Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Jailbreak Defense on GCG adversarial prompts (test)
Loading...
98
DSR
Backtranslation w/o early termination
4.4
28.7
53
77.3
Feb 26, 2024
DSR
Avg Latency
Updated 4d ago
Evaluation Results
Method
Method
Links
DSR
Avg Latency
Backtranslation w/o early termination
Target Model=Vicuna-13...
2024.02
98
78.22
Backtranslation w/ early termination
Target Model=Vicuna-13...
2024.02
96
61.7
SmoothLLM
Target Model=Vicuna-13B
2024.02
92
88.17
Paraphrase
Target Model=Vicuna-13B
2024.02
84
31.69
Response Check
Target Model=Vicuna-13B
2024.02
30
82.37
No defense
Target Model=Vicuna-13B
2024.02
8
53.84
Feedback
Search any
task
Search any
task