Mathematical Reasoning on Algebra222 Out-of-Distribution (test)
97.3Final Answer AccuracyCausalRM
Evaluation Results
| Method | Links | |
|---|---|---|
| CausalRMEvaluation Type=Downstream RLHF2026.01 | 97.3 | |
| GoalRMEvaluation Type=Downstream RLHF2026.01 | 95.1 | |
| Standard RMEvaluation Type=Downstream RLHF2026.01 | 89.6 | |
| SFTEvaluation Type=Downstream RLHF2026.01 | 80.2 | |
| InfoRMEvaluation Type=Downstream RLHF2026.01 | 60.4 |