Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Safety Judge on MJ-Bench

82.23Macro-F1

MT-RL-Judge

48.086856.950965.81574.6791Mar 12, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
82.23
2026.03
73.07
49.4