[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-273615-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-273615-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","meduag-unified-understanding-and-generation-for-medical-multimodal-models-paper","MedUAG - Unified Understanding and Generation for Medical Multimodal Models - paper","","Recent Multimodal Large Language Models (MLLMs) are evolving toward unified understanding and generation (UAG) frameworks, but medical adoption is limited by missing comprehensive training and evaluation benchmarks and the lack of a broadly validated unified medical model. MedUAG addresses these gaps by introducing MedUAGCorpus with 6M+ instances across 14 imaging modalities, MedUAGBench covering 12 generation tasks under standardized protocols, and MedUAG, an end-to-end unified medical model.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/healthcare/","Healthcare",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/meduag-unified-understanding-and-generation-for-medical-multimodal-models-paper/273615/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/meduag-unified-understanding-and-generation-for-medical-multimodal-models-paper/273615.png","ImageObject",300,407,{"name":42,"@type":43},"Asher","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-23","2026-09-15",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",8,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What main problem does MedUAG address in medical unified understanding and generation?","Question",{"text":63,"@type":64},"It targets the lack of comprehensive medical UAG training corpora and evaluation benchmarks, as well as the absence of a broadly validated unified medical model across tasks and modalities.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What is MedUAGCorpus and how large is it?",{"text":68,"@type":64},"MedUAGCorpus is the unified medical understanding-and-generation dataset, containing over 6 million instances across 14 imaging modalities and diverse task types.",{"name":70,"@type":61,"acceptedAnswer":71},"What does MedUAGBench evaluate and how is it structured?",{"text":72,"@type":64},"MedUAGBench is a systematic benchmark that expands medical generation evaluation to 12 diverse tasks under standardized prompts, metrics, and evaluation settings.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},273615,1789472283,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":108,"slug":109},7,40,"healthcare",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":111,"show_sort_weight":112,"slug":113},"Research & Report",30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":107,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":115,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":80,"read_time":143},687197207639,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","MedUAG: Unified Understanding and Generation for Medical Multimodal Models  \nZijie Meng 1,⋆, Yuncheng Zhang 1,⋆, Hualiang Wang2,⋆, Yitian Tang 1 , Xiaotang Gai 1 , Chen Shen 1 , Songtao Jiang 1 , Shaosheng Cao3 , Jian Wu 1 , Xian Wu4,†, Zuozhu Liu 1,†  \n1Zhejiang University, 2 Hong Kong University of Science and Technology,  \n3 Tsinghua University, 4 Tencent Jarvis Lab  \narXiv : 2608 . 18937v 1 [cs .CL] 19 Aug 2026  \nAbstract—Recent Multimodal Large Language Models (MLLMs) are rapidly evolving into unified understanding and generation (UAG) frameworks. However, extending these unified paradigms to the medical domain is hindered by: the absence of comprehensive training and evaluation benchmarks, and the lack of broadly validated unified medical model. To address these gaps, we present a comprehensive foundation for medical UAG. First, we construct MedUAGCorpus, the largest unified medical understanding and generation dataset to date, comprising over 6 million instances across 14 imaging modalities. Second, we introduce MedUAGBench, a systematic benchmark that expands medical generation evaluation to 12 diverse tasks under standardized protocols. Finally, leveraging these resources, we develop MedUAG, an end-to-end trained unified medical model. Extensive experiments demonstrate that MedUAG achieves strong performance across a wide array of understanding and generation tasks, establishing a competitive baseline and paving the way for next-generation medical multimodal systems.  \nIndex Terms—Medical Multimodal Learning; Unified Understanding and Generation; Benchmark and Dataset  \nI. INTRODUCTION  \nRecent Multimodal Large Language Models (MLLMs) have evolved from isolated understanding or generation systems into unified understanding and generation (UAG) frameworks across heterogeneous modalities [1], as shown in Figure 1 . Driven by the scaling and integration of data across diverse UAG tasks, these unified training paradigms have demonstrated the potential to learn universal representations within a single system, substantially reducing the reliance on taskspecific fine-tuning for downstream applications [2]–[6] .  \nHowever, the few pioneering works in medical UAG [7],[8] are hindered by two significant limitations. (1) The absence of comprehensive UAG training corpora and evaluation benchmarks. While training and evaluating MLLMs necessitate large-scale data across diverse tasks, current studies often rely on a limited scope of medical imaging modalities and clinical applications. They resort to a narrow selection of common or readily accessible tasks, such as basic textconditioned image synthesis, CT-MRI translation or MRI super resolution. Consequently, some challenging generation tasks of profound clinical significance remain overlooked. (2) The lack of a broadly validated unified medical model. Beyond data and benchmarks, current medical UAG studies  \n⋆ Equal contribution.  \n†Corresponding authors.  \nFig. 1. Comparison of different medical multimodal modeling paradigms. Our framework unifies understanding and generation within a single model.  \nhave yet to establish a unified model that is trained on largescale and diverse unified medical corpora and systematically validated across comprehensive understanding and generation benchmarks. Although recent approaches [7], [8] have shown promising initial results, their empirical validation remains limited in task breadth, modality coverage, and evaluation consistency. As a result, it remains unclear whether a single end-to-end model can robustly handle diverse medical multimodal demands, ranging from semantic understanding and clinical reasoning to structurally faithful and clinically meaningful image generation. This limitation not only leaves the generality and robustness of unified medical modeling insufficiently demonstrated, but also makes it difficult to assess its potential for practical application.  \nTo address these gaps, we introduce MedUAGBench, a comprehensive ben","cbCail9NbOpYIqGr","https://ap.wps.com/l/cbCail9NbOpYIqGr","pdf",2290112,"English","# Introduction\n## Motivation and limitations in medical UAG\n## Proposed MedUAGBench\n## Proposed MedUAGCorpus\n## End-to-end model MedUAG","[{\"question\":\"What main problem does MedUAG address in medical unified understanding and generation?\",\"answer\":\"It targets the lack of comprehensive medical UAG training corpora and evaluation benchmarks, as well as the absence of a broadly validated unified medical model across tasks and modalities.\"},{\"question\":\"What is MedUAGCorpus and how large is it?\",\"answer\":\"MedUAGCorpus is the unified medical understanding-and-generation dataset, containing over 6 million instances across 14 imaging modalities and diverse task types.\"},{\"question\":\"What does MedUAGBench evaluate and how is it structured?\",\"answer\":\"MedUAGBench is a systematic benchmark that expands medical generation evaluation to 12 diverse tasks under standardized prompts, metrics, and evaluation settings.\"}]","MedUAG - Unified Understanding and Generation for Medical Multimodal Models - paper | PDF",23]