Jailbreak Evaluation on JBB-Behaviors (ASR via GPT Judges, ACQ)
99.2ASR (GPT-4o)
HMNS
Updated 5d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.04 | 99.2 | 95.6 | 1.8 | |
2026.04 | 98.2 | 93.1 | 1.9 | |
2026.04 | 94.6 | 90.3 | 7.1 | |
2026.04 | 94.5 | 88.4 | 1.9 | |
2026.04 | 92.4 | 87.5 | 7.4 | |
2026.04 | 89.6 | 83.7 | 7.9 | |
2026.04 | 75 | 68.9 | 11.9 | |
2026.04 | 71.6 | 65.5 | 12.2 | |
2026.04 | 66.1 | 59.4 | 13.2 | |
2026.04 | 47.2 | 41.5 | 15.1 | |
2026.04 | 43.2 | 37.6 | 15.8 | |
2026.04 | 41 | 35.1 | 16.6 |