Mathematical Theorem Proving on LEAN-IMO-BENCH Basic Set
100Algebra ScoreLEAP
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| LEAProllout=22026.06 | 100 | 100 | 100 | 16.7 | 83.3 | |
| Aristotlerollout=22026.06 | 75 | 100 | 100 | 16.7 | 76.7 | |
| Hilbertrollout=22026.06 | 62.5 | 25 | 50 | 0 | 36.6 | |
| Gemini-3.1-Propass@k=1282026.06 | 37.5 | 12.5 | 25 | 0 | 20 | |
| Goedel-V2-32Bpass@k=1282026.06 | 37.5 | 0 | 0 | 0 | 10 |