MAGIC: Generating Self-Correction Guideline for In-Context Text-to-SQL

TL;DR

MAGIC is an automated multi-agent system that creates self-correction guidelines for text-to-SQL tasks, outperforming human-crafted guidelines and improving interpretability of LLM corrections.

Contribution

Introduces MAGIC, a novel multi-agent framework that automates self-correction guideline generation for LLM-based text-to-SQL, reducing human effort and enhancing correction quality.

Findings

MAGIC's guidelines outperform human-crafted ones.

Guidelines improve interpretability of LLM corrections.

Automated process reduces manual effort in guideline creation.

Abstract

Self-correction in text-to-SQL is the process of prompting large language model (LLM) to revise its previously incorrectly generated SQL, and commonly relies on manually crafted self-correction guidelines by human experts that are not only labor-intensive to produce but also limited by the human ability in identifying all potential error patterns in LLM responses. We introduce MAGIC, a novel multi-agent method that automates the creation of the self-correction guideline. MAGIC uses three specialized agents: a manager, a correction, and a feedback agent. These agents collaborate on the failures of an LLM-based method on the training set to iteratively generate and refine a self-correction guideline tailored to LLM mistakes, mirroring human processes but without human involvement. Our extensive experiments show that MAGIC's guideline outperforms expert human's created ones. We empirically find out that the guideline produced by MAGIC enhances the interpretability of the corrections made, providing insights in analyzing the reason behind the failures and successes of LLMs in self-correction. All agent interactions are publicly available at https://huggingface.co/datasets/microsoft/MAGIC.