| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Question Answering | QuaRTz | Accuracy87.8 | 11 | |
| Multi-hop Open-domain Question Answering | Quartz | Accuracy94.1 | 6 | |
| Question Answering | Quartz (test) | Accuracy90.2 | 4 | |
| Ranking correlation with full dataset evaluation | QuaRTz | Kendall Correlation0.37 | 4 |