SeSQL: Yet Another Large-scale Session-level Chinese Text-to-SQL Dataset

TL;DR

SeSQL is a large-scale, high-quality Chinese session-level text-to-SQL dataset designed to improve session-based parsing, featuring extensive manual annotation and comprehensive benchmarking of state-of-the-art models.

Contribution

This work introduces SeSQL, the first large-scale Chinese session-level text-to-SQL dataset with manual construction and quality control, enabling advanced research in session-based parsing.

Findings

SeSQL contains 5,028 sessions with 27,012 question/SQL pairs.

Benchmark experiments reveal the performance of three state-of-the-art session-level parsers.

The dataset supports single-round multi-DB text-to-SQL parsing.

Abstract

As the first session-level Chinese dataset, CHASE contains two separate parts, i.e., 2,003 sessions manually constructed from scratch (CHASE-C), and 3,456 sessions translated from English SParC (CHASE-T). We find the two parts are highly discrepant and incompatible as training and evaluation data. In this work, we present SeSQL, yet another large-scale session-level text-to-SQL dataset in Chinese, consisting of 5,028 sessions all manually constructed from scratch. In order to guarantee data quality, we adopt an iterative annotation workflow to facilitate intense and in-time review of previous-round natural language (NL) questions and SQL queries. Moreover, by completing all context-dependent NL questions, we obtain 27,012 context-independent question/SQL pairs, allowing SeSQL to be used as the largest dataset for single-round multi-DB text-to-SQL parsing. We conduct benchmark session-level text-to-SQL parsing experiments on SeSQL by employing three competitive session-level parsers, and present detailed analysis.