RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers
This addresses the problem of database querying for users by improving generalization in semantic parsing, though it is an incremental advancement over existing methods.
The paper tackles the challenge of generalizing text-to-SQL parsers to unseen database schemas by introducing a unified framework for schema encoding and linking, achieving a state-of-the-art exact match accuracy of 65.6% on the Spider dataset.
When translating natural language questions into SQL queries to answer questions from a database, contemporary semantic parsing models struggle to generalize to unseen database schemas. The generalization challenge lies in (a) encoding the database relations in an accessible way for the semantic parser, and (b) modeling alignment between database columns and their mentions in a given query. We present a unified framework, based on the relation-aware self-attention mechanism, to address schema encoding, schema linking, and feature representation within a text-to-SQL encoder. On the challenging Spider dataset this framework boosts the exact match accuracy to 57.2%, surpassing its best counterparts by 8.7% absolute improvement. Further augmented with BERT, it achieves the new state-of-the-art performance of 65.6% on the Spider leaderboard. In addition, we observe qualitative improvements in the model's understanding of schema linking and alignment. Our implementation will be open-sourced at https://github.com/Microsoft/rat-sql.