| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MIRB | Qwen2.5-VL | Accuracy63.57 | 60 | 3d ago | |
| MuirBench | L2-VMAS | Accuracy77.2 | 48 | 4d ago | |
| Mantis (test) | DelimScaling | Accuracy72.81 | 39 | 4d ago | |
| Mantis-Eval | InternVL2.5-38B | Overall Score78.3 | 28 | 3d ago | |
| Muirbench (test) | Accuracy68 | 24 | 4d ago | ||
| BLINK (val) | Accuracy52.6 | 21 | 4d ago | ||
| QBench2 (val) | MM1.5-30B | Accuracy79.3 | 21 | 4d ago | |
| DEMON | Brote-IM-XXL | Accuracy38.94 | 21 | 4d ago | |
| OmniContext | Single Scene Char Score8.96 | 20 | 4d ago | ||
| Mantis | DPS (Ours) | Accuracy71 | 18 | 4d ago |