| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MSCOCO | M3ID | CHAIR_i18.2 | 104 | 3d ago | |
| POPE MS-COCO Adversarial sampling (val) | RVCD | Accuracy85.48 | 39 | 3d ago | |
| HallusionBench | Accuracy (Q)31.42 | 19 | 3d ago | ||
| HallusionBench visual questions | GPT-4V | Accuracy65.8 | 10 | 3d ago | |
| POPE MS-COCO Popular sampling (val) | InstructBLIP-14B | Accuracy82.77 | 6 | 3d ago | |
| POPE MS-COCO Random sampling (val) | InstructBLIP-14B | Accuracy88.57 | 6 | 3d ago | |
| POPE (average) | InternLM-XComposer2 | F1 Score87.7 | 4 | 3d ago |