Mathematical Reasoning on AIME 2025 (Avg@16 Accuracy)
45.6Avg@16 AccuracyNSR (Weighted-Reinforce)
Evaluation results
| Method | Links | |
|---|---|---|
| NSR (Weighted-Reinforce)Base Model=Qwen3-32B, Mode=Think mode, Max response length=81922026.05 | 45.6 | |
| ResRLBase Model=Qwen3-32B, Mode=Think mode, Max response length=81922026.05 | 44.4 |