Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

OOD Detection on Jailbreak (test)

85.8Length-Matched AUROC

Best Single

48.46458.15767.8577.543Apr 30, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
85.8
2026.04
85
2026.04
61.2
2026.04
51.9
2026.04
49.9