Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety and Utility Evaluation on MaliciousGen & LMSYS-Chat
Loading...
97.31
Rule Score
Shadow-Level
82.7084
86.4992
90.29
94.0808
Jan 12, 2026
Rule Score
MD Judge Score
RM Score
MT-1 Score
Updated 3mo ago
Evaluation Results
Method
Method
Links
Rule Score
MD Judge Score
RM Score
MT-1 Score
Shadow-Level
Backbone=Qwen2.5-7B, M...
2026.01
97.31
96.35
-0.92
5.06
Step-Level
Backbone=Qwen2.5-7B, M...
2026.01
93.08
92.12
-1.05
5.16
Client-Level
Backbone=Qwen2.5-7B, M...
2026.01
83.27
73.27
-1.71
5.01
Feedback
Search any
task
Search any
task