Performance Score on QA (Question Answering)
63.31Performance ScoreQAs
Evaluation Results
| Method | Links | |
|---|---|---|
| QAsCheckpoint Type=Single-domain expert, Expert Domain=QA2026.06 | 63.31 | |
| Re-MathConfiguration=Re-Math refresh from CWo2026.06 | 62.49 | |
| CGPOType=Mixed-domain baseline2026.06 | 62.48 | |
| QAoCheckpoint Type=Sequential checkpoint, Training Stage=QA2026.06 | 62.34 | |
| CWoCheckpoint Type=Sequential checkpoint, Training Stage=CW2026.06 | 62.34 | |
| JTType=Joint training2026.06 | 62.11 | |
| CodesCheckpoint Type=Single-domain expert, Expert Domain=Code2026.06 | 60.89 | |
| CodeoCheckpoint Type=Sequential checkpoint, Training Stage=Code2026.06 | 60.89 | |
| MathsCheckpoint Type=Single-domain expert, Expert Domain=Math2026.06 | 60.76 | |
| BaseCheckpoint Type=Base model2026.06 | 60.64 | |
| MathoCheckpoint Type=Sequential checkpoint, Training Stage=Math2026.06 | 60.52 | |
| CWsCheckpoint Type=Single-domain expert, Expert Domain=CW2026.06 | 60.5 |