Saved in:
Bibliographic Details
Main Authors: Sui, Songyuan, Liu, Hongyi, Liu, Serena, Li, Li, Choi, Soo-Hyun, Chen, Rui, Hu, Xia
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.15809
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915646209523712
author Sui, Songyuan
Liu, Hongyi
Liu, Serena
Li, Li
Choi, Soo-Hyun
Chen, Rui
Hu, Xia
author_facet Sui, Songyuan
Liu, Hongyi
Liu, Serena
Li, Li
Choi, Soo-Hyun
Chen, Rui
Hu, Xia
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2508_15809
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Chain-of-Query: Unleashing the Power of LLMs in SQL-Aided Table Understanding via Multi-Agent Collaboration
Sui, Songyuan
Liu, Hongyi
Liu, Serena
Li, Li
Choi, Soo-Hyun
Chen, Rui
Hu, Xia
Computation and Language
Artificial Intelligence
Databases
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.
title Chain-of-Query: Unleashing the Power of LLMs in SQL-Aided Table Understanding via Multi-Agent Collaboration
topic Computation and Language
Artificial Intelligence
Databases
url https://arxiv.org/abs/2508.15809