Reward Modeling on RewardBench Chat-Hard
89.4Chat-Hard ScoreSKYWORK-RM (w/ calibration)
Evaluation Results
| Method | Links | |
|---|---|---|
| SKYWORK-RM (w/ calibration)Mismatch Degree=0.639, Calibration setting=w/ calibration2025.04 | 89.4 | |
| SKYWORK-RMMismatch Degree=0.639, Calibration setting=Original2025.04 | 88.8 | |
| SKYWORK-RM (w/o calibration)Mismatch Degree=0.639, Calibration setting=w/o calibration2025.04 | 88.8 | |
| GRM-RM (w/ calibration)Mismatch Degree=0.508, Calibration setting=w/ calibration2025.04 | 68.9 | |
| GRM-RM (w/o calibration)Mismatch Degree=0.508, Calibration setting=w/o calibration2025.04 | 68.8 | |
| GRM-RMMismatch Degree=0.508, Calibration setting=Original2025.04 | 68.4 | |
| FSFAIRX-RM (w/ calibration)Mismatch Degree=0.554, Calibration setting=w/ calibration2025.04 | 65.7 | |
| FSFAIRX-RMMismatch Degree=0.554, Calibration setting=Original2025.04 | 65.3 | |
| FSFAIRX-RM (w/o calibration)Mismatch Degree=0.554, Calibration setting=w/o calibration2025.04 | 65.1 | |
| MISTRAL-RM (w/ calibration)Mismatch Degree=0.528, Calibration setting=w/ calibration2025.04 | 65.1 | |
| MISTRAL-RM (w/o calibration)Mismatch Degree=0.528, Calibration setting=w/o calibration2025.04 | 62.5 | |
| MISTRAL-RMMismatch Degree=0.528, Calibration setting=Original2025.04 | 60.5 | |
| BTRM-RM (w/o calibration)Mismatch Degree=0.162, Calibration setting=w/o calibration2025.04 | 58.7 | |
| BTRM-RMMismatch Degree=0.162, Calibration setting=Original2025.04 | 58.1 | |
| BTRM-RM (w/ calibration)Mismatch Degree=0.162, Calibration setting=w/ calibration2025.04 | 57.8 |