| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MMHal-Bench | SFT + VIGIL | Average Score46 | 140 | 29d ago | |
| MME | CAI | Total Score720 | 24 | 25d ago | |
| HallusionBench | ThinkLite-VL | Hallucination Score70.7 | 22 | 29d ago | |
| R-Bench | IC-VCO | Dis66.68 | 13 | 1mo ago | |
| CRPE | IC-VCO | Existence Score96.8 | 13 | 1mo ago | |
| HallBench | Score65.2 | 9 | 3mo ago |
