$R^3$: "This is My SQL, Are You With Me?" A Consensus-Based Multi-Agent System for Text-to-SQL Tasks
Abstract: Large Language Models (LLMs) have demonstrated strong performance on various tasks.
To unleash their power on the Text-to-SQL task, we propose $R^3$ (Review-Rebuttal-Revision), a consensus-based multi-agent system for Text-to-SQL tasks.
$R^3$ outperforms the existing single LLM Text-to-SQL systems and as well as the multi-agent Text-to-SQL systems by $1.3\%$ to $8.1\%$ on Spider and Bird.
Surprisingly, we find that for Llama-3-8B, $R^3$ outperforms chain-of-thought prompting by over 20\%,
even outperforming GPT-3.5 on the development set of Spider.
Paper Type: Short
Research Area: Generation
Research Area Keywords: Text-to-SQL,Multi-Agent System
Contribution Types: NLP engineering experiment
Languages Studied: English
Submission Number: 3738
Loading