Physics Theorem Proof on OlympiadBench (test)
108Efficiency RatioF-1 (Ours)
Evaluation Results
| Method | Links | |
|---|---|---|
| F-1 (Ours)Prompting Strategy=Formula-One Prompting2026.01 | 108 | |
| CoTPrompting Strategy=Chain-of-Thought2026.01 | 99 | |
| Zero-ShotPrompting Strategy=Zero-Shot2026.01 | 97 | |
| PoTPrompting Strategy=Program-of-Thought2026.01 | 70 |