Numerical Reasoning on HotpotQA (test)
68.7EMPOET-SQL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| POET-SQLBackbone=RoBERTa-Large2022.01 | 68.7 | 81.6 | |
| RoBERTa-LargeBackbone=RoBERTa-Large2022.01 | 67.6 | 81.1 | |
| POET-SQLBackbone=BART-Large2022.01 | 66.5 | 79.7 | |
| BART-LargeBackbone=BART-Large2022.01 | 65.6 | 78.9 |