RareDxR1: Autonomous Medical Reasoning for Rare Disease Diagnosis Beyond Human Annotation

2026-06-30

AI Summary

This work addresses the limitations in rare disease differential diagnosis—namely, information loss and insufficient reasoning capabilities stemming from reliance on structured phenotypes, retrieval bottlenecks, and absent diagnostic logic—by proposing RareDxR1, an end-to-end large language model capable of performing open-domain diagnosis directly from unstructured clinical narratives. Through synergistic training that combines knowledge internalization with autonomous evolutionary learning, the model eliminates dependence on manual annotations and closed decision sets. It innovatively employs a novel, annotation-free Reflection-Enhanced Reasoning Sampling (RERS) strategy to synthesize expert-level diagnostic trajectories and integrates a two-tier curriculum-based reinforcement learning framework to embed fragmented rare disease knowledge into its parameters. Experiments demonstrate that this approach achieves state-of-the-art accuracy across multiple benchmarks, substantially advancing the automation and intelligence of open-domain rare disease diagnosis.

Abstract

Rare disease differential diagnosis is a critical yet arduous clinical task, requiring physicians to identify precise phenotypes from complex, unstructured patient symptoms and execute intricate reasoning within a vast search space. However, existing AI approaches typically rely on pipeline-based phenotype extraction or retrieval-augmented generation, which suffer from critical information loss due to predefined ontologies, retrieval bottlenecks, and a lack of diagnostic logic. To address these challenges, we introduce RareDxR1, an end-to-end reasoning-centric large language model designed for open-domain rare disease diagnosis directly from unstructured clinical notes. We design a progressive end-to-end training framework by synergizing knowledge internalization with autonomous evolutionary learning, thereby bypassing reliance on structured phenotypes and closed-set decision-making. To overcome the limitations of RAG and phenotype restriction, we enabled the deep internalization of fragmented rare-disease knowledge directly into the model's parameters. Moreover, to bridge the gap between model generation and expert reasoning, we propose Reflection-Enhanced Reasoning Sampling (RERS), a strategy that synthesizes expert-level diagnostic trajectories by learning from failures without human annotation. Additionally, we propose a dual-level curriculum reinforcement learning approach for gradually mastering rare disease diagnosis. Experimental results demonstrate that RareDxR1 achieves state-of-the-art accuracy across different benchmarks, marking a significant breakthrough in open-domain rare disease diagnosis. Our code and dataset will be publicly available.

Problem

• rare disease diagnosis
• differential diagnosis
• unstructured clinical notes
• medical reasoning
• phenotype extraction

Innovation

• end-to-end reasoning
• knowledge internalization
• Reflection-Enhanced Reasoning Sampling
• autonomous evolutionary learning
• curriculum reinforcement learning