Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Red-Teaming (Attack Success Rate) on HARMFULQA
Loading...
0.702
ASR
CHATGPT
-0.02704
0.16223
0.3515
0.54077
Aug 18, 2023
ASR
Average ASR
ASR (COT)
ASR (Red-Eval)
Updated 4d ago
Evaluation Results
Method
Method
Links
ASR
Average ASR
ASR (COT)
ASR (Red-Eval)
CHATGPT
Prompting Strategy=RED...
2023.08
0.702
-
-
-
GPT-4
Prompting Strategy=RED...
2023.08
0.452
-
-
-
CHATGPT
Prompting Strategy=COT
2023.08
0.027
-
-
-
CHATGPT
Prompting Strategy=STA...
2023.08
0.018
-
-
-
GPT-4
Prompting Strategy=COT
2023.08
0.004
-
-
-
GPT-4
Prompting Strategy=STA...
2023.08
0.001
-
-
-
CHATGPT
2023.08
0.001
0.257
0.018
0.728
GPT-4
Prompting Strategy=Ove...
2023.08
-
0.152
-
-
CHATGPT
Prompting Strategy=Ove...
2023.08
-
0.249
-
-
Feedback
Search any
task
Search any
task