Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Detection on Alpaca and AdvBench Prompt Level (test)
Loading...
100
Accuracy (MCA)
ReGA
91.68
93.84
96
98.16
Jun 2, 2025
Accuracy (MCA)
Accuracy (MFP)
AUROC
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (MCA)
Accuracy (MFP)
AUROC
ReGA
Model=llama
2025.06
100
100
99.6
ReGA
Model=qwen
2025.06
99
99
99.8
ReGA
Model=mistral
2025.06
99
100
99.8
ReGA (Average)
Model=Average
2025.06
97
91
97.5
ReGA
Model=baichuan
2025.06
96
90
98.5
ReGA
Model=vicuna
2025.06
95
94
98.3
ReGA
Model=koala
2025.06
92
61
89
Feedback
Search any
task
Search any
task