Yuntao Li, Zhenpeng Su, Yutian Li, Hanchu Zhang, Sirui Wang, Wei Wu, Yan Zhang
Translating natural language queries into SQLs in a seq2seq manner has attracted much attention recently. However, compared with abstract-syntactic-tree-based SQL generation, seq2seq semantic parsers face much more challenges, including poor quality on schematical information prediction and poor semantic coherence between natural language queries and SQLs. This paper analyses the above difficulties and proposes a seq2seq-oriented decoding strategy called SR, which includes a new intermediate representation SSQL and a reranking method with score re-estimator to solve the above obstacles respectively. Experimental results demonstrate the effectiveness of our proposed techniques and T5-SR-3b achieves new state-of-the-art results on the Spider dataset.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Semantic Parsing | spider | Exact Match Accuracy (Dev) | 77.2 | T5-SR |
| Semantic Parsing | spider | Exact Match Accuracy (Test) | 72.4 | T5-SR |
| Semantic Parsing | spider | Execution Accuracy (Test) | 75.2 | T5-SR |
| Text-To-SQL | spider | Exact Match Accuracy (Dev) | 77.2 | T5-SR |
| Text-To-SQL | spider | Exact Match Accuracy (Test) | 72.4 | T5-SR |
| Text-To-SQL | spider | Execution Accuracy (Test) | 75.2 | T5-SR |