| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Question Answering | 5 QA tasks | Accuracy54.02 | 78 | |
| Question Answering | 7 QA tasks | Accuracy69.44 | 42 | |
| Question Answering | QA Tasks (unseen) | AN' Score42.12 | 14 | |
| Question Answering | QA Tasks seen (test) | AN59.23 | 14 | |
| Question Answering | QA Tasks (OpenBookQA, PIQA, ARC-E, ARC-C, SciQ, WebQs) (test) | OpenBookQA0.394 | 12 | |
| Question Answering | Seen QA Tasks | SQuAD 279.99 | 8 |