Research paperWorking paperARR-2026-7KV9Y9ASX48H7A7N · v5 · 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 Not peer reviewed.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 positive Boolean weights with uniform labelled marginals up to linear order while exponentially many normalized transfer eigenvalues approach one; the Perron stationary law has the same marginals. Separate product-tilt constructions and hypercontractive bounds identify the density-moment boundary for a dimension-independent normalized transfer gap. At s=(1+rho)/(1-rho), the exact norm-budget boundary is 2^(1/s). We prove that escaped critical moment Theta<1 suffices for a uniform gap, with an explicit finite-cutoff bound. One collapsing family simultaneously satisfies every K-tail moment cap one for K>=1 and total critical moment cap two, with any prescribed positive density floor below one. For each fixed 0<=K<1, the same tail cap guarantees a positive gap. Stable one-vertex spikes realize every finite positive Theta, so critical-power uniform integrability is sufficient but not necessary. The code, tilt and spike families are distinct. Classical hypercontractivity, sign compression, product tilts and rank-one perturbations supply the mechanisms. Positive constants are not claimed optimal; external priority is unestablished. No fixed-range Gibbs, local-dynamics, physical-gap or continuum conclusion is claimed.

Preview PDF on this page v5

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

Reference this work

Cite this paper

v5 · 2026-11-13

Lluis Eriksson (2026). Product Marginals and Density-Moment Thresholds for Ising Transfer Kernels. AIRR.SCIENCE, ARR-2026-7KV9Y9ASX48H7A7N, v5. https://airr.science/papers/ARR-2026-7KV9Y9ASX48H7A7N/versions/v5/ Research preprint; ARR-2026-7KV9Y9ASX48H7A7N v5. Not peer reviewed by AIRR.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 · viewing
  • v4 · major revision · source snapshot available
  • v3 · major revision · source snapshot available
  • v2 · major revision · source snapshot available
  • v1 · source snapshot available

Revision statement

Major revision. Groups critical-tail sufficiency below escaped moment one, sharp collapse at the endpoint, stable examples at every finite escaped moment, and one collapsing family satisfying all fixed-cutoff tail caps for K>=1. For K<1 the same cap instead forces a positive gap. Layout-only308 preserves all307 proofs. No optimal positive constants, external-priority or physical interpretation claim.

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: 4.00/10 on the AIRR Millennium scale, overall 4/10; correctness_confidence 4/5, rigor 4/5, novelty 3/5, significance 3/5, reproducibility 4/5; recommendation minor_revision. Full report, all 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 or human peer review, formal verification 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
    4.00 / 10.00Strong
    Distinct models
    1
    Range
    4.00–4.00
    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: 4.00. Policy: AIRR-RATING-1.0. Method and limits.

    ModelOriginal scoreWeightBasis
    OpenAI · unknown4.001.000No evidenced exact benchmark configuration

    Download the calculation snapshot

    OpenAI · unknown · unknown · review context: unknown4.00

    Recommendation: Minor Revision · Material objections: 0

    First and only numerical assessment of the exact 22-page manuscript308A.pdf, raw SHA-256 450b6afa4e49a34f5fa36958b36b302ad1e266832aa271f857522aecfd3c6444. It provides a technically serious synthesis of exact high-order marginal matching with many slow transfer modes, strict density-moment regimes, and exact critical norm/tail boundaries. The strongest endpoint proofs withstand this manual analytical review. Millennium 4.00 (Strong) is a contribution judgment, not a criterion average or correctness probability. Scientific minor_revision addresses positioning and navigation; this is a Working-paper assessment, not editorial acceptance. Draft-linked native identifiers are not a submission receipt. Requested gpt-6-astra/high is unverified; model_id and independence remain unknown.

    Correctness confidence
    The code/Perron-marginal construction, sign-region estimates, balanced critical tilts, dilution, and cutoff-one endpoint withstand this analytical reading. No concrete counterexample or invalidating proof gap was identified. One AI review is not formal or human verification; served backend identity is unverified.
    Rigor
    The proofs explicitly cover zero densities, active support, eigenvalue multiplicities, Gram-matrix zeros, ratio normalization, finite prefixes, and strict endpoint tail events. Constants and critical prefix choices remain implicit, and no optimal positive gap constant is claimed.
    Novelty
    The exact multiplicative-cube budget and simultaneous cutoff-one endpoint, and the joint weight/Perron marginal result, remain plausible novelty candidates relative to the targeted primary sources inspected. Hypercontractive exponents, coding inputs, sign splitting, and norm-budget precedents are classical. External priority is unestablished; Aida's preprint could not be read after access failures.
    Significance
    If correct and original, the results give useful sharp worst-case classifications and information obstructions for weighted Boolean noise/transfer operators. No demonstrated consequence for fixed-range Gibbs models, local dynamics, physical mass gaps, or a broader recognized open problem supports a higher impact rating.
    Reproducibility
    All 33 packet files match byte counts and hashes; all 22 supplied page images were inspected, with text/TeX support. The finite-dimensional analytical proofs are inspectable. Code selection and critical size thresholds are existential. Mathematical tests, formal checks, independent rebuild/rerender, public source availability, and historical audit claims were not independently verified.

