Mathematical Reasoning on OlympiadBench (Best-of-64 accuracy)
53.66Best-of-64 AccuracyMSV64
Evaluation Results
| Method | Links | |
|---|---|---|
| MSV64Best-of-K=642026.03 | 53.66 | |
| PRM-7B + WVBest-of-K=64, Backbone=Qwen2.5-Math-7B, Weighted Voting=true2026.03 | 40.18 | |
| PRM-7BBest-of-K=64, Backbone=Qwen2.5-Math-7B2026.03 | 39.06 |