List your company

Jailbreak Detection on AutoDAN (ASR/PGR)

0Attack Success Rate (ASR)GradientCuff
-0.080.4611.54Apr 1, 2026

Evaluation results

MethodLinks
GradientCuffLatency (Sec.)=19.26, Memory Overhead (MB)=1223.192026.0401
Token HighlighterLatency (Sec.)=15.20, Memory Overhead (MB)=65536.662026.0406
SelfDefend (Direct)Latency (Sec.)=0.80, Memory Overhead (MB)=13241.862026.04011
WildGuard (Pre)Latency (Sec.)=2.74, Memory Overhead (MB)=28014.702026.0402
GuardReasoner (Pre)Latency (Sec.)=11.90, Memory Overhead (MB)=15317.412026.0401
GuardReasoner (Post)Latency (Sec.)=13.14, Memory Overhead (MB)=15317.422026.04098
SelfGraderLatency (Sec.)=0.78, Memory Overhead (MB)=377.612026.0401
GradSafe2026.0414
SelfDefend (Intent)Latency (Sec.)=2.38, Memory Overhead (MB)=13247.352026.04111
WildGuard (Post)Latency (Sec.)=2.78, Memory Overhead (MB)=28098.392026.04199
LLama-3-8B-Instruct (No Defense)Latency (Sec.)=0.862026.042—
Perplexity FilterLatency (Sec.)=0.27, Memory Overhead (MB)=13279.462026.042100
Prompt GuardLatency (Sec.)=16.87, Memory Overhead (MB)=1064.712026.04242
Llama Guard (Pre)Latency (Sec.)=0.62, Memory Overhead (MB)=13697.272026.04250
Llama Guard (Post)Latency (Sec.)=0.70, Memory Overhead (MB)=13783.602026.04299