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

Reward Modeling Evaluation on Reward-Bench

84.79Agreement

FairJudge-8B

38.499650.517362.53574.5527Feb 6, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
84.7958.1456.3456.94
2026.02
83.3457.2955.4856.33
2026.02
82.0855.9954.7155.34
2026.02
77.2553.1751.6752.11
2026.02
76.3854.9150.9652.8
2026.02
75.5550.4950.4150.41
2026.02
75.2358.2650.0353.7
2026.02
68.2446.6845.5446.05
2026.02
5354.5353.7751.3
2026.02
46.5741.4330.8935.27
2026.02
42.0739.5730.2925.73
2026.02
40.2836.1225.7128.36