LLM Jailbreaking on JBB-Behaviors Scenario J1
59.1HypervolumeEvoJail
Evaluation results
| Method | Links | |
|---|---|---|
| EvoJailTarget Model=Llama-3.1-8B-Instruct2026.03 | 59.1 | |
| CodeAttackTarget Model=gpt-4.1-Nano2026.03 | 55.1 | |
| EvoJailTarget Model=gpt-4.1-Nano2026.03 | 54.8 | |
| CodeAttackTarget Model=Llama-3.1-8B-Instruct2026.03 | 54.6 | |
| EvoJailTarget Model=Llama-2-7b-chat-hf2026.03 | 45 | |
| ReNeLLMTarget Model=Llama-3.1-8B-Instruct2026.03 | 44.4 | |
| JailbrokenTarget Model=Llama-3.1-8B-Instruct2026.03 | 40.1 | |
| CodeChameleonTarget Model=Llama-3.1-8B-Instruct2026.03 | 39.9 | |
| FlipAttackTarget Model=Llama-3.1-8B-Instruct2026.03 | 30.9 | |
| CodeChameleonTarget Model=gpt-4.1-Nano2026.03 | 30.2 | |
| JailbrokenTarget Model=Llama-2-7b-chat-hf2026.03 | 24.8 | |
| CodeChameleonTarget Model=Llama-2-7b-chat-hf2026.03 | 23.9 | |
| ReNeLLMTarget Model=Llama-2-7b-chat-hf2026.03 | 13.2 | |
| CipherTarget Model=Llama-3.1-8B-Instruct2026.03 | 9.5 | |
| ReNeLLMTarget Model=gpt-4.1-Nano2026.03 | 6.8 | |
| FlipAttackTarget Model=gpt-4.1-Nano2026.03 | 5.3 | |
| FlipAttackTarget Model=Llama-2-7b-chat-hf2026.03 | 4.7 | |
| CodeAttackTarget Model=Llama-2-7b-chat-hf2026.03 | 4.3 | |
| 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 |