasi-everywhere equality delivered by the upper envelope theorem (Theorem 2.2), and the citation [2,8] is not pinpointed. Since the introduction stresses the failure of full limits and the possibility of positive-capacity exceptional sets, the authors should either prove (14) for exterior powers of T_± or state it in the quasi-everywhere form and take the union of capacity-zero exceptional sets in Proposition 4.2. The repair is transparent, but as written the central estimate is not fully justified.","section":"§4.3, Lemma 4.4"}],"minor_comments":[{"comment":"Typo: 'easily check' should be 'easily checked'. Also Lemma 3.2 contains 'equivalance' for 'equivalence'.","section":"§3, Lemma 3.1"},{"comment":"The symbol J(ω) for the symplectic form clashes with the flipped operator J_ω; consider renaming the form, e.g. Ω(ω), to avoid confusion.","section":"§4.3, proof of Lemma 4.2"},{"comment":"The notation '/∈' should be '∉' throughout.","section":"§4.2, Lemma 4.1"},{"comment":"Reference [2] has a typo: 'Lyaponov' should be 'Lyapunov'.","section":"References"},{"comment":"The identity det(E - J_{ω,[0,n-1]}) = (∏_{j=0}^{n-1} det A_j(ω)) det D_n(E,ω) is standard, but the sign/index conventions are worth a one-line derivation or an exact reference, since the absolute value is eventually used.","section":"§4.2, determinant identity"},{"comment":"The paper proves the result for a particular maximal spectral measure μ_ω (the trace of the two-site spectral measure). The implication for 'any maximal spectral measure', as stated in Theorem 1.1, follows from mutual absolute continuity of maximal spectral measures; this should be explicitly noted.","section":"Theorem 1.2 / Section 1"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely correct and the flipping construction is genuinely novel. The only substantive issue is the precise status of Lemma 4.4; if the authors provide a direct proof or a clear reference for the strong 'every E' form for exterior powers of both transfer matrices, the paper should be acceptable. The current version leaves a load-bearing point under-specified, hence major_revision rather than minor_revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Zhenfu Wang, Bei Zhang, and Qi Zhou prove the whole-line ultimate Ishii–Pastur theorem for ergodic block Jacobi operators, settling Damanik–Fillman's Problem 4.7.18 in the scalar case and going well beyond it. The flipping construction is the right move: it maps the whole-line operator to a half-line block Jacobi operator of block size 2m in a way that preserves finite-volume spectra exactly, so the DOS of the flipped non-ergodic family is still governed by the original ergodic system. The determinant asymptotics and the exterior-power bound then yield the needed lower bound on the smallest singular value of the Dirichlet fundamental matrix, and the Schnol theorem does the rest. The proof is coherent, forward-derived, and honestly discusses the whole-line difficulty. That is real progress.\n\nThe soft spots are mostly presentation. Lemma 4.4 states a Craig–Simon type limsup upper bound for every E for almost every omega. At first glance the 'every E' looks suspicious, because the upper envelope theorem only gives equality quasi-everywhere. But the inequality alone is standard: Craig–Simon prove the limsup bound for all E, and Goldsheid–Sodin carry it over to the matrix case. The stress-test that says this is only true off a capacity-zero set is confusing the equality statement with the one-sided bound. Still, the paper should not just cite [2,8] with no page or theorem number; it should either quote the exact statement or give a short proof, especially since the backward cocycle T_- is handled via its relation to T_+. Similarly, the Schnol theorem is quoted loosely. These are fixable.\n\nOne more minor point: the paper says the flipped family is no longer ergodic, which is true, but Proposition 4.1 compensates exactly by the finite-volume unitary equivalence. That part is solid. I did not find any circular step; the Thouless formula is used in the standard direction.\n\nSo: the main theorem is new, the proof is believable, and the open problem is genuinely resolved. This deserves a proper peer review. I would recommend sending it to a good spectral theory journal after the authors tighten the references and make Lemma 4.4 self-contained.","headline":"A convincing resolution of the whole-line Ishii–Pastur problem; the proof is strong, though one quoted bound needs a precise reference.","tokens_in":12975,"tokens_out":12999,"would_cite":true,"duration_ms":119841,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["47B36","47A10","81Q10"],"pacs":[],"model":"deepseek-v4-flash","headline":"For almost every realization of a whole-line ergodic block Jacobi operator, the spectral measure's support in the positive-Lyapunov region has zero logarithmic capacity.","keywords":["ergodic block Jacobi operators","Lyapunov exponents","logarithmic capacity","spectral measure","Ishii-Pastur theorem","whole-line operators","zero-capacity support","Thouless formula"],"falsifier":"Take the almost-Mathieu operator with coupling $\\lambda>2$ and compute, for typical $\\omega$, the logarithmic capacity of the set of energies in $S^+$ where $\\limsup (1/n) \\log s_{2m}(D_n(E,\\omega)) < L_m(E)$. If this exceptional set has positive capacity and carries positive spectral measure, the theorem is false; if it is zero-capacity, the paper's key estimate is confirmed for that model.","tokens_in":11899,"feed_emoji":"🧮","tokens_out":15296,"duration_ms":141263,"temperature":0.7,"texified_at":"2026-08-05T21:48:09.851579+00:00","pith_summary":"The paper establishes the ultimate Ishii-Pastur theorem for whole-line ergodic block Jacobi operators: for almost every realization, any maximal spectral measure is supported, within the positive-Lyapunov region, on a Borel set of zero logarithmic