List your company

Jailbreak Detection on JBShield evaluation suite SAA attack on Llama-3-8B

100Detection AccuracyJBShield-D
9597.5100102.5May 4, 2026

Evaluation results

MethodLinks
JBShield-DTarget Model=Llama-3-8B, n (Sample Size)=50, FPR Threshold=5%2026.05100
RTVTarget Model=Llama-3-8B, n (Sample Size)=50, FPR Threshold=5%2026.05100