AmbiSQL: Interactive Ambiguity Detection and Resolution for Text-to-SQL

2025-08-21

AI Summary

In text-to-SQL translation, natural language ambiguity frequently causes large language models (LLMs) to misinterpret user intent, severely degrading SQL generation accuracy. To address this, we propose an interactive ambiguity detection and resolution framework. First, we introduce the first fine-grained ambiguity taxonomy jointly modeling database element mapping and LLM inference behavior. Second, we design an LLM-based automated ambiguity identification algorithm and employ multi-turn multiple-choice prompts to elicit user clarifications on ambiguous terms. Third, we dynamically rewrite the natural language query by integrating user feedback, seamlessly embedding the process into existing text-to-SQL pipelines. Evaluated on a dedicated ambiguity benchmark, our method achieves 87.2% precision in ambiguity detection and improves end-to-end exact-match SQL accuracy by 50%, significantly enhancing system robustness and practical interactivity.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Text-to-SQL systems translate natural language questions into SQL queries, providing substantial value for non-expert users. While large language models (LLMs) show promising results for this task, they remain error-prone. Query ambiguity has been recognized as a major obstacle for LLM-based Text-to-SQL systems, leading to misinterpretation of user intent and inaccurate SQL generation. We demonstrate AmbiSQL, an interactive system that automatically detects query ambiguities and guides users through intuitive multiple-choice questions to clarify their intent. Our approach introduces a fine-grained ambiguity taxonomy for identifying ambiguities that affect database element mapping and LLM reasoning, then incorporates user feedback to rewrite ambiguous questions. Evaluation on an ambiguous query dataset shows that AmbiSQL achieves 87.2% precision in ambiguity detection and improves SQL exact match accuracy by 50% when integrated with Text-to-SQL systems. Our demonstration showcases the significant performance gains and highlights the system's practical usability. Code repo and demonstration are available at: https://github.com/JustinzjDing/AmbiSQL.

Problem

• Detecting query ambiguities in Text-to-SQL systems
• Resolving ambiguity through interactive user clarification
• Improving SQL generation accuracy via intent disambiguation

Innovation

• Interactive ambiguity detection and resolution system
• Fine-grained ambiguity taxonomy for database mapping
• User feedback integration to rewrite ambiguous questions