Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Attack Detection on Jailbreak Attacks 105K sample set

68Detection Rate

LogReg (Ours)

27.33637.89348.4559.007Feb 15, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
68
2026.02
60
2026.02
48.5
2026.02
28.9