capacity. This replaces the classical statement that the spectral measure is singular in that region with a strictly stronger smallness statement. It covers scalar Schrödinger operators as the $m=1$ case, thereby resolving the open whole-line scalar problem. The mechanism is a unitary 'flip' of the negative half-line onto the nonnegative half-line, which encodes the two-ended problem through a single Dirichlet fundamental matrix; its smallest singular value is then controlled by determinant and exterior-power asymptotics.","texify_model":"deepseek-v4-flash","texify_usage":{"total_tokens":8098,"prompt_tokens":746,"completion_tokens":7352,"prompt_tokens_details":{"cached_tokens":0},"prompt_cache_hit_tokens":0,"prompt_cache_miss_tokens":746,"completion_tokens_details":{"reasoning_tokens":6654}},"feed_headline":"Positive-Lyapunov spectral support has zero capacity","feed_subtitle":"Proves the ultimate Ishii-Pastur theorem for whole-line block Jacobi operators, resolving an open scalar problem.","key_machinery":"The argument rests on a unitary flipping map $U$ that identifies the whole-line operator with a half-line block Jacobi operator $J_\\omega$ of twice the block size, preserving the coupling across the origin. The two-ended evolution is captured by a single Dirichlet fundamental matrix $D_n(E,\\omega)$, whose smallest singular value $s_{2m}$ controls all initial data. The key estimate—$\\limsup (1/n) \\log s_{2m}(D_n) \\geq L_m(E)$ off a capacity-zero set—follows from the identity $s_{2m}(D) = |\\det D| / \\|\\wedge^{2m-1}D\\|$, with the determinant controlled by the Thouless formula and upper envelope theorem, and the exterior-power factor bounded by Craig-Simon-type estimates on forward and backward transfer matrices. Shnol's theorem th","core_discovery":"For $P$-almost every $\\omega$, there is a Borel set $Q_\\omega$ of zero logarithmic capacity such that $\\mu_\\omega(S^+ \\setminus Q_\\omega)=0$, where $S^+=\\{E: L_m(E)>0\\}$ and $\\mu_\\omega$ is a maximal spectral measure of the whole-line ergodic block Jacobi operator $H_\\omega$. In other words, the spectral measure of the positive-Lyapunov region is carried by a set invisible to logarithmic capacity. Under unique ergodicity and continuity of coefficients, the conclusion holds for each realization individually. In the scalar case $m=1$, this settles the whole-line problem.","pith_inferences":["The flipping map may be reusable beyond spectral measures: any two-ended linear problem whose growth in both directions matters could be folded into a single half-line problem at the cost of doubling dimension and losing ergodicity of the coefficients.","The proof suggests that, in this class, the set of energies where the transfer-matrix cocycle is non-regular has zero capacity; verifying this directly would connect the result to recent studies of non-Lyapunov exceptional sets.","If the invertibility assumption on the hopping matrices could be relaxed by an approximation argument, the conclusion would extend to a broader class of ergodic block operators.","The methods may transfer to other fine notions of spectral smallness (e.g., Hausdorff measures with dimension functions) by replacing the potential-theoretic estimates with appropriate upper envelope theorems."],"forward_implications":["Resolves the open scalar whole-line problem: the restriction of a maximal spectral measure to the positive-Lyapunov region is carried by a zero-capacity set.","Extends the zero-capacity conclusion from half-line to whole-line operators, and from scalar to matrix-valued block Jacobi and strip operators.","Under unique ergodicity and continuity, provides a deterministic zero-capacity support for every realization rather than almost every.","Since zero-capacity sets have zero Hausdorff dimension, the positive-Lyapunov spectral support is also of zero Hausdorff dimension in this general ergodic setting.","The flipping construction offers a template for reducing whole-line spectral questions to half-line ones while keeping the origin coupling intact."],"fun_headline_variants":["Zero-capacity set carries spectral measure for positive Lyapunov","Whole-line Jacobi: spectral support has zero capacity","Positive Lyapunov spectrum lies in zero-capacity sets","Spectral measure vanishes outside zero-capacity set"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof relies on the bound that, for almost every realization and every energy, the normalized logarithm of any exterior power of the transfer matrix has $\\limsup$ no larger than the corresponding sum of Lyapunov exponents; if the set of energies where this fails has positive logarithmic capacity, the exterior-power estimate and the main theorem collapse.","fun_headline_variants_meta":{"raw":{"variants":["Zero-capacity set carries spectral measure for positive Lyapunov","Whole-line Jacobi: spectral support has zero capacity","Positive Lyapunov spectrum lies in zero-capacity sets","Spectral measure vanishes outside zero-capacity set"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000116,"raw_usage":{"total_tokens":835,"prompt_tokens":591,"completion_tokens":244,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":335,"completion_tokens_details":{"reasoning_tokens":179}},"tokens_in":335,"tokens_out":244,"duration_ms":3123,"temperature":1.0,"reasoning_tokens":179,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T01:23:18.210156+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the almost-Mathieu operator with coupling $\\lambda>2$ and compute, for typical $\\omega$, the logarithmic capacity of the set of energies in $S^+$ where $\\limsup (1/n) \\log s_{2m}(D_n(E,\\omega)) < L_m(E)$. If this exceptional set has positive capacity and carries positive spectral measure, the theorem is false; if it is zero-capacity, the paper's key estimate is confirmed for that model.","supporting_citations":[],"review_version":1}