Faster and Better Grammar-based Text-to-SQL Parsing via Clause-level Parallel Decoding and Alignment Loss
This work addresses efficiency and performance issues in cross-domain text-to-SQL parsing, which is incremental as it enhances existing high-performance parsers.
The paper tackled the low decoding efficiency and alignment challenges in grammar-based text-to-SQL parsing by proposing clause-level parallel decoding and alignment loss, resulting in consistent improvements in accuracy and decoding speed for RATSQL and LGESQL parsers.
Grammar-based parsers have achieved high performance in the cross-domain text-to-SQL parsing task, but suffer from low decoding efficiency due to the much larger number of actions for grammar selection than that of tokens in SQL queries. Meanwhile, how to better align SQL clauses and question segments has been a key challenge for parsing performance. Therefore, this paper proposes clause-level parallel decoding and alignment loss to enhance two high-performance grammar-based parsers, i.e., RATSQL and LGESQL. Experimental results of two parsers show that our method obtains consistent improvements both in accuracy and decoding speed.