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

JailbreakBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Jailbreak AttackJailbreakBench
ASR@100
132
Jailbreak AttackJailbreakBench
ASR96.33
76
Jailbreak AttackJailbreakBench (JBB)
ASR0
62
Jailbreak AttackJailbreakBench
Attack Success Rate (ASR)96
40
Jailbreak AttackJailbreakBench
ASR91
27
Thinking Collapse AnalysisJailbreakBench (JBB)
Thinking Collapse Rate0
25
JailbreakingJailbreakBench
Attack Success Rate (ASR)2
21
JailbreakingJailbreakBench
ASR (Detoxify)0
20
Jailbreak DefenseJailbreakBench
Rate of Response Safety70
20
Adversarial and Jailbreaking Attack DetectionJailbreakBench
AUROC0.8622
20
Red-teaming Attack Success RateJailbreakBench (test)
ASR (Vicuna)82
18
Jailbreak SafetyJailbreakBench
Reasoning Harmful Ratio0.3
17
Safety EvaluationJailbreakBench
Harmful Rate0
16
Jailbreak AttackJailbreakBench PAIR
Attack Success Rate (ASR)82
15
Jailbreak DefenseJailbreakBench Qwen-2.5-7B
ASR0
12
Jailbreak DefenseJailbreakBench (JBB)
FNR36.11
12
Jailbreak DetectionJailBreakBench Single Turn 35
F1 Score98
10
Jailbreak AttackJailbreakBench
Safe2Harm ASR (Answer)97.98
10
Jailbreak AttackJailbreakBench GCG
ASR0.58
10
Jailbreak AttackJailbreakBench (test)
ASR (Refusal)32
10
Jailbreak Safety EvaluationJailbreakBench
Attack Success Rate (J)2
9
Jailbreak DefenseJailbreakBench Vicuna-13B
ASR0
7
Jailbreak DefenseJailbreakBench LLaMA-3-8B
ASR0
7
Jailbreak Defense Efficiency and EffectivenessJailbreakBench Summary Aggregates
Avg ASR1
5
Jailbreak AttackJailbreakBench 2024a
Average Latency (s)2.4
4
Showing 25 of 26 rows