Conversational text-to-SQL on SParC (dev)
64.1Question MatchRAT-SQL-TC
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| RAT-SQL-TCBackbone=GAP2021.12 | 64.1 | 44.1 | — | |
| RAT-SQL + ScoreBackbone=BERT2021.12 | 62.5 | 42.5 | — | |
| RAT-SQLBackbone=GAP2021.12 | 59.6 | 40.5 | — | |
| RAT-SQLBackbone=BERT2021.12 | 56.8 | 33.4 | — | |
| R2SQLBackbone=BERT2021.12 | 54.1 | 35.2 | — | |
| IGSQLBackbone=BERT2021.12 | 50.7 | 32.5 | — | |
| EditSQLBackbone=BERT2021.12 | 47.2 | 29.5 | — | |
| Hui et al. (2021)Extra Pre-training=false2022.01 | — | — | 54.1 | |
| T5-3BBackbone=T5-3B2022.01 | — | — | 61.51 | |
| T5-baseBackbone=T5-base2022.01 | — | — | 50.54 | |
| T5-largeBackbone=T5-large2022.01 | — | — | 56.69 | |
| Yu et al. (2021)Extra Pre-training=true2022.01 | — | — | 62.2 |