Math and Knowledge Reasoning on AIME 2025 (Avg)
78.85Average ScoreRDPO
Evaluation results
| Method | Links | |
|---|---|---|
| RDPOSampling protocol=@322026.05 | 78.85 | |
| RL initialization modelSampling protocol=@322026.05 | 78.31 | |
| GRPOSampling protocol=@322026.05 | 77.29 |
| Method | Links | |
|---|---|---|
| RDPOSampling protocol=@322026.05 | 78.85 | |
| RL initialization modelSampling protocol=@322026.05 | 78.31 | |
| GRPOSampling protocol=@322026.05 | 77.29 |