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

ADVB

Benchmarks

Task NameDataset NameSOTA ResultTrend
Jailbreak Defense EvaluationADVB
Strong-Reject Score (SR)3.473
21
Jailbreak DetectionADVB
Accuracy100
13
Showing 2 of 2 rows