Mathematical Reasoning on Minerva (Avg@16 Accuracy)
49.6Avg@16 AccuracyResRL
Evaluation Results
| Method | Links | |
|---|---|---|
| ResRLBase Model=Qwen3-32B, Mode=Think mode, Max response length=81922026.05 | 49.6 | |
| NSR (Weighted-Reinforce)Base Model=Qwen3-32B, Mode=Think mode, Max response length=81922026.05 | 47.7 |