C3: Zero-shot Text-to-SQL with ChatGPT

TL;DR

This paper introduces C3, a zero-shot Text-to-SQL approach using ChatGPT, achieving state-of-the-art accuracy on the Spider dataset through systematic prompting, calibration, and output consistency techniques.

Contribution

The paper presents a novel zero-shot Text-to-SQL method with a systematic framework and demonstrates its effectiveness with extensive experiments.

Findings

Achieves 82.3% execution accuracy on Spider test set.

Sets new state-of-the-art in zero-shot Text-to-SQL on Spider.

Proves effectiveness and efficiency of the proposed method.

Abstract

This paper proposes a ChatGPT-based zero-shot Text-to-SQL method, dubbed C3, which achieves 82.3\% in terms of execution accuracy on the holdout test set of Spider and becomes the state-of-the-art zero-shot Text-to-SQL method on the Spider Challenge. C3 consists of three key components: Clear Prompting (CP), Calibration with Hints (CH), and Consistent Output (CO), which are corresponding to the model input, model bias and model output respectively. It provides a systematic treatment for zero-shot Text-to-SQL. Extensive experiments have been conducted to verify the effectiveness and efficiency of our proposed method.