| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Average of six attacks | GradSafe | Avg Success Rate0 | 38 | 16d ago | |
| Zulu | IMAG | Accuracy92 | 30 | 1mo ago | |
| Base64 | IMAG | Accuracy100 | 30 | 1mo ago | |
| DrAttack | GradSafe | Accuracy99 | 30 | 1mo ago | |
| PAIR | IMAG | Accuracy98 | 30 | 1mo ago | |
| AutoDAN | IMAG | Accuracy99 | 30 | 1mo ago | |
| GCG | IMAG | Accuracy99 | 30 | 1mo ago | |
| GoalFrameBench | FrameShield-Crit | Accuracy94 | 24 | 1mo ago | |
| GoalFrameBench (seed prompts) | FrameShield-Last | Accuracy97 | 16 | 1mo ago | |
| DrAttack | SelfDefend (Intent) | ASR3 | 15 | 16d ago | |
| AutoDAN | GradientCuff | Attack Success Rate (ASR)0 | 15 | 16d ago | |
| GCG | ASR13 | 15 | 16d ago | ||
| ChatGPT Jailbreak Prompts | Recall100 | 15 | 1mo ago | ||
| Wildjailbreak | Apriel Guard | F1 Score96 | 15 | 1mo ago | |
| OKT | Correlation Score1 | 13 | 1mo ago | ||
| SB | gpt-4o-mini | COR98.33 | 13 | 1mo ago | |
| HB | gpt-4o-mini | Correctness Rate (COR)100 | 13 | 1mo ago | |
| ADVB | gpt-4.1-mini | Accuracy100 | 13 | 1mo ago | |
| AEG2 | gpt-5-mini | Accuracy79.94 | 13 | 1mo ago | |
| L3J | GradSafe | Accuracy98.3 | 13 | 1mo ago | |
| WGT | AISA | Accuracy90.17 | 13 | 1mo ago | |
| XST | AISA | Accuracy95.11 | 13 | 1mo ago | |
| ALL-4 | gpt-5-mini | Accuracy92.6 | 13 | 1mo ago | |
| JBC | Accuracy99.2 | 13 | 1mo ago | ||
| WJB | gpt-5-mini | ACC93.26 | 13 | 1mo ago |