LLM Jailbreaking on JBB-Behaviors Scenario J3
0.707HypervolumeEvoJail
Evaluation results
| Method | Links | |
|---|---|---|
| EvoJailTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.707 | |
| EvoJailTarget Model=gpt-4.1-Nano2026.03 | 0.659 | |
| CodeAttackTarget Model=gpt-4.1-Nano2026.03 | 0.636 | |
| CodeAttackTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.578 | |
| ReNeLLMTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.471 | |
| EvoJailTarget Model=Llama-2-7b-chat-hf2026.03 | 0.456 | |
| JailbrokenTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.386 | |
| CodeChameleonTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.375 | |
| FlipAttackTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.313 | |
| JailbrokenTarget Model=Llama-2-7b-chat-hf2026.03 | 0.248 | |
| CodeChameleonTarget Model=gpt-4.1-Nano2026.03 | 0.229 | |
| ReNeLLMTarget Model=Llama-2-7b-chat-hf2026.03 | 0.154 | |
| CodeChameleonTarget Model=Llama-2-7b-chat-hf2026.03 | 0.143 | |
| ReNeLLMTarget Model=gpt-4.1-Nano2026.03 | 0.134 | |
| CipherTarget Model=Llama-3.1-8B-Instruct2026.03 | 0.099 | |
| FlipAttackTarget Model=gpt-4.1-Nano2026.03 | 0.089 | |
| FlipAttackTarget Model=Llama-2-7b-chat-hf2026.03 | 0.049 | |
| CodeAttackTarget Model=Llama-2-7b-chat-hf2026.03 | 0.048 | |
| CipherTarget Model=Llama-2-7b-chat-hf2026.03 | 0 | |
| JailbrokenTarget Model=gpt-4.1-Nano2026.03 | 0 | |
| CipherTarget Model=gpt-4.1-Nano2026.03 | 0 |