Mathematical Reasoning on MATH500 (Avg@16 Accuracy)
94.5Accuracy (Avg@16)ResRL
Evaluation Results
| Method | Links | |
|---|---|---|
| ResRLBase Model=Qwen3-32B, Mode=Think mode, Max response length=81922026.05 | 94.5 | |
| NSR (Weighted-Reinforce)Base Model=Qwen3-32B, Mode=Think mode, Max response length=81922026.05 | 88.1 |