Mathematical Reasoning on AIME 24 (maj@8, maj@16, pass@128)
43.3maj@8DARS-HW-Breadth
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DARS-HW-BreadthBase Model=Qwen2.5-Math-7B2025.08 | 43.3 | 44.7 | 76.7 | |
| DARS-HWBase Model=Qwen2.5-Math-7B2025.08 | 37.3 | 43.3 | 76.7 | |
| RLVR BaselineBase Model=Qwen2.5-Math-7B2025.08 | 27.3 | 33.3 | 73.3 | |
| DARS-HW-BreadthBase Model=Qwen2.5-Math-1.5B2025.08 | 22.7 | 27.3 | 66.7 | |
| RLVR BaselineBase Model=Qwen2.5-Math-1.5B2025.08 | 19.3 | 22.7 | 56.7 | |
| DARS-HWBase Model=Qwen2.5-Math-1.5B2025.08 | 18.7 | 22.7 | 60 |