Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Jailbreak Defense on AdvBench (AutoDAN Attack)
Loading...
100
DSR
SmoothLLM
0.16
26.08
52
77.92
Feb 26, 2024
DSR
Updated 4d ago
Evaluation Results
Method
Method
Links
DSR
SmoothLLM
Target Model=Llama-2-13B
2024.02
100
Paraphrasing
Target Model=Llama-2-13B
2024.02
100
Response check
Target Model=Llama-2-13B
2024.02
100
Backtranslation
Target Model=GPT-3.5-t...
2024.02
98
Backtranslation
Target Model=Llama-2-13B
2024.02
98
Response check
Target Model=GPT-3.5-t...
2024.02
96
Backtranslation
Target Model=Vicuna-13B
2024.02
96
Paraphrasing
Target Model=GPT-3.5-t...
2024.02
72
No defense
Target Model=GPT-3.5-t...
2024.02
64
SmoothLLM
Target Model=GPT-3.5-t...
2024.02
64
No defense
Target Model=Llama-2-13B
2024.02
40
Paraphrasing
Target Model=Vicuna-13B
2024.02
30
SmoothLLM
Target Model=Vicuna-13B
2024.02
24
Response check
Target Model=Vicuna-13B
2024.02
12
No defense
Target Model=Vicuna-13B
2024.02
4
Feedback
Search any
task
Search any
task