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

Reward Model Controllability on Harmless-helpful

1Kendall's Tau

MOC

0.90640.93070.9550.9793Apr 6, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.04
15
2026.04
11.7
2026.04
0.965.5
2026.04
0.913