Chain-of-Query: Unleashing the Power of LLMs in SQL-Aided Table Understanding via Multi-Agent Collaboration

TL;DR

Chain-of-Query (CoQ) is a multi-agent framework that improves SQL-based table understanding by using natural-language schema representations, clause-by-clause SQL generation, and hybrid reasoning to enhance accuracy and reduce invalid queries.

Contribution

Introduces CoQ, a novel multi-agent approach that enhances table understanding with natural-language schemas and a hybrid reasoning division, addressing structural and error propagation issues.

Findings

Significant accuracy improvements across benchmarks.

Lower invalid SQL rates compared to baselines.

Effective separation of mechanical and logical reasoning.

Abstract

Table understanding requires structured, multi-step reasoning. Large Language Models (LLMs) struggle with it due to the structural complexity of tabular data. Recently, multi-agent frameworks for SQL generation have shown promise in tackling the challenges of understanding tabular data, but existing approaches often suffer from limitations such as the inability to comprehend table structure for reliable SQL generation, error propagation that results in invalid queries, and over-reliance on execution correctness. To address these issues, we propose Chain-of-Query (CoQ), a novel multi-agent framework for SQL-aided table understanding. CoQ adopts natural-language-style representations of table schemas to abstract away structural noise and enhance understanding. It employs a clause-by-clause SQL generation strategy to improve query quality and introduces a hybrid reasoning division that separates SQL-based mechanical reasoning from LLM-based logical inference, thereby reducing reliance on execution outcomes. Extensive experiments across four models and five widely used benchmarks demonstrate that CoQ achieves substantial accuracy improvements and significantly lowers invalid SQL rates compared to prior generic LLM-based, SQL-aided, and hybrid baselines, confirming its superior effectiveness in table understanding. The code is available at https://github.com/SongyuanSui/ChainofQuery.