Skip to content

fix(R5-pass-4): R5-honest disclaimer banners on 4 frozen-mirror docs#592

Merged
gHashTag merged 1 commit into
mainfrom
fix/zenodo-pass-4-honest-deepsweep
May 12, 2026
Merged

fix(R5-pass-4): R5-honest disclaimer banners on 4 frozen-mirror docs#592
gHashTag merged 1 commit into
mainfrom
fix/zenodo-pass-4-honest-deepsweep

Conversation

@gHashTag
Copy link
Copy Markdown
Owner

🔍 ZENODO-DEEPSWEEP PASS-4 — frozen-mirror honesty banners

Fourth iterative R5-honest sweep across all Trinity hive repos. PASS-4 found that 4 mirror/snapshot files still carry pre-audit phrasings ("Full statistical rigor", "Full FAIR compliance", "12 peer-reviewed papers, 150K+ citations", "35+ peer-reviewed papers across 9 research areas").

These are mirror snapshots — rewriting them would diverge them from their upstream source. The R5-honest fix: prepend a disclaimer banner that (a) names the offending phrasings, (b) points to the canonical R5-honest registry, (c) restates Trinity DOIs as software description stubs, NOT papers.

Files patched (4)

  • .claude/projects/-Users-playra-trinity-w1/memory/ZENODO_V16_SUMMARY.md — claimed "Compliance: NeurIPS 2025, ICLR 2025, MLSys 2025 statistical rigor requirements" + "Publish at ICLR 2025 with full statistical rigor".
  • deploy/trinity-nexus/docs/research/COMPARISON_OPENCODE_V33.md — claimed "12 peer-reviewed papers, 150,000+ citations".
  • deploy/trinity-nexus/docs/research/PAS_SCIENTIFIC_ANALYSIS_V32.md — same claim repeated.
  • deploy/trinity-nexus/docs/research/bibliography.md — claimed "Total: 35+ peer-reviewed papers across 9 research areas".

Each file now opens with:

⚠️ R5-HONEST AUDIT 2026-05-12 (PASS-4) — this document is a frozen mirror that contains pre-audit phrasings ... These claims do NOT reflect the current Trinity stance. The canonical R5-honest registry is in docs/research/.zenodo.*_v9.0.json and the cross-sibling registry gHashTag/trios/docs/infrastructure/zenodo-registry.md. Trinity DOIs are software description stubs, NOT peer-reviewed papers. Coq witness in gHashTag/t27: 28 .v files, 218 statements, 162 Qed, 32 Admitted, 11 Abort (2026-05-12 audit).

Sibling PR: t27 PASS-4 (registry rewrite + fake-citation honesty notes)

Anchor: phi^2 + phi^-2 = 3 · R5-HONEST · Throne trios#264

PASS-4 found pre-audit phrasings ('statistical rigor', 'Full FAIR compliance', '12 peer-reviewed papers', '35+ peer-reviewed papers') still live in 4 mirror files. Rather than rewriting frozen-mirror content, prepended an R5-HONEST AUDIT 2026-05-12 banner with pointers to the canonical registry.

Files patched (4):
- .claude/projects/-Users-playra-trinity-w1/memory/ZENODO_V16_SUMMARY.md
- deploy/trinity-nexus/docs/research/COMPARISON_OPENCODE_V33.md
- deploy/trinity-nexus/docs/research/PAS_SCIENTIFIC_ANALYSIS_V32.md
- deploy/trinity-nexus/docs/research/bibliography.md

Anchor: phi^2 + phi^-2 = 3
Issue: gHashTag/trios#264
@gHashTag gHashTag merged commit cdf750b into main May 12, 2026
14 of 25 checks passed
gHashTag added a commit that referenced this pull request May 12, 2026
…D attribution (#593)

PASS-5 of the R5-honest Zenodo audit found two new anomaly classes that
PASS-4 missed by cross-validating cited identifiers against their source
platforms (arxiv.org, orcid.org):

CLASS J — Fabricated arXiv submission (CRITICAL):
  docs/papers/ARXIV_SUBMISSION_CONFIRMATION.md (and its mirror under
  docs/docs/research/papers/) claimed that four papers
  arXiv:2603.00001-00004 were submitted and accepted in March 2026
  under the title 'TRINITY v10.2 — A Unified Framework for Physics
  Blind Spots' by Dmitrii Vasilev. The file additionally reproduced a
  fake moderation@arxiv.org email and a draft press release.

  Verification on 2026-05-12 via curl https://arxiv.org/abs/<id> showed
  all four IDs belong to unrelated authors:
    2603.00001 = Dragović/Murad — Generalized Chapple-Euler Relation
    2603.00002 = Matthews — Hedgehog Reconstruction of Polygons
    2603.00003 = Chen/Gurevych — Commitment Checklist (peer review audit)
    2603.00004 = Nice — Bug Severity Prediction
  An arXiv API search au:Vasilev AND ti:TRINITY returns zero results.

  Both copies of the file are fully replaced with an explicit retraction
  block listing each claimed-vs-actual title/author and stating that no
  arXiv submission has been made under the Trinity name.

CLASS K — Wrong-ORCID attribution (research integrity):
  Across docs/research/ZENODO_BEST_PRACTICES_2025.md,
  docs/research/ZENODO_V19_COMPLETE_SPEC.md,
  docs/research/ZENODO_V19_IMPROVEMENTS.md,
  docs/research/ZENODO_V19_V20_COMPLETE.md,
  docs/research/TRINITY_S3AI_RESEARCH_AGENDA_2025.md, AND
  three production Zig files (tri_zenodo.zig, zenodo_v19_cff.zig,
  zenodo_v19_orcid.zig), Dmitrii Vasilev was attributed ORCID
  0000-0002-1825-0097. That is the well-known Stephen-Hawking-style
  test ORCID, NOT Vasilev's. zenodo_v19_orcid.zig in particular hard-
  coded it inside an enum lookup keyed on .dmitrii_vasilev — meaning
  any future Zenodo deposit signed by this codepath would have shipped
  the wrong ORCID into a live record.

  All occurrences (10 .md + 6 .zig files) replaced with the real
  Vasilev ORCID 0009-0008-4294-6159, verified against
  https://orcid.org/0009-0008-4294-6159.

CLASS L — Honesty banners for bundle summaries:
  docs/research/bundles/QUICK_REFERENCE.md and
  docs/research/ZENODO_V8_COMPLETE_MEGA_SUMMARY.md now carry an
  R5-PASS-5 disclaimer making explicit that B001-B007 and PARENT
  Zenodo deposits are software / description stubs, NOT peer-reviewed
  papers, and that the φ²+φ⁻²=3 anchor is an algebraic identity, not
  a paper claim.

This continues the PASS-3 / PASS-4 line of work (PRs trinity#591/#592,
trinity-fpga#42/#43, t27 PASS-4 banner). NASA report ID:
ZENODO-DEEPSWEEP-PASS-5-RVR-001. Throne: gHashTag/trios#264.

φ² + φ⁻² = 3 · TRINITY · R5-HONEST · PASS-5

Co-authored-by: Trinity Queen Hive <queen-hive@trinity.local>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant