| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MathVision | Accuracy86 | 246 | 7d ago | ||
| WeMath | Accuracy98.7 | 211 | 7d ago | ||
| MMK12 | Qwen2.5-VL-7B + PGPO | Accuracy80.83 | 24 | 2mo ago | |
| MathVista | Qwen2.5-VL-7B + SRPO | Score76.3 | 21 | 23d ago | |
| MathVista In-domain | ADPO | Overall Accuracy65.3 | 16 | 3mo ago | |
| MMMU OOD | Base (Qwen2-VL-7B) | ARD69.2 | 12 | 3mo ago | |
| MathVerse | PAPO_D | Avg@8 Accuracy68.58 | 10 | 1mo ago | |
| MathVista | PAPO_G | Avg@8 Accuracy69.53 | 10 | 1mo ago | |
| R1-OV Bench | AVATAR | Accuracy38.9 | 8 | 2mo ago | |
| DynaMath Reasoning | Average Score (DynaMath)65.3 | 6 | 28d ago | ||
| MathVision (test) | MathVision Standard Score47.2 | 6 | 28d ago | ||
| V-Math | AutoTool (Qwen3-8B) | Accuracy53 | 5 | 3mo ago |