Hybrid-IR: Dual-Path Hybrid Retrieval with Iterative Reasoning for Complex Medical Question Answering

2026-06-23

AI Summary

This work addresses the limitations of existing retrieval-augmented generation methods in complex medical question answering, which struggle to jointly capture fine-grained semantics and structured knowledge while being constrained by static retrieval that hinders deep reasoning. To overcome these challenges, we propose Hybrid-IR, a dual-path hybrid retrieval framework that innovatively integrates graph-based retrieval with dense retrieval. Furthermore, we introduce an iterative retrieve-and-reason mechanism that dynamically refines the reasoning trajectory based on evolving query understanding. By moving beyond the constraints of single-path static retrieval, our approach significantly improves both answer accuracy and knowledge consistency across three mainstream medical question-answering benchmarks.

Abstract

Large language models (LLMs) have shown promising performance across a wide range of biomedical applications, including medical question answering (QA), yet they remain prone to hallucinations and outdated knowledge. Although retrieval-augmented generation (RAG) can alleviate this issue by incorporating external documents, there still exist two fundamental limitations. First, medical knowledge is often fragmented across documents, while most RAG methods rely on a single retrieval path, which makes it challenging to jointly preserve fine-grained semantic information and structured global associations. Second, static retrieval strategies are typically insufficient to support deep reasoning that is important in complex medical QA. In this paper, we present a dual-path retrieval framework with an iterative retrieval-reasoning mechanism termed "Hybrid-IR" for complex medical QA. The proposed Hybrid-IR integrates graph-based retrieval for exploration of structured knowledge and dense retrieval for fine-grained semantic matching. Moreover, the reasoning trajectory can be progressively refined through an iterative retrieve-reason loop. Experiments on three widely used medical QA benchmarks demonstrate the effectiveness of our Hybrid-IR.

Problem

• medical question answering
• retrieval-augmented generation
• knowledge fragmentation
• deep reasoning
• hybrid retrieval

Innovation

• dual-path retrieval
• graph-based retrieval
• dense retrieval
• iterative reasoning
• retrieval-augmented generation