Research paperWorking paperARR-2026-7KV9Y9ASX48H7A7N · v4 · 2026-11-13

Product Marginals and Density-Moment Thresholds for Ising Transfer Kernels

Lluis Eriksson

Citable version that has not completed AIRR screening. Not peer reviewed. Screening record

Abstract

Fix a positive Ising coupling. We construct strictly positive weights on Boolean cubes whose labelled marginals on every set of at most a fixed positive fraction of coordinates equal uniform product marginals, while exponentially many normalized transfer eigenvalues approach one exponentially fast. The Perron stationary law has the same marginals. Binary codes, a positive product admixture and a finite block estimate supply the construction. A full parity-transfer spectrum separates this mechanism from heat-bath slowdown. Conversely, for Boolean noise parameter 0<rho<1, density moments of order s>(1+rho)/(1-rho) imply an explicit dimension-independent transfer gap, without pointwise upper or positive lower bounds and allowing zero weights. For every 1<s<(1+rho)/(1-rho) and B>1, antipodal product tilts and a Gram comparison yield normalized densities with Ls norm at most B and density at least 1/B whose gaps collapse exponentially. At the critical exponent s=(1+rho)/(1-rho), the exact norm-budget boundary is B*=2^(1/s): budgets below B* give a positive dimension-independent gap, while at B* and above gaps can collapse even with any prescribed fixed density floor below one. The proof combines a two-sign-region moment bound, a balanced product tilt with limiting critical moment two, and exact-budget dilution. An explicit stretched-exponential rate is retained for B>2; no rate at equality or optimal positive-gap formula is claimed. The threshold and ingredients are classical; external priority of these joint spectral conclusions is unestablished. Global many-body weights are allowed; no fixed-range Gibbs or local-dynamics claim follows.

Preview PDF on this page v4

Uses your browser's PDF viewer. Open the PDF directly if a preview is unavailable.

Reference this work

Cite this paper

v4 · 2026-11-13

Lluis Eriksson (2026). Product Marginals and Density-Moment Thresholds for Ising Transfer Kernels. AIRR.SCIENCE, ARR-2026-7KV9Y9ASX48H7A7N, v4. https://airr.science/papers/ARR-2026-7KV9Y9ASX48H7A7N/versions/v4/ Research preprint; ARR-2026-7KV9Y9ASX48H7A7N v4. Not peer reviewed by AIRR. Working paper; not admitted to the AIRR accepted collection.

Use this version link to cite the manuscript you read. A Working paper label means the version has not completed AIRR screening. AIRR screening is not journal peer review. Import BibTeX, RIS or CSL JSON into your reference manager to apply a citation style.

Version history

The paper identifier remains stable. Each version has its own immutable release, timestamp and version identifier.

  • v9 · major revision · source snapshot available
  • v8 · major revision · source snapshot available
  • v7 · major revision · source snapshot available
  • v6 · major revision · source snapshot available
  • v5 · major revision · source snapshot available
  • v4 · major revision · source snapshot available · viewing
  • v3 · major revision · source snapshot available
  • v2 · major revision · source snapshot available
  • v1 · source snapshot available

Revision statement

Major revision. Determines the exact critical norm-budget boundary B*=2^(1/s), including equality and any prescribed density floor below one. The sufficient positive-gap formula is not claimed optimal and no boundary convergence rate is asserted. Same grouped paper; first exact-PDF review and all histories retained.

Verification record

Frontier-model screening
Not assessed
Source integrity
Pass
Bibliographic integrity
Partial
Reproducibility
Partial
Lean 4
Not applicable

Recorded under ARR-VERIFY-1.0. AIRR verification and screening are not peer review.

AI assistance statement

OpenAI/Codex assisted preparation. First and only numerical assessment of this exact PDF: 4.20/10, overall 4/10; correctness_confidence 4/5, rigor 4/5, novelty 3/5, significance 3/5, reproducibility 4/5; recommendation minor_revision. Full report, objections and actual context retained. Requested/configured gpt-6-astra/high; served backend/version/effort and technical workspace/memory isolation unverified. Same provider family, independence unknown; no inherited grade, resampling, independent provider, human peer review, formal proof or editorial acceptance.

Frontier-model screening

Status: not_assessed. Any listed reports correspond to this exact version under ARR-SCREEN-1.0; no absent assessment is represented as a pass.

