MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains

TL;DR

MIRAGE is a scalable reasoning framework that enhances medical question-answering by executing parallel, structured inference chains over knowledge graphs, improving accuracy, traceability, and interpretability.

Contribution

It introduces a novel multi-chain inference approach over knowledge graphs, addressing error propagation and enhancing interpretability in medical reasoning tasks.

Findings

Outperforms GPT-4o, Tree-of-Thought, and other baselines in medical QA

Improves interpretability with explicit reasoning chains

Enhances accuracy and traceability in complex medical reasoning

Abstract

Large reasoning models (LRMs) have shown significant progress in test-time scaling through chain-of-thought prompting. Current approaches like search-o1 integrate retrieval augmented generation (RAG) into multi-step reasoning processes but rely on a single, linear reasoning chain while incorporating unstructured textual information in a flat, context-agnostic manner. As a result, these approaches can lead to error accumulation throughout the reasoning chain, which significantly limits its effectiveness in medical question-answering (QA) tasks where both accuracy and traceability are critical requirements. To address these challenges, we propose MIRAGE (Multi-chain Inference with Retrieval-Augmented Graph Exploration), a novel test-time scalable reasoning framework that performs dynamic multi-chain inference over structured medical knowledge graphs. Specifically, MIRAGE 1) decomposes complex queries into entity-grounded sub-questions, 2) executes parallel inference chains, 3) retrieves evidence adaptively via neighbor expansion and multi-hop traversal, and 4) integrates answers using cross-chain verification to resolve contradictions. Experiments on three medical QA benchmarks (GenMedGPT-5k, CMCQA, and ExplainCPE) show that MIRAGE consistently outperforms GPT-4o, Tree-of-Thought variants, and other retrieval-augmented baselines in both automatic and human evaluations. Additionally, MIRAGE improves interpretability by generating explicit reasoning chains that trace each factual claim to concrete chains within the knowledge graph, making it well-suited for complex medical reasoning scenarios. The code will be available for further research.