Mediator-Guided Multi-Agent Collaboration among Open-Source Models for Medical Decision-Making

• URL: http://arxiv.org/abs/2508.05996v2
• Date: Sat, 11 Oct 2025 07:01:05 GMT
• Title: Mediator-Guided Multi-Agent Collaboration among Open-Source Models for Medical Decision-Making
• Authors: Kaitao Chen, Mianxin Liu, Daoming Zong, Chaoyue Ding, Shaohao Rui, Yankai Jiang, Mu Zhou, Xiaosong Wang,
• Abstract summary: A blind combination of diverse vision-language models (VLMs) can amplify an erroneous outcome interpretation.<n>We propose MedOrch, a mediator-guided multi-agent collaboration framework for medical multimodal decision-making.<n>We show that the collaboration within distinct VLM-based agents can surpass the capabilities of any individual agent.
• Score: 18.640622974004724
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Complex medical decision-making involves cooperative workflows operated by different clinicians. Designing AI multi-agent systems can expedite and augment human-level clinical decision-making. Existing multi-agent researches primarily focus on language-only tasks, yet their extension to multimodal scenarios remains challenging. A blind combination of diverse vision-language models (VLMs) can amplify an erroneous outcome interpretation. VLMs in general are less capable in instruction following and importantly self-reflection, compared to large language models (LLMs) of comparable sizes. This disparity largely constrains VLMs' ability in cooperative workflows. In this study, we propose MedOrch, a mediator-guided multi-agent collaboration framework for medical multimodal decision-making. MedOrch employs an LLM-based mediator agent that enables multiple VLM-based expert agents to exchange and reflect on their outputs towards collaboration. We utilize multiple open-source general-purpose and domain-specific VLMs instead of costly GPT-series models, revealing the strength of heterogeneous models. We show that the collaboration within distinct VLM-based agents can surpass the capabilities of any individual agent. We validate our approach on five medical vision question answering benchmarks, demonstrating superior collaboration performance without model training. Our findings underscore the value of mediator-guided multi-agent collaboration in advancing medical multimodal intelligence.

Related papers

• MedSAM-Agent: Empowering Interactive Medical Image Segmentation with Multi-turn Agentic Reinforcement Learning [53.37068897861388]
  MedSAM-Agent is a framework that reformulates interactive segmentation as a multi-step autonomous decision-making process.<n>We develop a two-stage training pipeline that integrates multi-turn, end-to-end outcome verification.<n>Experiments across 6 medical modalities and 21 datasets demonstrate that MedSAM-Agent achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-02-03T09:47:49Z)
• A Knowledge-driven Adaptive Collaboration of LLMs for Enhancing Medical Decision-making [49.048767633316764]
  KAMAC is a knowledge-driven Adaptive Multi-Agent Collaboration framework.<n>It enables agents to dynamically form and expand expert teams based on the evolving diagnostic context.<n> Experiments on two real-world medical benchmarks demonstrate that KAMAC significantly outperforms both single-agent and advanced multi-agent methods.
  arXiv  Detail & Related papers  (2025-09-18T14:33:36Z)
• MAM: Modular Multi-Agent Framework for Multi-Modal Medical Diagnosis via Role-Specialized Collaboration [57.98393950821579]
  We introduce the Modular Multi-Agent Framework for Multi-Modal Medical Diagnosis (MAM)<n>Inspired by our empirical findings, MAM decomposes the medical diagnostic process into specialized roles: a General Practitioner, Specialist Team, Radiologist, Medical Assistant, and Director.<n>This modular and collaborative framework enables efficient knowledge updates and leverages existing medical LLMs and knowledge bases.
  arXiv  Detail & Related papers  (2025-06-24T17:52:43Z)
• MMedAgent-RL: Optimizing Multi-Agent Collaboration for Multimodal Medical Reasoning [63.63542462400175]
  We propose MMedAgent-RL, a reinforcement learning-based multi-agent framework that enables dynamic, optimized collaboration among medical agents.<n> Specifically, we train two GP agents based on Qwen2.5-VL via RL: the triage doctor learns to assign patients to appropriate specialties, while the attending physician integrates the judgments from multi-specialists.<n>Experiments on five medical VQA benchmarks demonstrate that MMedAgent-RL not only outperforms both open-source and proprietary Med-LVLMs, but also exhibits human-like reasoning patterns.
  arXiv  Detail & Related papers  (2025-05-31T13:22:55Z)
• MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks [17.567786780266353]
  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)
• MIND: Modality-Informed Knowledge Distillation Framework for Multimodal Clinical Prediction Tasks [50.98856172702256]
  We propose the Modality-INformed knowledge Distillation (MIND) framework, a multimodal model compression approach.<n>MIND transfers knowledge from ensembles of pre-trained deep neural networks of varying sizes into a smaller multimodal student.<n>We evaluate MIND on binary and multilabel clinical prediction tasks using time series data and chest X-ray images.
  arXiv  Detail & Related papers  (2025-02-03T08:50:00Z)
• Inquire, Interact, and Integrate: A Proactive Agent Collaborative Framework for Zero-Shot Multimodal Medical Reasoning [21.562034852024272]
  The adoption of large language models (LLMs) in healthcare has attracted significant research interest. Most state-of-the-art LLMs are unimodal, text-only models that cannot directly process multimodal inputs. We propose a multimodal medical collaborative reasoning framework textbfMultiMedRes to solve medical multimodal reasoning problems.
  arXiv  Detail & Related papers  (2024-05-19T18:26:11Z)
• MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making [45.74980058831342]
  We introduce a novel multi-agent framework, named Medical Decision-making Agents (MDAgents) The assigned solo or group collaboration structure is tailored to the medical task at hand, emulating real-world medical decision-making processes. MDAgents achieved the best performance in seven out of ten benchmarks on tasks requiring an understanding of medical knowledge.
  arXiv  Detail & Related papers  (2024-04-22T06:30:05Z)
• Med-MoE: Mixture of Domain-Specific Experts for Lightweight Medical Vision-Language Models [17.643421997037514]
  We propose a novel framework that tackles both discriminative and generative multimodal medical tasks. The learning of Med-MoE consists of three steps: multimodal medical alignment, instruction tuning and routing, and domain-specific MoE tuning. Our model can achieve performance superior to or on par with state-of-the-art baselines.
  arXiv  Detail & Related papers  (2024-04-16T02:35:17Z)
• Large Multimodal Agents: A Survey [78.81459893884737]
  Large language models (LLMs) have achieved superior performance in powering text-based AI agents. There is an emerging research trend focused on extending these LLM-powered AI agents into the multimodal domain. This review aims to provide valuable insights and guidelines for future research in this rapidly evolving field.
  arXiv  Detail & Related papers  (2024-02-23T06:04:23Z)
• Building Cooperative Embodied Agents Modularly with Large Language Models [104.57849816689559]
  We address challenging multi-agent cooperation problems with decentralized control, raw sensory observations, costly communication, and multi-objective tasks instantiated in various embodied environments. We harness the commonsense knowledge, reasoning ability, language comprehension, and text generation prowess of LLMs and seamlessly incorporate them into a cognitive-inspired modular framework. Our experiments on C-WAH and TDW-MAT demonstrate that CoELA driven by GPT-4 can surpass strong planning-based methods and exhibit emergent effective communication.
  arXiv  Detail & Related papers  (2023-07-05T17:59:27Z)

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.