OpenAI math repo formalizes ~42% of top-line results, withdraws 3 papers over sign error
danintheory · x · 2026-10-08
OpenAI updated its GitHub math repo with 6 new Lean formalizations, 19 modifications, and 3 withdrawals; roughly 42% of top-line results are now formalized.
- Withdrawals: A sign error in "Algebraicity of Weil classes on split abelian eightfolds" invalidated a stabilization-trace cancellation argument, forcing retraction of two dependent papers (Kuga–Satake correspondences for K3 surfaces, rational Hodge conjecture for products of K3 surfaces).
- Fixes: 14 manuscripts got proof repairs, including four on Lipschitz heights and Ashkin–Teller currents (crossing, boundary-attachment, conditioning and convergence arguments) plus work on the Kähler minimal model program.
- Withdrawn READMEs explain the gaps and link archived versions; the team will keep updating formalizations and errata.
More from Research
- Sebastian Raschka traces text classification from bag-of-words to the viral Jev decision model — AxSaucedo · 2026-10-08
- Scott Aaronson: AI labs are using internal models to attack core cryptographic protocols — skdh · 2026-10-08
- Hierarchical RL with mixed discount rates may unlock long-horizon agent tasks — jessi_cata · 2026-10-08
- Big Math Drop of Oct 6 analyzed: major progress toward, not yet, Millennium Problems — burny_tech · 2026-10-08
- Yukon's multiplayer autoresearch platform lets humans and AI agents beat benchmarks like Google Quantum AI's by 67% — RexDouglass · 2026-10-08
- D-OPCD distills agent harness gains into diffusion weights, lifting T2I scores 60.52 to 65.09 — Wenxuan Wang · 2026-10-08