UniSAr: a unified structure-aware autoregressive language model for text-to-SQL semantic parsing

Existing text-to-SQL semantic parsers are typically designed for particular settings such as handling queries that span multiple tables, domains, or turns which makes them ineffective when applied to different settings. We present UniSAr ( Uni fied S tructure- A ware Auto r egressive Language Model), which benefits from directly using an off-the-shelf language model architecture and demonstrates consistently high performance under different settings. Specifically, UniSAr extends existing autoregressive language models to incorporate two non-invasive extensions to make them structure-aware : (1) adding structure mark to encode database schema, conversation context, and their relationships; (2) constrained decoding to decode well-structured SQL for a given database schema. On seven well-known text-to-SQL datasets covering multi-domain, multi-table, and multi-turn, UniSAr demonstrates highly comparable or better performance to the most advanced specifically-designed text-to-SQL models.