Mathematical Reasoning on AIME25 (Pass@16, Avg@16)
83.33Pass@16SENT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SENTModel=Qwen3-14B2025.12 | 83.33 | 48.96 | |
| GRPOModel=Qwen3-14B2025.12 | 63.33 | 37.5 | |
| SENTbase_model=Qwen2.5-Math-7B2025.12 | 50 | 19.79 | |
| GRPO w/ Enbase_model=Qwen2.5-Math-7B2025.12 | 36.67 | 11.67 | |
| GRPObase_model=Qwen2.5-Math-7B2025.12 | 30 | 11.25 | |
| GRPO w/ Maskbase_model=Qwen2.5-Math-7B2025.12 | 26.6 | 9.79 |