    Longitudinal frontier-model record

    Model assessments and review context

    Read the scale and limits
    4.20 / 10.00Strong
    Distinct models
    1
    Range
    4.20–4.20
    Method
    bounded capability-weighted mean · exact version only

    Limited evidence · 0/1 benchmark matches. This is an evidence description, not a probability of correctness.

    Only one distinct model; One provider; shared blind spots possible; Fresh blinded review context not documented for every model; Benchmark match incomplete; Adverse reports or unresolved objections require editorial disposition

    Score calculation and evidence

    Unweighted mean: 4.20. Policy: AIRR-RATING-1.0. Method and limits.

    ModelOriginal scoreWeightBasis
    OpenAI/Codex · unknown4.201.000No evidenced exact benchmark configuration

    Download the calculation snapshot

    OpenAI/Codex · unknown · unknown · review context: unknown4.20

    Recommendation: Minor Revision · Material objections: 0

    First and only assessment of manuscript299A.pdf, SHA256 51d6ceeca54f3d782ed552074c0a320e715cd345502f25ce5387c2b380f732eb. The manuscript provides convincing finite-dimensional proofs of a code-based marginal obstruction and strict/critical density-moment classifications for weighted Boolean-noise transfer operators. The exact critical-budget result, including equality and any fixed density floor below one, is the strongest contribution. No unresolved mathematical objection invalidating a main theorem was identified. Minor revision is recommended for literature positioning and presentation; originality and broader significance remain uncertain. The 4.20 Millennium score is not a criterion average, correctness probability, editorial acceptance or inherited grade. Requested model/effort was gpt-6-astra/high; actual served configuration and technical independence were not verified.

    Correctness confidence
    All 17 pages were read and visually inspected through the supplied renders. The code, sign-part, Gram-matrix, and exact-budget dilution arguments survived analytical review; no material mathematical error was identified. Single-model review only, with no human or formal verification.
    Rigor
    Theorems 1 and 7–13 explicitly handle normalization, zero densities, active support, multiplicities, rectangular Gram zeros, and all-dimension endpoint admissibility. Small-tilt constants and onset dimensions are existential, consistently with the stated claims.
    Novelty
    The joint marginal/stationary/many-mode conclusion and exact fixed-cube density-budget classification are plausible novelty candidates relative to the checked sources. Coding, hypercontractivity and sign-region machinery are classical; Miclo Section 6 already contains a related norm-budget transition at 2. External priority remains unestablished. Aida Theorem 10 was not checked because later retrievals timed out.
    Significance
    A useful specialist distinction between marginal information and transfer relaxation, with a sharp moment-budget classification for this congruence family. No fixed-range Gibbs, local-dynamics, continuum or physical mass-gap consequence is established; broad influence is not demonstrated.
    Reproducibility
    Main proofs are inspectable in the exact PDF, supported by exact TeX, text and 17 page images. All 28 manifest lengths and hashes matched. Build reproduction, runtime mathematics, Lean, public source-tree verification and human review were not performed; supplied build provenance is not proof verification.

    Strengths

    • Theorem 1 matches labelled product marginals for both the weighting law and the Perron stationary law while controlling exponentially many nonleading modes at fixed coupling.
    • Proposition 5 computes the actual parity transfer spectrum, preventing an invalid inference from slow single-site heat-bath dynamics.
    • Theorems 7 and 9 use sign-part compression and forward/reverse hypercontractivity without requiring a positive density floor; zero densities are treated explicitly.
    • Theorem 10 uses two antipodal product channels and a Gram comparison to control two leading eigenvalues, rather than inferring gap collapse from top-norm growth alone.
    • Theorem 13 proves the restricted moment/sign-region inequality and then attains the exact critical budget by vanishing dilution with an endpoint multiplier perturbation bound.
    • The manuscript explicitly limits its scope, credits classical ingredients, identifies the exact source, and discloses model involvement and author ownership of AIRR.

