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

Reward Modeling on RubricBench

59.7Instruction Following (IF) Score

EvoRubrics

43.73647.880552.02556.1695Jun 22, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
59.749.259.43549.551.9
2026.06
57.2658.454.2431.2551.953.01
2026.06
545450.628.847.248.9
2026.06
5060.856.0941.2547.8752.4
2026.06
47.585255.724051.951.53
2026.06
46.7757.656.464046.4550.83
2026.06
45.2505957.54650.7
2026.06
44.355854.6137.546.2149.96