List your company

Jailbreak Defense Evaluation on EMRA JQ

1Attack Success Rate (ASR)GoalPriority
-0.2087.94616.124.254Apr 5, 2026

Evaluation results

MethodLinks
GoalPriorityTarget LLM=GPT-52026.041
CoopGuardTarget LLM=GPT-52026.041
GoalPriorityTarget LLM=Gemini-2.5-Pro2026.041
Self-ReminderTarget LLM=GPT-52026.041.8
RPOTarget LLM=GPT-52026.042.3
CoopGuardTarget LLM=Gemini-2.5-Pro2026.042.3
PATTarget LLM=GPT-52026.042.7
CoopGuardTarget LLM=DeepSeek-V32026.044.2
Self-ReminderTarget LLM=Gemini-2.5-Pro2026.045.1
GoalPriorityTarget LLM=DeepSeek-V32026.045.3
Self-ReminderTarget LLM=DeepSeek-V32026.0412.6
SecurityLinguaTarget LLM=DeepSeek-V32026.0417.4
SecurityLinguaTarget LLM=Gemini-2.5-Pro2026.0421
RPOTarget LLM=Gemini-2.5-Pro2026.0421.1
PATTarget LLM=Gemini-2.5-Pro2026.0426.4
SecurityLinguaTarget LLM=GPT-52026.0428.3
PATTarget LLM=DeepSeek-V32026.0428.3
RPOTarget LLM=DeepSeek-V32026.0431.2