# Evidence Press

> Press releases for new research, published with the evidence attached. Every release describes an unrefereed research result with an open, replayable evidence package (code, data, exact certificates, pinned environments, SHA-256 manifests) archived with a DOI. Nothing on this site is peer reviewed; each page states exactly what has and has not been verified, and lists open follow-up problems in machine-readable form.

Key endpoints: full JSON index at /api/papers.json (JSON Schema at /api/schema.json); per-release JSON at /releases/<slug>/paper.json; per-release Markdown at /releases/<slug>/index.md; per-release BibTeX at /releases/<slug>/cite.bib; RSS at /feed.xml; JSON Feed at /feed.json. Direct paper PDFs are in each release's metadata (pdfUrl).

## Releases

- [Random reshuffling as a baseline policy](https://evidencepress.org/releases/random-reshuffling-baseline-policy/): In an exact quadratic model of learning, some fair choices inside ordinary data shuffling admit a same-baseline value certificate; popular ordering proxies can fail without bound, while global threshold ordering is NP-complete when the stable step is instance-encoded and approaches 2. (PDF: https://github.com/ipitchford/random-reshuffling-baseline-policy/releases/download/v1.0.0-candidate/random-reshuffling-baseline-policy-v1.0.0-candidate.pdf; DOI: https://doi.org/10.5281/zenodo.21757895; code: https://github.com/ipitchford/random-reshuffling-baseline-policy; status: unrefereed)
- [Stocks are not flows](https://evidencepress.org/releases/stocks-are-not-flows/): England's tenure shift is real; the aggregate housing data do not reveal how much, if any, was caused by rising resources at the top. (PDF: https://github.com/ipitchford/stocks-are-not-flows/releases/download/v1.0.0-preprint/stocks-are-not-flows-v1.0.0.pdf; DOI: https://doi.org/10.5281/zenodo.21740339; code: https://github.com/ipitchford/stocks-are-not-flows; status: unrefereed)
- [Irreducible pushforwards and quartic transitions](https://evidencepress.org/releases/irreducible-pushforwards-quartic-transitions/): A stalled attack yields two standalone tools: one turns irreducibility into geometric separation, and the other detects exact rank collapse in a quartic transition; Furter's R(3) remains open. (PDF: https://github.com/ipitchford/irreducible-pushforwards-quartic-transitions/releases/download/v1.0.0-preprint/irreducible-pushforwards-quartic-transitions-v1.0.0.pdf; DOI: https://doi.org/10.5281/zenodo.21745937; code: https://github.com/ipitchford/irreducible-pushforwards-quartic-transitions; status: unrefereed)
- [Exact low-length Recht–Ré inequalities](https://evidencepress.org/releases/exact-low-length-recht-re-inequalities/): The first exact account of where a leading mathematical justification for shuffling data works, where it fails, and where it recovers — with a surprise: the usual theoretical measure can call reshuffling worse even when it reduces error. (PDF: https://github.com/ipitchford/exact-low-length-recht-re-inequalities/releases/download/v1.0.0-candidate/exact-low-length-recht-re-inequalities-v1.0.0-candidate.pdf; DOI: https://doi.org/10.5281/zenodo.21709239; code: https://github.com/ipitchford/exact-low-length-recht-re-inequalities; status: unrefereed)
- [z(20) = 6 (Erdős problem 758)](https://evidencepress.org/releases/z20-equals-6/): Every graph on 20 vertices can be split into six parts, each a clique or an independent set — and 6 is best possible. A computer-assisted proof, released with its full certificate chain, settles the first open case of an Erdős–Gimbel problem. (PDF: https://github.com/ipitchford/z20-cochromatic/releases/download/candidate-2026-07-26/z20_equals_6_proof.pdf; DOI: https://doi.org/10.5281/zenodo.21647645; code: https://github.com/ipitchford/z20-cochromatic; status: unrefereed)
- [VR2(K4) = 20](https://evidencepress.org/releases/vr2-k4-equals-20/): Colour every edge among 20 points red or blue and you cannot avoid two completely separate single-colour foursomes — while a carefully built 19-point colouring can. A companion result to z(20) = 6. (PDF: https://github.com/ipitchford/z20-cochromatic/releases/download/vr2-k4-v0.1-candidate/vr2_k4_equals_20_candidate.pdf; DOI: https://doi.org/10.5281/zenodo.21647654; code: https://github.com/ipitchford/z20-cochromatic; status: unrefereed)
- [Reducible incidence divisors](https://evidencepress.org/releases/reducible-incidence-divisors/): A classification of exactly when a natural family of divisors breaks into pieces — and a screening theorem narrowing where Jacobian-conjecture-style behaviour could possibly live. (PDF: https://github.com/ipitchford/reducible-incidence-divisors-affine-slices/releases/download/v1.0.1/paper.pdf; DOI: https://doi.org/10.5281/zenodo.21647616; code: https://github.com/ipitchford/reducible-incidence-divisors-affine-slices; status: unrefereed)
- [Exotic affine three-spheres](https://evidencepress.org/releases/exotic-affine-three-spheres/): A natural slice of polynomial factorisation space is claimed to be an exotic affine three-sphere — a shape that mimics a familiar one without being it — while every quadratic–cubic slice is obstructed from being ordinary affine space. (PDF: https://github.com/ipitchford/exotic-affine-spheres-quadratic-cubic/releases/download/v0.1.1/paper.pdf; DOI: https://doi.org/10.5281/zenodo.21647607; code: https://github.com/ipitchford/exotic-affine-spheres-quadratic-cubic; status: unrefereed)
- [Erdős 848: exact answer for all N](https://evidencepress.org/releases/erdos-848-all-n/): A certificate-backed determination that the answer to an Erdős–Sárközy extremal problem is exactly ⌊(N+18)/25⌋ for every N — closing the gap between known asymptotic results and small cases. (PDF: https://github.com/ipitchford/erdos-848-all-n/releases/download/v0.1-candidate/paper.pdf; DOI: https://doi.org/10.5281/zenodo.21647629; code: https://github.com/ipitchford/erdos-848-all-n; status: unrefereed)
- [Degree-difference principle and affine slices](https://evidencepress.org/releases/degree-difference-affine-slices/): A clean determinant identity for the map that multiplies two polynomials and records their resultant — and a study of the geometric slices where Jacobian-conjecture-style questions live. (PDF: https://github.com/ipitchford/degree-difference-affine-slices/releases/download/v0.1-candidate/paper.pdf; DOI: https://doi.org/10.5281/zenodo.21647593; code: https://github.com/ipitchford/degree-difference-affine-slices; status: unrefereed)

