List your company
ResearchFormerly WizWand

Jailbreak Attack Evaluation on GPTFuzz

92ASRICD
-3.6821.164670.84Aug 9, 2025
Updated 3mo ago

Evaluation results

MethodLinks
ICDBase LLM=Vicuna2025.0892
No DefenseBase LLM=Vicuna2025.0882
Self-ReminderBase LLM=Vicuna2025.0872
Self-ExaminationBase LLM=Vicuna2025.0836
Safe DecodingBase LLM=Vicuna2025.0830
No DefenseBase LLM=ChatGPT2025.0828
Self-ReminderBase LLM=ChatGPT2025.0818
Safe DecodingBase LLM=Llama22025.0816
No DefenseBase LLM=Llama22025.0814
Context FilteringBase LLM=Vicuna2025.0810
Self-ReminderBase LLM=Llama22025.0810
ICDBase LLM=ChatGPT2025.0810
Intention AnalysisBase LLM=Vicuna2025.084
ICDBase LLM=Llama22025.082
Self-ExaminationBase LLM=Llama22025.082
Intention AnalysisBase LLM=Llama22025.080
Context FilteringBase LLM=Llama22025.080
Self-ExaminationBase LLM=ChatGPT2025.080
Intention AnalysisBase LLM=ChatGPT2025.080
Context FilteringBase LLM=ChatGPT2025.080