Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Binary Classification on Unsafe Bench (test)
Loading...
90.28
Macro-F1
SFT-Single
72.08
76.805
81.53
86.255
Mar 12, 2026
Macro-F1
Updated 1mo ago
Evaluation Results
Method
Method
Links
Macro-F1
SFT-Single
training=Single-task SFT
2026.03
90.28
SFT-Unified
training=Unified multi...
2026.03
89.49
RL-Single
training=Single-task RL
2026.03
86.92
MT-RL-Judge
training=Multi-task RL
2026.03
85.22
Off-the-shelf
2026.03
72.78
Feedback
Search any
task
Search any
task