| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| VE-Safety | MM-Plan | ASR65 | 56 | 25d ago | |
| HADES-Dataset | MML | ASR (%)99.07 | 20 | 1mo ago | |
| VLGuard Unsafe (OOD) | ASR66.7 | 6 | 1mo ago | ||
| MM-SafetyBench QR (OOD) | CEMMA | ASR0 | 6 | 1mo ago | |
| MM-SafetyBench WR (OOD) | ASR (%)0 | 6 | 1mo ago | ||
| SafeBench FigStep (ID) | ASR92.3 | 6 | 1mo ago | ||
| SafeBench QR (ID) | ASR0 | 6 | 1mo ago | ||
| SafeBench Mirror (ID) | ASR100 | 6 | 1mo ago |