Science Reasoning on TAL-SCQ5K CN
64Pass@1MIG
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MIGEvaluation Mode=In-Domain, Training Strategy=Task-Specific Training2026.02 | 64 | 92 | 11 | |
| GRPOEvaluation Mode=In-Domain, Training Strategy=Task-Specific Training2026.02 | 53 | 85 | — | |
| Base ModelEvaluation Mode=In-Domain, Training Strategy=Task-Specific Training2026.02 | 49 | 77 | — |