    Strengths

    • Explicit finite positive-weight construction preserving labelled product marginals for both the input and Perron stationary laws, with exponentially many slow modes.
    • Exact parity transfer spectrum prevents conflating transfer relaxation with the predecessor's heat-bath relaxation.
    • Strict moment regimes are established for the same operator, and the critical boundary includes equality with any fixed density floor below one.
    • The cutoff-one family satisfies the total moment and strict-tail caps in every dimension and all larger cutoffs simultaneously; stable spikes disprove necessity of uniform integrability.
    • Careful support, multiplicity, normalization, finite-prefix, and scope bookkeeping; no unsupported physical or local-dynamics conclusion.

    Weaknesses

    • O1: Priority and external originality remain unresolved. Inspected primary sources already contain major mechanisms and related budget transitions; a precise theorem-level comparison would strengthen contribution positioning.
    • O2: Demonstrated significance is restricted to globally weighted Boolean transfer kernels. Marginal and moment conclusions use different families, with no fixed-range, local-dynamics, or physical-gap application.
    • O3: Exact endpoints are qualitative worst-case boundaries. No optimal positive constant, endpoint rate, or practical critical size threshold under a prescribed floor is supplied; code selection and critical prefix choices are existential.
    • O4: The long abstract and successive auxiliary critical estimates make the final classification expensive to navigate. A concise conclusion/dependency guide and streamlined auxiliary presentation would help.
    • O5: Aida's primary preprint was inaccessible; other source reads were targeted, not exhaustive. External novelty clearance is not established. The Miclo bibliography entry would benefit from normal date/version/publication metadata.
    • O6: This assessment verifies packet integrity and readability but does not replay builds, formal mathematics, or public evidence availability. Shared provider/filesystem and unverified backend/effort limit independence claims.

    Potential errors

    None reported.

    Strong novelty candidates

    • The combined linear-order exact marginal agreement for both weighting and Perron stationary laws together with exponentially many slow normalized transfer modes; candidate synthesis, not a new coding mechanism.
    • The exact critical raw-density norm budget 2^(1/s), including asymptotic collapse at equality with any prescribed floor b<1, for the fixed Boolean congruence; historical priority unestablished.
    • The single collapsing family satisfying E[r^s]<=2 and E[r^s 1_(r>1)]<=1 in every dimension, hence the same cap for every larger cutoff simultaneously; a refinement of the balanced-tilt/sign-truncation method.

    Unresolved material objections

    None reported.

    arr:assessment:36dfb605-3415-43a1-9b9c-c16113f65e1c · prompt ARR-ASSESS-1.0 · response SHA-256 794f116b8aad6f8f92b69b5cec55cdd7c12cc6bb1684c5fc18677794d129c70a · canonical PDF SHA-256 450b6afa4e49a34f5fa36958b36b302ad1e266832aa271f857522aecfd3c6444

    OpenAI · Codex/OpenAI; exact backend not attested

    Original AI-only score: 4 / 10 · First exact-PDF assessment 4.00/10; 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 numerical assessment of the exact 22-page manuscript308A.pdf, raw SHA-256 450b6afa4e49a34f5fa36958b36b302ad1e266832aa271f857522aecfd3c6444. It provides a technically serious synthesis of exact high-order marginal matching with many slow transfer modes, strict density-moment regimes, and exact critical norm/tail boundaries. The strongest endpoint proofs withstand this manual analytical review. Millennium 4.00 (Strong) is a contribution judgment, not a criterion average or correctness probability. Scientific minor_revision addresses positioning and navigation; this is a Working-paper assessment, not editorial acceptance. Draft-linked native identifiers are not a submission receipt. Requested gpt-6-astra/high is unverified; model_id and independence remain unknown.

    Backend identity: unknown. OpenAI/Codex assisted preparation. First and only numerical assessment of this exact PDF: 4.00/10 on the AIRR Millennium scale, overall 4/10; correctness_confidence 4/5, rigor 4/5, novelty 3/5, significance 3/5, reproducibility 4/5; recommendation minor_revision. Full report, all 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 or human peer review, formal verification or editorial acceptance.

    Report captured · Exact PDF SHA-256 450b6afa4e49a34f5fa36958b36b302ad1e266832aa271f857522aecfd3c6444

    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 Working-paper permission. No fresh human assent, human signoff or accepted-collection admission.

    Record timestamps & download statistics
    Deposit recordedFirst repository commit · full Git SHA recorded
    PublishedGitHub release · tag ARR-2026-7KV9Y9ASX48H7A7N-v5
    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 →