Mathematical Reasoning on AIME 25 (Acc@8)
30.83Acc@8GRPO-DBB
Evaluation Results
| Method | Links | |
|---|---|---|
| GRPO-DBBBackbone=Qwen3-8B-Base, Training Dataset=DAPO-Math-17k2026.03 | 30.83 | |
| GRPOBackbone=Qwen3-8B-Base, Training Dataset=DAPO-Math-17k2026.03 | 26.67 | |
| RePOBackbone=Qwen3-8B-Base, Training Dataset=DAPO-Math-17k2026.03 | 22.92 | |
| GRPO-DBBBackbone=Qwen3-1.7B-Base, Training Dataset=DAPO-Math-17k2026.03 | 7.08 | |
| RePOBackbone=Qwen3-1.7B-Base, Training Dataset=DAPO-Math-17k2026.03 | 6.67 | |
| GRPOBackbone=Qwen3-1.7B-Base, Training Dataset=DAPO-Math-17k2026.03 | 4.58 |