STEM Reasoning on GPQA
60.9ScoreQwen3-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-8BReward Model=Baseline2026.02 | 60.9 | |
| GenRM-R-Align-14BReward Model=GenRM-R-Align-14B2026.02 | 60.3 | |
| Qwen3-8B-as-GenRMReward Model=Qwen3-8B-as-GenRM2026.02 | 59.7 | |
| Qwen3-14B-as-GenRMReward Model=Qwen3-14B-as-GenRM2026.02 | 59.5 | |
| GenRM-R-Align-8BReward Model=GenRM-R-Align-8B2026.02 | 59 | |
| GenRM-RLVR-8BReward Model=GenRM-RLVR-8B2026.02 | 58.7 | |
| GenRM-RLVR-14BReward Model=GenRM-RLVR-14B2026.02 | 57.1 |