MedGPT-oss: 为生物医学训练一个通用的视觉-语言模型
MedGPT-oss: Training a General-Purpose Vision-Language Model for Biomedicine
AI总结:
MEDGPT-oss是一款开放式权重的200亿参数视觉-语言模型,通过优化训练课程实现多模态和文本推理能力,适用于隐私保护的临床AI研究。
摘要:
生物医学多模态助手有望统一放射学、病理学和临床文本推理,但仍然存在关键的部署缺口:表现最佳的系统要么是闭源的,要么计算上不可行,这阻止了在患者隐私和PHI合规性要求下所需的本地部署。我们介绍了MEDGPT-OSS,一个开放式权重、200亿参数的通用视觉-语言模型,旨在促进临床AI的开放研究。而不是依赖架构复杂性,MEDGPT-OSS通过优化的三阶段训练课程将GPT-oss语言骨干与视觉前端配对。通过严格的数据整理和长上下文多模态对齐,逐步域适应这些模块,我们证明一个200亿参数的模型可以弥补能力缺口。它在分布外(OOD)多模态推理和复杂的纯文本临床任务上均优于更大的开放医疗模型。通过在单一指令遵循接口下统一各种模态,MEDGPT-OSS保持了参数高效的足迹,完全兼容商用GPU。我们发布了完整的训练配方、开放式权重检查点和严谨的评估工具,以作为隐私保护、机构特定临床AI研究的可验证基础。
英文摘要:
Biomedical multimodal assistants have the potential to unify radiology, pathology, and clinical-text reasoning, yet a critical deployment gap remains: top-performing systems are either closed-source or computationally prohibitive, precluding the on-premises deployment required for patient privacy and PHI compliance. We introduce MEDGPT-OSS, an open-weight, 20B-parameter generalist vision-language model designed to facilitate open research in clinical AI. Rather than relying on architectural complexity, MEDGPT-OSS pairs the GPT-oss language backbone with a visual front-end via a optimized, three-stage training curriculum. By progressively domain-adapting these modules through rigorous data curation and long-context multimodal alignment, we demonstrate that a 20B model can bridge the capacity gap. It successfully outperforms larger open medical models on out-of-distribution (OOD) multimodal reasoning and complex text-only clinical tasks. By unifying diverse modalities under a single instruction-following interface, MEDGPT-OSS maintains a parameter-efficient footprint fully compatible with commodity GPUs. We release the complete training recipe, open-weight checkpoints, and a rigorous evaluation harness to serve as a verifiable foundation for privacy-preserving, institution-specific clinical AI research.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。