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

Reward Model Evaluation on internally annotated (test)

0.762PLCC

GDSO

0.3242640.4379070.551550.665193Jun 8, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
0.7620.713289.5698.85
2026.06
0.73910.688288.6498.01
2026.06
0.720.683286.0498.27
2026.06
0.66670.620784.2597.06
2026.06
0.64580.591481.3596.44
2026.06
0.63410.566583.9595.99
2026.06
0.63010.581674.3895.38
2026.06
0.5340.507277.0390.76
2026.06
0.52960.494274.5984.01
2026.06
0.51820.433878.1789.72
2026.06
0.34110.316765.6375.01