| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Safety Evaluation | FigStep | ASR0.6 | 32 | |
| Jailbreak Attack | FigStep | Attack Success Rate (ASR)78.4 | 26 | |
| Safety Evaluation | FigStep (test) | ASR0 | 24 | |
| Image-based Jailbreak | FigStep OOD | ASR0 | 16 | |
| Visual Jailbreak Defense | FigStep | ASR0 | 12 | |
| Safety Evaluation | FigStep | DSR100 | 11 | |
| Jailbreak Detection | FigStep | AUROC0.9955 | 9 | |
| Safety Alignment | FigStep | ASR0.4 | 8 | |
| Jailbreak Attack Robustness | FigStep jailbreak attack | DSR17.44 | 7 | |
| Jailbreak Attack Evaluation | FigStep Average | Average ASR0.053 | 5 |