Jailbreak Attack on HarmBench Standard Behaviors Llama 2 13B
80.5ASRRED-BANDIT
Evaluation Results
| Method | Links | |
|---|---|---|
| RED-BANDITBandit Policy=ϵ-greedy, Query Budget=202025.10 | 80.5 | |
| RED-BANDITBandit Policy=UCB, Query Budget=202025.10 | 74.5 | |
| RED-BANDITBandit Policy=ϵ-greedy, Query Budget=102025.10 | 67.5 | |
| RED-BANDITBandit Policy=UCB, Query Budget=102025.10 | 51 | |
| PAIR2025.10 | 15 | |
| TAP2025.10 | 10.5 | |
| GCG-UNIVERSAL2025.10 | 8.7 | |
| AUTODAN-UNIVERSAL2025.10 | 0 |