SQL-of-Thought: Multi-agentic Text-to-SQL with Guided Error Correction

2025-08-30

AI Summary

This work addresses the challenge of accurate natural language-to-SQL (NL2SQL) translation. We propose a multi-agent collaborative framework that decomposes the task into five sequential, interdependent stages: schema linking, subproblem identification, reasoning-driven query planning, SQL generation, and classification-guided dynamic error correction. Our key innovation is a semantic error category–aware dynamic correction mechanism that integrates in-context learning and chain-of-thought reasoning—departing from conventional static correction paradigms reliant solely on execution feedback. Evaluated on the Spider benchmark and its variants, our approach achieves state-of-the-art performance, particularly excelling on complex cross-domain queries. It demonstrates substantial improvements in both accuracy and robustness, empirically validating the effectiveness of classification-aware error correction and multi-stage collaborative reasoning for NL2SQL.

Technology Category

• Natural Language Processing: Question Answering
• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Converting natural language queries into SQL queries is a crucial challenge in both industry and academia, aiming to increase access to databases and large-scale applications. This work examines how in-context learning and chain-of-thought can be utilized to develop a robust solution for text-to-SQL systems. We propose SQL-of-Thought: a multi-agent framework that decomposes the Text2SQL task into schema linking, subproblem identification, query plan generation, SQL generation, and a guided correction loop. Unlike prior systems that rely only on execution-based static correction, we introduce taxonomy-guided dynamic error modification informed by in-context learning. SQL-of-Thought achieves state-of-the-art results on the Spider dataset and its variants, combining guided error taxonomy with reasoning-based query planning.

Problem

• Improving natural language to SQL query conversion accuracy
• Developing dynamic error correction for text-to-SQL systems
• Enhancing database accessibility through multi-agent framework

Innovation

• Multi-agent framework decomposing Text2SQL tasks
• Taxonomy-guided dynamic error modification via learning
• Combining error taxonomy with reasoning-based planning