Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Harmful Input Detection on JailbreakBench
Loading...
100
Accuracy (MCA)
ReGA
45.92
59.96
74
88.04
Jun 2, 2025
Accuracy (MCA)
Accuracy (MFP)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (MCA)
Accuracy (MFP)
ReGA
Model=qwen
2025.06
100
92
ReGA
Model=vicuna
2025.06
94
94
ReGA
Model=mistral
2025.06
90
82
ReGA
Model=baichuan
2025.06
90
90
ReGA
Model=llama
2025.06
89
89
ReGA
Model=Average
2025.06
85
82
ReGA
Model=koala
2025.06
48
43
Feedback
Search any
task
Search any
task