Conversational Text-to-SQL: An Odyssey into State-of-the-Art and Challenges Ahead
This work addresses the problem of generating accurate SQL queries from multi-turn natural language dialogues for database users, representing an incremental advance in a specific domain.
The paper tackles the conversational text-to-SQL task by improving state-of-the-art systems using multi-tasking and reranking techniques, achieving absolute accuracy improvements of 1.0% in exact match and 3.4% in execution match over a baseline on CoSQL.
Conversational, multi-turn, text-to-SQL (CoSQL) tasks map natural language utterances in a dialogue to SQL queries. State-of-the-art (SOTA) systems use large, pre-trained and finetuned language models, such as the T5-family, in conjunction with constrained decoding. With multi-tasking (MT) over coherent tasks with discrete prompts during training, we improve over specialized text-to-SQL T5-family models. Based on Oracle analyses over n-best hypotheses, we apply a query plan model and a schema linking algorithm as rerankers. Combining MT and reranking, our results using T5-3B show absolute accuracy improvements of 1.0% in exact match and 3.4% in execution match over a SOTA baseline on CoSQL. While these gains consistently manifest at turn level, context dependent turns are considerably harder. We conduct studies to tease apart errors attributable to domain and compositional generalization, with the latter remaining a challenge for multi-turn conversations, especially in generating SQL with unseen parse trees.