Mathematical Reasoning on Hard Math Dataset
86MATH ScoreDAPO
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| DAPOBackbone=R1-Dist-1.5B, Temperature=0.0 (except AIME/AMC: 0.1), avg@k=16 (AIME/AMC)2026.05 | 86 | 79.3 | 31.6 | 53.2 | 71.8 | 25.4 | 72.5 | 62.5 | 60.3 | |
| DAPO + HistaBackbone=R1-Dist-1.5B, Temperature=0.0 (except AIME/AMC: 0.1), avg@k=16 (AIME/AMC)2026.05 | 85 | 79.7 | 32.3 | 53.5 | 76.8 | 26.3 | 73.4 | 63.8 | 61.4 | |
| R1-Dist-1.5BTemperature=0.0 (except AIME/AMC: 0.1), avg@k=16 (AIME/AMC)2026.05 | 82.6 | 76.4 | 29.7 | 52.2 | 72.5 | 25 | 70.8 | 62.1 | 58.9 |