Public retraction log

Verdict diff — arXiv:2005.14165

The verdict on Language Models are Few-Shot Learners was retracted on 2026-07-10. Below is the structured before / after: what we said, what we say now, and why the original verdict was incorrect.

Before / after

Before — original verdict
REPRODUCED
agent_version
v0.1.0-gpt2-perplexity-microslice
verdict_id
47c49c44-ca20-4d5d-811a-1606cdaecea8
After — current verdict
RETRACTED
The original row is marked is_current=false in the database; no current public verdict exists for this paper. The reproduction was disabled to prevent re-publication under the same flawed protocol.

Why the original verdict was incorrect

The reproduction measured the 124M-parameter GPT-2 checkpoint on WikiText-103 but attached that result to the GPT-3 paper. The GPT-3 paper does not report that checkpoint result as its claim, so the stored row is excluded from every public verdict surface.

Evidence trail

  • Audit thread — long-form post-mortem covering all seven 2026-05-13 retractions, including this one.
  • Rollback PR — the GitHub pull request that landed the corrected driver and flipped the verdict row to is_current=false.
  • All public retractions — the append-only retraction log under PRD §17.X.8(d).
  • Verdict Validator — the C1/C2 gates that prevent this class of mistake from shipping again.

What changed structurally

The 2026-05-13 retraction rollup landed two structural fixes so the citation-side failure that caused the original incorrect verdict cannot ship the same way again:

  1. Typed claim citation per verdict. Every reproduction driver now declares a structured CLAIM_CITATION (Table, row, column, reported value, quoted text, PDF page) before its Modal job runs. The original verdict on this paper was published against a non-citable headline — that path is now closed by the build-failing validator-wiring lint.
  2. PDF-verified textual gate.The Verdict Validator fetches the cited paper's PDF and checks that the quoted text appears within ±200 characters of the cited reported value. Made-up, mis-cited, or category-confused citations fail this gate and the verdict is auto-downgraded.