[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85114-en":3,"doc-seo-85114-105":30,"detail-sidebar-cat-0-en-105":91},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},85114,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","DexVerse: A Modular Benchmark for Multi-Task, Multi-Embodiment Dexterous Manipulation","DexVerse addresses the need for general-purpose dexterous manipulation benchmarks that evaluate policies across diverse interaction modes, sensory conditions, and robot embodiments. It introduces a large-scale modular benchmark suite with 100 dexterous manipulation tasks, covering grasping and relocation, articulated-object interaction, tool use, bimanual coordination, nonprehensile control, contact-rich behaviors, and long-horizon multi-stage execution. DexVerse supports multiple robot arms and dexterous hands, configurable visual variations, and 3,180 VR teleoperation demonstrations with synchronized multi-modal observations, enabling robust visuomotor generalization testing.","arXiv :2607 .0875 1v 1 [ cs .RO] 9 Jul 2026  \nDexVerse: A Modular Benchmark for Multi-Task, Multi-Embodiment Dexterous Manipulation  \nYunchao Yao 1∗ Zhuxiu Xu 1 ,2∗ Tianqi Zhang 1 Zixian Liu 1 Sikai Li 1 Zhenyu Wei 1 Feng Chen2 Dihong Huang 1 Kechang Wan 1 Chenyang Ma 1 Shuqi Zhao3 Shenghua Gao2 Masayoshi Tomizuka3 Yi Ma2 Mingyu Ding 1†  \n1UNC-Chapel Hill 2The University of Hong Kong 3UC Berkeley  \n∗Equal contribution. †Corresponding author  \nFigure 1: Overview of DexVerse, a modular benchmark for multi-task, multi-embodiment dexterous manipulation with diverse tasks, visual variations, demonstration datasets, and baseline evaluations.  \nAbstract: Building general-purpose dexterous manipulation policies requires benchmarks that go beyond isolated tasks to systematically evaluate policies across diverse interaction modes, sensory conditions, and robot embodiments.  \nHowever, existing benchmarks remain limited in task and data diversity, embodiment coverage, or controllable visual variation, hindering studies of cross-task and cross-embodiment generalization. We present DexVerse, a large-scale and modular benchmark for dexterous manipulation. DexVerse includes 100 tasks spanning a broad range of manipulation skills, including object grasping and relocation, articulated-object interaction, functional tool use, bimanual coordination, nonprehensile control, contact-rich behaviors, multi-goal execution, and long-horizon multi-stage task completion. It supports 3 robot arms and 6 dexterous hands, and is extensible to new tasks, assets, and embodiments. To evaluate visuomotor generalization, DexVerse provides configurable visual variations in textures, background, lighting, and camera viewpoints. We further provide a VR-based teleoperation interface and 3,180 demonstrations with synchronized proprioceptive, RGB, depth, point-cloud, and state observations. We benchmark representative methods, including Diffusion Policy, DP3, OpenVLA, and π0.5 , across 19 tasks. Results  \nreveal substantial challenges in task generalization and visuomotor robustness, establishing DexVerse as a promising testbed for general-purpose dexterous manipulation. Project page: [https://ycyao216.github.io/DexVerse.site/](https://ycyao216.github.io/DexVerse.site/)  \nKeywords: Dexterous Manipulation, Benchmark Suite, Diverse Tasks  \n1 Introduction  \nDexterous manipulation is a central capability for building general-purpose robots [1, 2] . Moving beyond contact-poor, gripper-based manipulation that can often be approximated by reaching, grasping, transporting, and releasing, dexterous manipulation requires coordinated control of highDoF hands and arms under intermittent, contact-rich interactions, while grounding actions in object geometry, visual affordances, force closure, and long-horizon task structure [3, 4] . Recent generalist robot policies have advanced this frontier from several complementary directions: imageconditioned imitation learning, such as action chunking and diffusion policies [5, 6, 7]; scalable demonstration and data generation pipelines [8, 9, 10, 11]; 3D-aware policies that exploit voxel, multi-view, or point-cloud representations [12, 13, 14, 15]; and vision-language-action policies trained on increasingly heterogeneous robot data [16, 17, 18, 19, 20, 21] . Yet it remains unclear how well these methods scale from isolated skills and controlled task distributions to functionally diverse, long-horizon, contact-rich dexterous manipulation across embodiments and environments.  \nA key bottleneck is the lack of benchmarks that jointly evaluate the major axes of dexterous generalization. Existing long-horizon manipulation benchmarks such as CALVIN [22], RoboTwin 2.0 [23], ManiSkill3 [24], and LIBERO [25] primarily focus on gripper-based manipulation, while dexterous benchmarks often specialize in narrower settings without expert demonstrations. DexMimicGen [10] focuses on dexterous demonstration generation, Bi-DexHands [26] emphasizes RL-based bima","cbCaieaZCJjxIprl","https://ap.wps.com/l/cbCaieaZCJjxIprl","pdf",20271645,4,1,22,"English","en",105,"# Introduction\n## Dexterous manipulation and generalization challenges\n## DexVerse: modular benchmark design\n## Dataset and evaluation setup","[{\"question\":\"What problem does DexVerse aim to solve in dexterous robotics research?\",\"answer\":\"DexVerse targets the lack of benchmarks that jointly test task diversity, sensory/visual variation, and robot embodiment coverage, which limits evaluation of cross-task and cross-embodiment generalization.\"},{\"question\":\"What tasks and robot embodiments does DexVerse include?\",\"answer\":\"DexVerse provides 100 dexterous tasks across a wide range of skills and supports multiple robot arms (including Franka Research 3, UR10e, xArm 7) and multiple dexterous hands (including Sharpa Wave, WUJI Hand, Shadow Hand, Inspire Hand, Allegro Hand, and LEAP Hand).\"},{\"question\":\"How does DexVerse evaluate visuomotor generalization and what data is released?\",\"answer\":\"It offers configurable visual variations such as textures, lighting, and camera viewpoints, and releases 3,180 VR teleoperation demonstrations with synchronized proprioceptive, RGB, depth, point-cloud, and state observations for unified policy evaluation.\"}]",1784201188,55,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"dexverse-a-modular-benchmark-for-multi-task-multi-embodiment-dexterous-manipulation","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/dexverse-a-modular-benchmark-for-multi-task-multi-embodiment-dexterous-manipulation/85114/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does DexVerse aim to solve in dexterous robotics research?","Question",{"text":75,"@type":76},"DexVerse targets the lack of benchmarks that jointly test task diversity, sensory/visual variation, and robot embodiment coverage, which limits evaluation of cross-task and cross-embodiment generalization.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What tasks and robot embodiments does DexVerse include?",{"text":80,"@type":76},"DexVerse provides 100 dexterous tasks across a wide range of skills and supports multiple robot arms (including Franka Research 3, UR10e, xArm 7) and multiple dexterous hands (including Sharpa Wave, WUJI Hand, Shadow Hand, Inspire Hand, Allegro Hand, and LEAP Hand).",{"name":82,"@type":73,"acceptedAnswer":83},"How does DexVerse evaluate visuomotor generalization and what data is released?",{"text":84,"@type":76},"It offers configurable visual variations such as textures, lighting, and camera viewpoints, and releases 3,180 VR teleoperation demonstrations with synchronized proprioceptive, RGB, depth, point-cloud, and state observations for unified policy evaluation.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]