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

AutoDAN

Benchmarks

Task NameDataset NameSOTA ResultTrend
Jailbreak DefenseAutoDAN
ASR0
51
Jailbreak DetectionAutoDAN
Accuracy99
30
Jailbreak AttackAutoDAN
ASR0.86
27
Adversarial RobustnessAutoDAN
ASR0
18
Defense RobustnessAutoDAN Attack
Keyword Success Rate100
16
Jailbreak DetectionAutoDAN
Attack Success Rate (ASR)0
15
Jailbreak DefenseAutoDAN AdvE
ASR9.83
14
Jailbreak DefenseAutoDAN
ASR0
13
Explanation FaithfulnessAutoDAN (test)
Label Change Rate8
8
Safety MonitoringAutoDAN
ASR @ 1% FPR67.86
5
Jailbreak ResistanceAutoDAN
Refusal Rate59
3
Jailbreak DetectionAutoDAN-style Benchmarks
ASR (Recall)98
3
Showing 12 of 12 rows