[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84155-en":3,"doc-seo-84155-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84155,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Vision as Unified Multimodal Generation","Vision as Unified Multimodal Generation formulates computer vision as unified multimodal generation, expressing diverse visual tasks through a single unified multimodal model’s native text and image generation spaces rather than task-specific architectures. Natural-language instructions and optional visual prompts define task, regions/views, and decoding. SenseNova-Vision converts heterogeneous vision annotations into instruction-response pairs for large-scale training and produces text, dense image outputs, or mixed responses. The resulting model covers detection, OCR, keypoints, segmentation, depth, surface normals, point maps, and camera pose.","arXiv :2607 .06560v 1 [ cs .CV] 7 Jul 2026  \nVision as Unified Multimodal Generation  \nXiaoyang Han∗ , 1 , Jianhua Li∗ , 1 , Kewang Deng∗ , 1 , Zukai Chen∗ , 1 , Xuanke Shi∗ , 1 , Sihan Wang∗ , 1 Boxuan Li∗ , 1 , Linyan Wang∗ , 1 , Siyi Xie 1 ,4 , Xin You 1 ,5 , Jinsheng Quan 1 ,6 , Zhongang Cai 1 Haiwen Diao2 , Ziwei LiuB,2 , Lei YangB, 1 , Dahua LinB, 1 ,3 , Quan Wang∗ ,B, 1  \n∗ Core Contributors, B Corresponding Authors  \n1 SenseTime Research, 2Nanyang Technological University, 3The Chinese University of Hong Kong  \n4Peking University, 5 Shanghai Jiao Tong University, 6Zhejiang University  \nAbstract  \nWe formulate computer vision as unified multimodal generation, where heterogeneous visual tasks are expressed through the native text and image generation spaces of a unified multimodal model (UMM), without task-specific architectures. With this formulation, the single model SenseNova-Vision matches leading task-specialized systems across structured visual understanding, dense geometric prediction, segmentation, and multi-view visual geometry. Natural-language instructions and optional visual prompts specify the task, target regions or views, and decoding convention. Responses are then generated as text for symbolic records, images for dense spatial targets, or mixed outputs for compositional tasks. To enable large-scale training, we convert heterogeneous computer vision annotations into instruction-response examples compatible with these native generation spaces. This conversion yields the SenseNova-Vision Corpus, a computer-vision instruction-response corpus spanning text, image, and mixed text-and-image targets. Starting from an off-the-shelf pretrained UMM, SenseNova-Vision is trained primarily on the SenseNova-Vision Corpus, using auxiliary multimodal data as a capability-preserving mixture and requiring no task-specific prediction heads or architectural changes. The resulting model covers detection, OCR, keypoints, segmentation, depth, surface normals, point maps, and camera pose estimation, and can follow language-defined variants that combine category, color, region, and other visual cues. These results suggest unified multimodal generation as a scalable route for integrating computer vision into general-purpose foundation models. The SenseNova-Vision model and SenseNova-Vision Corpus are publicly available.  \nCodebase: [https://github.com/OpenSenseNova/SenseNova-Vision](https://github.com/OpenSenseNova/SenseNova-Vision)  \nModel and Dataset: [https://huggingface.co/collections/sensenova/sensenova-vision](https://huggingface.co/collections/sensenova/sensenova-vision)  \nSenseNova-Vision Rex-Omni LocateAnything Lotus-2 MoGe-2 LISA X-SAM G2VLM VGGT  \nFigure 1 Benchmark overview across representative computer vision task families. Despite using a single unified multimodal generation interface and no task-specific heads, SenseNova-Vision achieves competitive performance across heterogeneous output formats, including text-serialized records, dense image outputs, mixed text-mask responses, and multi-view geometric predictions.  \nFigure 2 SenseNova-Vision integrates diverse computer vision tasks into a single UMM, producing outputs for structured visual understanding, dense geometric prediction, segmentation, and multi-view visual geometry through unified multimodal generation.  \n1 Introduction  \nLarge language models [12] have shown that diverse language tasks can be consolidated through prompting and generation, and unified multimodal models (UMMs) [18, 33, 35, 184, 191] further extend this paradigm to both text and image generation. We consider whether the entire spectrum of classical computer vision—from detection to multi-view 3D—can be expressed as unified multimodal generation within a single UMM, without task-specific heads. Computer vision has made remarkable progress through specialist systems [17, 59, 86, 160, 171, 197] tailored to individual task families. Their outputs range from boxes and masks to motion fields, d","cbCaivH5IcpgoLAr","https://ap.wps.com/l/cbCaivH5IcpgoLAr","pdf",38228895,1,48,"English","en",105,"# Abstract\n# Introduction","[{\"question\":\"How does SenseNova-Vision unify different computer vision tasks?\",\"answer\":\"It formulates classical vision tasks as unified multimodal generation, using the model’s native text and image generation spaces instead of task-specific heads or architectures.\"},{\"question\":\"What do natural-language instructions and visual prompts control?\",\"answer\":\"They specify the task, target regions or views, and the decoding convention so the model can generate the appropriate output format and schema.\"},{\"question\":\"What kinds of outputs can the unified model produce?\",\"answer\":\"It can generate text for symbolic records, dense image outputs for spatial targets, or mixed text-and-image responses for compositional tasks, covering detection, OCR, segmentation, depth, normals, point maps, and camera pose.\"}]",1784193489,121,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"vision-as-unified-multimodal-generation","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/vision-as-unified-multimodal-generation/84155/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"How does SenseNova-Vision unify different computer vision tasks?","Question",{"text":74,"@type":75},"It formulates classical vision tasks as unified multimodal generation, using the model’s native text and image generation spaces instead of task-specific heads or architectures.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What do natural-language instructions and visual prompts control?",{"text":79,"@type":75},"They specify the task, target regions or views, and the decoding convention so the model can generate the appropriate output format and schema.",{"name":81,"@type":72,"acceptedAnswer":82},"What kinds of outputs can the unified model produce?",{"text":83,"@type":75},"It can generate text for symbolic records, dense image outputs for spatial targets, or mixed text-and-image responses for compositional tasks, covering detection, OCR, segmentation, depth, normals, point maps, and camera pose.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]