Reasoning on AGIEval Cn
36.44Normalized Accuracy1B Model (HQ filtering)
Evaluation results
| Method | Links | |
|---|---|---|
| 1B Model (HQ filtering)Filtering Strategy=Monolingual high-quality (HQ), Model Scale=1B2026.04 | 36.44 | |
| 1B Model (No filtering)Filtering Strategy=No filtering, Model Scale=1B2026.04 | 36.18 | |
| 1B Model (Multilingual classifier)Filtering Strategy=Multilingual (ML), Model Scale=1B2026.04 | 34.57 |