MedResearcher-R1: Expert-Level Medical Deep Researcher via A Knowledge-Informed Trajectory Synthesis Framework

• URL: http://arxiv.org/abs/2508.14880v3
• Date: Mon, 01 Sep 2025 15:33:47 GMT
• Title: MedResearcher-R1: Expert-Level Medical Deep Researcher via A Knowledge-Informed Trajectory Synthesis Framework
• Authors: Ailing Yu, Lan Yao, Jingnan Liu, Zhe Chen, Jiajun Yin, Yuan Wang, Xinhao Liao, Zhiling Ye, Ji Li, Yun Yue, Hansong Xiao, Hualei Zhou, Chunxiao Guo, Peng Wei, Junwei Liu, Jinjie Gu,
• Abstract summary: General-purpose deep research agents struggle with medical domain challenges, as evidenced by leading proprietary systems.<n>We present a medical deep research agent that addresses these challenges through two core innovations.<n>Our approach generates 2100+ diverse trajectories across 12 medical specialties, each averaging 4.2 tool interactions.
• Score: 24.399778346443757
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent developments in Large Language Model (LLM)-based agents have shown impressive capabilities spanning multiple domains, exemplified by deep research systems that demonstrate superior performance on complex information-seeking and synthesis tasks. While general-purpose deep research agents have shown impressive capabilities, they struggle significantly with medical domain challenges, as evidenced by leading proprietary systems achieving limited accuracy on complex medical benchmarks. The key limitations are: (1) the model lacks sufficient dense medical knowledge for clinical reasoning, and (2) the framework is constrained by the absence of specialized retrieval tools tailored for medical contexts. We present a medical deep research agent that addresses these challenges through two core innovations. First, we develop a novel data synthesis framework using medical knowledge graphs, extracting the longest chains from subgraphs around rare medical entities to generate complex multi-hop question-answer pairs. Second, we integrate a custom-built private medical retrieval engine alongside general-purpose tools, enabling accurate medical information synthesis. Our approach generates 2100+ diverse trajectories across 12 medical specialties, each averaging 4.2 tool interactions. Through a two-stage training paradigm combining supervised fine-tuning and online reinforcement learning with composite rewards, our MedResearcher-R1-32B model demonstrates exceptional performance, establishing new state-of-the-art results on medical benchmarks while maintaining competitive performance on general deep research tasks. Our work demonstrates that strategic domain-specific innovations in architecture, tool design, and training data construction can enable smaller open-source models to outperform much larger proprietary systems in specialized domains.

Related papers

• DR.EHR: Dense Retrieval for Electronic Health Record with Knowledge Injection and Synthetic Data [2.9929405444223205]
  EHRs are pivotal in clinical practices, yet their retrieval remains a challenge mainly due to semantic gap issues.<n>Recent advancements in dense retrieval offer promising solutions but existing models, both general-domain and biomedical-domain, fall short due to insufficient medical knowledge or mismatched training corpora.<n>This paper introduces textttDR.EHR, a series of dense retrieval models specifically tailored for EHR retrieval.
  arXiv  Detail & Related papers  (2025-07-24T17:02:46Z)
• Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning [57.873833577058]
  We build a multimodal dataset enriched with extensive medical knowledge.<n>We then introduce our medical-specialized MLLM: Lingshu.<n>Lingshu undergoes multi-stage training to embed medical expertise and enhance its task-solving capabilities.
  arXiv  Detail & Related papers  (2025-06-08T08:47:30Z)
• MEDMKG: Benchmarking Medical Knowledge Exploitation with Multimodal Knowledge Graph [28.79000907242469]
  We propose MEDMKG, a Medical Multimodal Knowledge Graph that unifies visual and textual medical information through a multi-stage construction pipeline.<n>We evaluate MEDMKG across three tasks under two experimental settings, benchmarking twenty-four baseline methods and four state-of-the-art vision-language backbones on six datasets.<n>Results show that MEDMKG not only improves performance in downstream medical tasks but also offers a strong foundation for developing adaptive and robust strategies for multimodal knowledge integration in medical artificial intelligence.
  arXiv  Detail & Related papers  (2025-05-22T18:41:46Z)
• MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks [27.717720332927296]
  We introduce MedAgentBoard, a comprehensive benchmark for the systematic evaluation of multi-agent collaboration, single-LLM, and conventional approaches.<n> MedAgentBoard encompasses four diverse medical task categories: medical (visual) question answering, lay summary generation, structured Electronic Health Record (EHR) predictive modeling, and clinical workflow automation.<n>Our extensive experiments reveal a nuanced landscape: while multi-agent collaboration demonstrates benefits in specific scenarios, it does not consistently outperform advanced single LLMs.
  arXiv  Detail & Related papers  (2025-05-18T11:28:17Z)
• Towards Artificial Intelligence Research Assistant for Expert-Involved Learning [64.7438151207189]
  Large Language Models (LLMs) and Large Multi-Modal Models (LMMs) have emerged as transformative tools in scientific research.<n>We present textbfARtificial textbfIntelligence research assistant for textbfExpert-involved textbfLearning (ARIEL)
  arXiv  Detail & Related papers  (2025-05-03T14:21:48Z)
• m-KAILIN: Knowledge-Driven Agentic Scientific Corpus Distillation Framework for Biomedical Large Language Models Training [22.996230737442254]
  Corpus Heading for biomedical large language models (LLMs) seeks to address the pressing challenge of insufficient quantity and quality in open-source scientific corpora.<n>This paper proposes a knowledge-driven, agentic framework for scientific corpus distillation, tailored explicitly for LLM training in the biomedical domain.
  arXiv  Detail & Related papers  (2025-04-28T08:18:24Z)
• Capabilities of Gemini Models in Medicine [100.60391771032887]
  We introduce Med-Gemini, a family of highly capable multimodal models specialized in medicine. We evaluate Med-Gemini on 14 medical benchmarks, establishing new state-of-the-art (SoTA) performance on 10 of them. Our results offer compelling evidence for Med-Gemini's potential, although further rigorous evaluation will be crucial before real-world deployment.
  arXiv  Detail & Related papers  (2024-04-29T04:11:28Z)
• Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicine [89.46836590149883]
  We build on a prior study of GPT-4's capabilities on medical challenge benchmarks in the absence of special training. We find that prompting innovation can unlock deeper specialist capabilities and show that GPT-4 easily tops prior leading results for medical benchmarks. With Medprompt, GPT-4 achieves state-of-the-art results on all nine of the benchmark datasets in the MultiMedQA suite.
  arXiv  Detail & Related papers  (2023-11-28T03:16:12Z)
• Towards Medical Artificial General Intelligence via Knowledge-Enhanced Multimodal Pretraining [121.89793208683625]
  Medical artificial general intelligence (MAGI) enables one foundation model to solve different medical tasks. We propose a new paradigm called Medical-knedge-enhanced mulTimOdal pretRaining (MOTOR)
  arXiv  Detail & Related papers  (2023-04-26T01:26:19Z)
• MedDG: An Entity-Centric Medical Consultation Dataset for Entity-Aware Medical Dialogue Generation [86.38736781043109]
  We build and release a large-scale high-quality Medical Dialogue dataset related to 12 types of common Gastrointestinal diseases named MedDG. We propose two kinds of medical dialogue tasks based on MedDG dataset. One is the next entity prediction and the other is the doctor response generation. Experimental results show that the pre-train language models and other baselines struggle on both tasks with poor performance in our dataset.
  arXiv  Detail & Related papers  (2020-10-15T03:34:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.