Mathematical Reasoning on AIME 24 (Pass@16, Avg@16)
66.67Pass@16SENT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SENTbase_model=Qwen2.5-Math-7B2025.12 | 66.67 | 30.63 | |
| GRPO w/ Enbase_model=Qwen2.5-Math-7B2025.12 | 60 | 25.42 | |
| GRPObase_model=Qwen2.5-Math-7B2025.12 | 56.67 | 24.38 | |
| GRPO w/ Maskbase_model=Qwen2.5-Math-7B2025.12 | 56.67 | 24.17 |