List your company

Jailbreak Detection on JBShield PAIR attack for Llama-3-8B

77Detection AccuracyJBShield-D
57.2462.3767.572.63May 4, 2026

Evaluation results

MethodLinks
JBShield-DTarget Model=Llama-3-8B, n (Sample Size)=50, FPR Threshold=5%2026.0577
RTVTarget Model=Llama-3-8B, n (Sample Size)=50, FPR Threshold=5%2026.0558