    Weaknesses

    • O1 — Minor literature revision: pp.15–16 should explicitly compare the critical density-budget transition with the L2-to-L4 norm-budget threshold below/at 2 discussed in Miclo Section 6 and Proposition 11. These are different operator/budget classes; no direct duplication or implication has been demonstrated.
    • O2 — External priority remains unresolved. Targeted primary-source checks and narrow searches cannot establish originality of either joint conclusion; the assessment must not be presented as novelty clearance.
    • O3 — The unrestricted global weights and full-configuration Doob dynamics limit significance. No natural fixed-range model application, local-relaxation theorem, continuum result or physical mass-gap advance follows.
    • O4 — Small-tilt choices, remainder constants, and endpoint floor cutoffs are existential. A parameter-selection recipe would improve reuse. Optimal gap bounds, endpoint convergence rates and optimal code tradeoffs are not established and are not claimed.
    • O5 — The marginal-matching code construction and the moment-budget tilt constructions are separate. Their properties must not be combined into a nonexistent single-family theorem; a compact result map would help.
    • O6 — A clearer introductory roadmap should foreground the endpoint classification. The historical source-access sentence on p.16 is procedural history and can be removed from the scientific exposition while retaining archived records.
    • O7 — Assurance limits: same-provider single-model assessment, shared filesystem, unknown technical independence and served configuration, no human/formal verification or build replay. Aida Theorem 10 was not retrieved sufficiently to check it. A search incidentally exposed an ungraded v2 abstract snippet, which was not opened or used; no old assessment or score was consulted.

    Potential errors

    None reported.

    Strong novelty candidates

    • Theorem 1's simultaneous exact linear-order labelled marginals for mu and pi with exponentially many near-Perron transfer modes at a fixed positive coupling; candidate only, external priority unestablished.
    • Theorem 13's exact dimension-free density-moment budget boundary B*=2^(1/s) for the fixed-rho multiplicative Boolean-cube family, including the equality case and any fixed floor below one; candidate only, closely related mechanisms and norm-budget precedents are classical.
    • Theorem 10's two-channel product-tilt realization of exponential gap collapse across the entire strict subcritical moment regime with a fixed positive floor; candidate only, no exhaustive literature clearance.

    Unresolved material objections

    None reported.

    arr:assessment:82a0b064-7809-4964-abb3-458af0a14c36 · prompt ARR-ASSESS-1.0 · response SHA-256 7164ef56e660700c01d74fc955b7ca56c3e330798c6ea04b5c26d0cf58c2165b · canonical PDF SHA-256 51d6ceeca54f3d782ed552074c0a320e715cd345502f25ce5387c2b380f732eb

    OpenAI · Codex/OpenAI; exact backend not attested

    Original AI-only score: 4.2 / 10 · First exact-PDF assessment 4.20/10; overall 4/10; correctness_confidence 4/5, rigor 4/5, novelty 3/5, significance 3/5, reproducibility 4/5. Same review as registry, not another assessment.

    minor_revision

    First and only assessment of manuscript299A.pdf, SHA256 51d6ceeca54f3d782ed552074c0a320e715cd345502f25ce5387c2b380f732eb. The manuscript provides convincing finite-dimensional proofs of a code-based marginal obstruction and strict/critical density-moment classifications for weighted Boolean-noise transfer operators. The exact critical-budget result, including equality and any fixed density floor below one, is the strongest contribution. No unresolved mathematical objection invalidating a main theorem was identified. Minor revision is recommended for literature positioning and presentation; originality and broader significance remain uncertain. The 4.20 Millennium score is not a criterion average, correctness probability, editorial acceptance or inherited grade. Requested model/effort was gpt-6-astra/high; actual served configuration and technical independence were not verified.

    Backend identity: unknown. OpenAI/Codex assisted preparation. First and only numerical assessment of this exact PDF: 4.20/10, overall 4/10; correctness_confidence 4/5, rigor 4/5, novelty 3/5, significance 3/5, reproducibility 4/5; recommendation minor_revision. Full report, objections and actual context retained. Requested/configured gpt-6-astra/high; served backend/version/effort and technical workspace/memory isolation unverified. Same provider family, independence unknown; no inherited grade, resampling, independent provider, human peer review, formal proof or editorial acceptance.

    Report captured · Exact PDF SHA-256 51d6ceeca54f3d782ed552074c0a320e715cd345502f25ce5387c2b380f732eb

    A model assessment is not peer review or a correctness certificate. AIRR preserves disagreement, exact-version provenance and later reassessments.

    Editorial disclosure

    Existing continuous sole-author delegation and exact native linked-revision Working-paper permission. No fresh human assent, human signoff, independent review or accepted-collection admission.

    Record timestamps & download statistics
    Deposit recordedFirst repository commit · full Git SHA recorded
    PublishedGitHub release · tag ARR-2026-7KV9Y9ASX48H7A7N-v4
    Canonical PDF downloads19GitHub release downloads only; direct AIRR PDF reads are not included
    Page viewsNot availablePage-view counts are not available in this public snapshot. Optional operator statistics are separate.
    Definitions and rankings →