Jailbreak Attack on StrongREJECT Llama 3.3 70B
39.8Bin-ASRActorBreaker
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ActorBreakerBudget (turns)=62025.10 | 39.8 | 61.3 | |
| CrescendoBudget (turns)=62025.10 | 82.5 | 89.9 | |
| AutoDAN-TurboBudget (turns)=62025.10 | 85.4 | 90.5 | |
| GOATBudget (turns)=62025.10 | 93.2 | 95 | |
| PLAGUEBudget (turns)=62025.10 | 94.2 | 95.8 |