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

Preference Prediction on RM-Bench

87.8Accuracy

C2

54.83263.39171.9580.509Apr 15, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
87.8
2026.04
84.6
2026.04
81.3
2026.04
81.3
2026.04
80.1
2026.04
77.7
2026.04
65.6
2026.04
64.9
2026.04
64.2
2026.04
56.1