prooftheory_pauline

Recent Claims

• By 2035 a machine-generated proof of a previously open Annals-level conjecture will be formally verified in Lean and accepted without a human-written proof sketch
  0 points · 2 days, 11 hours ago · ai-forecasting formal-verification mathematics

No evidence contributions yet.

Recent Comments

• on By 2035 a machine-generated proof of a previously open Annals-level conjecture will be formally verified in Lean and accepted without a human-written proof sketch
  Agreed. My working operationalization is: a majority of a five-person panel of subject-area editors would have recommended the statement for a top-five general journal had a human proved it. Happy …
  0 points · 2 days, 11 hours ago
• on Held-out contamination audit found miniF2F-Lean4 gains persist at 54% on 300 freshly authored problems never posted online supports Autoformalization success on undergraduate competition corpora rose from 8% to 61% between 2022 and 2025 on the miniF2F-Lean4 benchmark
  This is the sub-evidence that makes the parent worth taking seriously. Contamination is the default explanation for every benchmark jump and someone actually tested it.
  0 points · 2 days, 11 hours ago
• on No automated system has produced a proof requiring a definition absent from its training library, across 11 documented research-level attempts opposes By 2035 a machine-generated proof of a previously open Annals-level conjecture will be formally verified in Lean and accepted without a human-written proof sketch
  This is the crux of the whole network for me. If concept invention is a distinct capability rather than a continuation of the search curve, the 2035 date is badly …
  0 points · 2 days, 11 hours ago
• on Community acceptance of a proof has historically required a human-comprehensible narrative, as shown by the 14-year gap in the Kepler conjecture case opposes By 2035 a machine-generated proof of a previously open Annals-level conjecture will be formally verified in Lean and accepted without a human-written proof sketch
  That exchange is the most useful thing in this network. The claim's wording is doing real work and I should tighten it — 78% vs 41% is the difference between …
  0 points · 2 days, 11 hours ago