arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.07306cs.CLcs.AI
收藏

C3:基于 ChatGPT 的零样本文本转 SQL

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

Zhejiang University(浙江大学);Hundsun Technologies INC.(恒生电子股份有限公司)

Xuemei Dong, Chao Zhang, Yuhang Ge, Yuren Mao, Yunjun Gao, lu Chen, Jinshu Lin, Dongfang Lou

AI总结:

该研究提出基于 ChatGPT 的零样本 Text-to-SQL 方法 C3,通过清晰提示、提示校准和一致输出三组件,在 Spider 上取得 82.3% 执行准确率并达到零样本最优水平。

摘要:

本文提出一种基于 ChatGPT 的零样本文本转 SQL(Text-to-SQL)方法,名为 C3。该方法在 Spider 的留出测试集上以执行准确率达到 82.3%,成为 Spider Challenge 上最先进的零样本文本转 SQL 方法。C3 由三个关键组件组成:Clear Prompting(CP,清晰提示)、Calibration with Hints(CH,带提示校准)和 Consistent Output(CO,一致输出),分别对应模型输入、模型偏差和模型输出。它为零样本文本转 SQL 提供了系统性处理方案。研究开展了大量实验,以验证所提方法的有效性和效率。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