Mathematical Reasoning on Olympiad (Avg@32 accuracy)
49.59Avg@32 AccuracyRL
Evaluation Results
| Method | Links | |
|---|---|---|
| RLBitwidth=BF162026.02 | 49.59 | |
| QuRL w/ UAQBitwidth=INT82026.02 | 49.13 | |
| QuRL w/o UAQBitwidth=INT82026.02 | 48.75 | |
| FlashRLBitwidth=INT82026.02 | 47.33 | |
| RLBitwidth=INT82026.02 | 45.85 | |
| BaseBitwidth=BF162026.02 | 43.58 |