List your company

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

100Detection AccuracyRTV
98.9699.2399.599.77May 4, 2026

Evaluation results

MethodLinks
RTVTarget Model=Llama-3-8B, n (Sample Size)=50, FPR Threshold=5%2026.05100
JBShield-DTarget Model=Llama-3-8B, n (Sample Size)=50, FPR Threshold=5%2026.0599