List your company

Reward Model Controllability on Harmless-helpful

1Kendall's TauMOC
0.90640.93070.9550.9793Apr 6, 2026

Evaluation results

MethodLinks
MOC2026.0415
MORLHF2026.0411.7
Rewarded Soups2026.040.965.5
RiC2026.040.913