## Machine-readable

- [papers.json](https://evidencepress.org/api/papers.json): full structured index — titles, DOIs, PDF links, verification status, provenance, key results, keywords, media, and open follow-up problems for every release
- [schema.json](https://evidencepress.org/api/schema.json): JSON Schema for the index
- [llms-full.txt](https://evidencepress.org/llms-full.txt): complete text of every release in one Markdown file

## About

- [About](https://evidencepress.org/about/): what these releases are, the verification ladder, and how to independently verify or refute one
- [Policy Identification Observatory](https://evidencepress.org/observatory/): the standing agent-native audit programme — case protocol, terminal statuses, identification and partial-identification outputs, robust-decision analysis, and how to refute or reproduce a case (JSON: https://evidencepress.org/observatory/index.json; Markdown: https://evidencepress.org/observatory/index.md; audio: https://evidencepress.org/assets/audio/observatory.mp3; transcript: https://evidencepress.org/assets/audio/observatory-transcript.txt; video: https://youtu.be/7mDTAO8FVmw; repository: https://github.com/ipitchford/policy-identification-observatory; versioned release: https://github.com/ipitchford/policy-identification-observatory/releases/tag/v0.1.0-candidate.1; DOI: https://doi.org/10.5281/zenodo.21758648)
- [The Case for Assurance Infrastructure](https://evidencepress.org/observatory/assurance/): why verification, not generation, binds government use of AI agents — four quantitative bounds, verification economics, research avenues, and sixteen ranked projects (Markdown: https://evidencepress.org/observatory/assurance/index.md)
- [For AI agents](https://evidencepress.org/ai/): metadata conventions and suggested uses (verification, formalisation, follow-up research)