QueryGenie: Making LLM-Based Database Querying Transparent and Controllable

2025-08-20

AI Summary

Existing LLM-based database query tools suffer from frequent user intent misinterpretation, hallucinated SQL generation, and inadequate integration of human feedback—compromising reliability and controllability. To address these issues, we propose an interactive, progressive query generation framework that integrates incremental reasoning, real-time semantic validation, and responsive human-in-the-loop interaction, enabling users to dynamically monitor, detect, and correct semantic mismatches and logical errors across multi-turn dialogues. Our core contribution lies in closing the loop among natural language understanding, program synthesis, and explainable verification—thereby achieving robust intent alignment and hallucination suppression. Experimental results demonstrate significant reductions in erroneous query rate and hallucination incidence, alongside measurable improvements in query accuracy, user trust, and operational controllability. This work establishes a novel paradigm for trustworthy, human-centered database interaction.

Technology Category

• Data Mining & Knowledge Management: Intelligent Query Processing
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Humans and AI: Interaction Techniques and Devices

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Conversational user interfaces powered by large language models (LLMs) have significantly lowered the technical barriers to database querying. However, existing tools still encounter several challenges, such as misinterpretation of user intent, generation of hallucinated content, and the absence of effective mechanisms for human feedback-all of which undermine their reliability and practical utility. To address these issues and promote a more transparent and controllable querying experience, we proposed QueryGenie, an interactive system that enables users to monitor, understand, and guide the LLM-driven query generation process. Through incremental reasoning, real-time validation, and responsive interaction mechanisms, users can iteratively refine query logic and ensure alignment with their intent.

Problem

• Addressing misinterpretation of user intent in LLM-based querying
• Reducing hallucinated content generation during database interactions
• Implementing effective human feedback mechanisms for query refinement

Innovation

• Incremental reasoning for query refinement
• Real-time validation to prevent errors
• Responsive interaction for user control