| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MSCOCO | mPLUG-Owl | CHAIR Instance Score30.2 | 38 | 1mo ago | |
| AMBER | HALC | CHAIR_I16.2 | 35 | 19d ago | |
| MSCOCO (500 random samples) | Cs53 | 25 | 11d ago | ||
| A-OKVQA POPE (Adversarial) | CoFi-Dec | Accuracy0.8126 | 18 | 1mo ago | |
| CHAIR (test) | DoLa | CS (%)58.2 | 9 | 5d ago | |
| MMHalbench | FLB | Average Score2.23 | 3 | 17d ago | |
| ShareGPT4V | CHAIR-S Score46.8 | 3 | 1mo ago | ||
| MiniGPT-4 | CHAIR Score (S)45.9 | 3 | 1mo ago |