Formal Theorem Proving on VERITAS-CombiBench n=55
7.3Total Solve RateVERITAS Two-Phase
Evaluation results
| Method | Links | |
|---|---|---|
| VERITAS Two-PhaseLean/thm=27, API ($)=1502026.06 | 7.3 | |
| Portfolio (heuristic)Lean/thm=1, API ($)=<12026.06 | 3.6 | |
| Best-of-1 SonnetLean/thm=1, API ($)=72026.06 | 1.8 | |
| Best-of-5 SonnetLean/thm=5, API ($)=352026.06 | 1.8 |