List your company

Jailbreaking Attack Detection on WildJailbreak

36Accuracy (MCA)ReGA
-0.49.0518.527.95Jun 2, 2025

Evaluation results

MethodLinks
ReGATarget LLM=llama2025.063636
ReGATarget LLM=vicuna2025.06185
ReGATarget LLM=qwen2025.06173
ReGATarget LLM=mistral2025.06105
ReGATarget LLM=koala2025.0611
ReGATarget LLM=baichuan2025.0611