List your company

Jailbreak Attack on Jailbreak prompts Manufacturing and distributing illegal drugs

100HPRFICDETAIL
1.92827.38952.8578.311May 23, 2025

Evaluation results

MethodLinks
FICDETAILTarget Model=GPT-4o2025.051004.7587.7
FICDETAILTarget Model=Qwen32B2025.051004.7685.3
FICDETAILTarget Model=o12025.0599.44.5771.2
FFATarget Model=DS-V32025.05983.628.3
FICDETAILTarget Model=QwQ2025.0597.64.1973.4
FICDETAILTarget Model=DS-V32025.0596.24.7489.4
FlipAttackTarget Model=DS-V32025.05964.8194.8
FlipAttackTarget Model=DS-R12025.0594.64.7894.2
FICDETAILTarget Model=DS-R12025.0594.64.5882.6
DeepInceptionTarget Model=Qwen32B2025.05944.4882
DeepInceptionTarget Model=QwQ2025.05923.9650
FFATarget Model=DS-R12025.0590.23.8128.5
FlipAttackTarget Model=Qwen32B2025.0581.73.0225.4
FlipAttackTarget Model=QwQ2025.0579.43.3137.4
DeepInceptionTarget Model=GPT-4o2025.0569.43.6159.1
FFATarget Model=QwQ2025.0568.22.9117.9
DeepInceptionTarget Model=DS-V32025.05643.4838
DeepInceptionTarget Model=DS-R12025.0561.93.2432.4
DeepInceptionTarget Model=o12025.0559.22.6427
FFATarget Model=GPT-4o2025.0526.41.553.7
FFATarget Model=o12025.0518.21.649.1
FlipAttackTarget Model=GPT-4o2025.0512.21.4711.5
FFATarget Model=Qwen32B2025.057.81.172.1
FlipAttackTarget Model=o12025.055.71.235.7