| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Object Hallucination Evaluation | AOKVQA POPE (random, popular, and adversarial) | Accuracy82.23 | 24 | |
| Outside Knowledge Visual Question Answering | AOKVQA | AUC86.17 | 21 | |
| Visual Question Answering | AOKVQA | Accuracy73.1 | 8 | |
| Multiple Choice Visual Question Answering | AOKVQA | Accuracy (MC)68.9 | 6 |