MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQL

TL;DR

MAC-SQL introduces a multi-agent framework utilizing external tools and collaborative reasoning to significantly improve Text-to-SQL performance on complex databases, achieving state-of-the-art results.

Contribution

The paper presents MAC-SQL, a novel multi-agent framework that enhances Text-to-SQL tasks by integrating external tools and collaborative reasoning, with a fine-tuned open-source model rivaling GPT-4.

Findings

SQL-Llama achieves 43.94% accuracy, close to GPT-4's 46.35%.

MAC-SQL+GPT-4 achieves 59.59% accuracy on BIRD benchmark, setting new SOTA.

The framework effectively handles complex databases and multi-step reasoning.

Abstract

Recent LLM-based Text-to-SQL methods usually suffer from significant performance degradation on "huge" databases and complex user questions that require multi-step reasoning. Moreover, most existing methods neglect the crucial significance of LLMs utilizing external tools and model collaboration. To address these challenges, we introduce MAC-SQL, a novel LLM-based multi-agent collaborative framework. Our framework comprises a core decomposer agent for Text-to-SQL generation with few-shot chain-of-thought reasoning, accompanied by two auxiliary agents that utilize external tools or models to acquire smaller sub-databases and refine erroneous SQL queries. The decomposer agent collaborates with auxiliary agents, which are activated as needed and can be expanded to accommodate new features or tools for effective Text-to-SQL parsing. In our framework, We initially leverage GPT-4 as the strong backbone LLM for all agent tasks to determine the upper bound of our framework. We then fine-tune an open-sourced instruction-followed model, SQL-Llama, by leveraging Code Llama 7B, to accomplish all tasks as GPT-4 does. Experiments show that SQL-Llama achieves a comparable execution accuracy of 43.94, compared to the baseline accuracy of 46.35 for vanilla GPT-4. At the time of writing, MAC-SQL+GPT-4 achieves an execution accuracy of 59.59 when evaluated on the BIRD benchmark, establishing a new state-of-the-art (SOTA) on its holdout test set (https://github.com/wbbeyourself/MAC-SQL).