Text-to-SQL on WikiSQL Weakly-supervised (test)
84.7AccuracyGRAPPA (MLM+SSP)
Evaluation results
| Method | Links | |
|---|---|---|
| GRAPPA (MLM+SSP)Base Model=Wang et al. (2019), Pre-training Objective=MLM+SSP2020.09 | 84.7 | |
| Min et al. (2019)2020.09 | 83.9 | |
| GRAPPA (SSP)Base Model=Wang et al. (2019), Pre-training Objective=SSP2020.09 | 83.7 | |
| Herzig et al. (2020b)2020.09 | 83.6 | |
| GRAPPA (MLM)Base Model=Wang et al. (2019), Pre-training Objective=MLM2020.09 | 83.5 | |
| RoBERTa-largeBase Model=Wang et al. (2019)2020.09 | 82.3 | |
| Wang et al. (2019)2020.09 | 79.3 | |
| Agarwal et al. (2019)2020.09 | 74.8 | |
| Liang et al. (2018)2020.09 | 72.1 |