The Geometry of Queries: Query-Based Innovations in Retrieval-Augmented Generation

2024-07-25

arXiv.org

AI Summary

To address inaccurate semantic matching between user queries and knowledge bases, inefficient retrieval, and frequent hallucinations in medical RAG systems, this paper proposes QB-RAG—a theoretical framework featuring a novel pre-generated query database that maps structured medical knowledge to semantic query vectors. QB-RAG integrates LLM-driven query augmentation, efficient approximate nearest neighbor (ANN) retrieval, and geometric alignment modeling to achieve precise semantic matching between user questions and domain-specific knowledge. Evaluated on chronic disease management question answering, QB-RAG improves answer accuracy by +12.7%, reduces hallucination rate by 43.5%, and achieves clinically acceptable response reliability. The framework establishes an interpretable and verifiable paradigm for trustworthy medical large language models, advancing the deployment of reliable, evidence-grounded AI in clinical decision support.

Technology Category

• Data Mining & Knowledge Management: Intelligent Query Processing
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Natural Language Processing: Question Answering

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Digital health chatbots powered by Large Language Models (LLMs) have the potential to significantly improve personal health management for chronic conditions by providing accessible and on-demand health coaching and question-answering. However, these chatbots risk providing unverified and inaccurate information because LLMs generate responses based on patterns learned from diverse internet data. Retrieval Augmented Generation (RAG) can help mitigate hallucinations and inaccuracies in LLM responses by grounding it on reliable content. However, efficiently and accurately retrieving most relevant set of content for real-time user questions remains a challenge. In this work, we introduce Query-Based Retrieval Augmented Generation (QB-RAG), a novel approach that pre-computes a database of potential queries from a content base using LLMs. For an incoming patient question, QB-RAG efficiently matches it against this pre-generated query database using vector search, improving alignment between user questions and the content. We establish a theoretical foundation for QB-RAG and provide a comparative analysis of existing retrieval enhancement techniques for RAG systems. Finally, our empirical evaluation demonstrates that QB-RAG significantly improves the accuracy of healthcare question answering, paving the way for robust and trustworthy LLM applications in digital health.

Problem

• Enhancing healthcare QA accuracy via query-aligned retrieval-augmented generation
• Filtering irrelevant questions to ensure reliable healthcare query responses
• Evaluating retrieval and response quality for trustworthy health applications

Innovation

• Query-Based Retrieval Augmented Generation framework
• LLM-based filtering for relevant questions
• Comprehensive evaluation framework for QA