Conversational Text-to-SQL: An Odyssey into State-of-the-Art and Challenges Ahead

TL;DR

This paper advances conversational text-to-SQL systems by integrating multi-task learning and reranking techniques with large language models, achieving improved accuracy but highlighting ongoing challenges in context-dependent and compositional generalization.

Contribution

It introduces a combined approach of multi-task training and reranking with large models, improving state-of-the-art accuracy in conversational text-to-SQL tasks.

Findings

1. Achieved 1.0% absolute accuracy improvement in exact match.

2. Improved execution match by 3.4%.

3. Identified challenges in domain and compositional generalization.

Abstract

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.