[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-327796-105":59,"doc-detail-327796-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","omnitactune-policy-agnostic-real-world-rl-for-tactile-residual-adaptation-of-visual-policies","OMNITACTUNE - Policy-Agnostic Real-World RL for Tactile Residual Adaptation of Visual Policies","","Visual policies learned from human videos, teleoperation, and robot demonstrations provide scalable motion priors, but often fail in contact-rich manipulation where local force and contact geometry determine success. Tactile sensing supplies these complementary signals, yet tactile data are expensive to collect and difficult to generalize across sensors, robots, and tasks. OMNITACTUNE introduces a policy-agnostic real-world RL pipeline that adapts tactile feedback to pretrained visual policies via residual correction. Extensive experiments validate cross-task, cross-policy, and cross-representation generalization, improving success from 5–40% to 85–100% across four real-world tasks within 40–80 minutes.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/omnitactune-policy-agnostic-real-world-rl-for-tactile-residual-adaptation-of-visual-policies/327796/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/omnitactune-policy-agnostic-real-world-rl-for-tactile-residual-adaptation-of-visual-policies/327796.png","ImageObject",300,407,{"name":92,"@type":93},"Ezra","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-09-21",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",14,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does OMNITACTUNE address in contact-rich manipulation?","Question",{"text":113,"@type":114},"Visual policies often fail at contact because they cannot directly measure forces or contact geometry. OMNITACTUNE uses tactile feedback to refine execution during contact to reduce these failures.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does OMNITACTUNE adapt tactile feedback to visual policies?",{"text":118,"@type":114},"It uses a policy-agnostic RL pipeline with residual correction. The method warm-starts tactile-aware learning from base-policy rollouts and then learns a lightweight tactile residual policy via online interaction.",{"name":120,"@type":111,"acceptedAnswer":121},"What performance improvements does OMNITACTUNE achieve in real-world tasks?",{"text":122,"@type":114},"Across four real-world contact-rich tasks, OMNITACTUNE improves success rates of visual base policies from 5–40% to 85–100% within 40–80 minutes.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},327796,1790049127,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":146,"preview_page_count":4},1099514068035,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","arXiv :2607 .03723v 1 [ cs .RO] 4 Jul 2026  \nOMNITACTUNE: Policy-Agnostic Real-World RL for Tactile Residual Adaptation of Visual Policies  \nKelin Yu 1∗ Haode Zhang 1∗ Harish Ravichandar2 Yunhai Han2 Ruohan Gao 1  \n1University of Maryland, College Park 2 Georgia Institute of Technology  \n∗Equal contribution  \nProject page: [https://colinyu1.github.io/omnitactune-site/](https://colinyu1.github.io/omnitactune-site/)  \nAbstract: Visual policies learned from human videos, teleoperation, and robot demonstrations offer scalable motion priors, but often fail in contact-rich manipulation, where success significantly depends on local force and contact geometry. Tactile sensing provides these complementary signals, yet tactile data remain costly to collect and hard to generalize across sensors, robots, and tasks. We introduce OMNITACTUNE, a policy-agnostic real-world RL pipeline that adapts tactile feedback to pretrained visual policies through residual correction. OMNITACTUNE uses a two-stage design: it first warm-starts tactile-aware learning from autonomous base-policy rollouts, then learns a lightweight tactile residual policy through online interaction. Extensive experiments show that OMNITACTUNE generalizes across diverse contact-rich tasks, visual base policies, and tactile representations. Across four real-world contact-rich tasks, it improves visual base policies from 5–40% success to 85–100% within 40–80 minutes, demonstrating an efficient path for adapting tactile feedback to scalable visual robot policies.  \nKeywords: Visuo-Tactile Policy, Real-World RL, Contact-rich Manipulation  \nFigure 1: OMNITACTUNE adapts tactile feedback to diverse visual base policies trained from human videos or robot data (left) through a two-stage real-world tactile residual RL pipeline (middle), enabling efficient tactile adaptation across challenging contact-rich manipulation tasks (right) .  \n1 Introduction  \nVisual robot policy learning has made tremendous progress, driven in large part by scaling. The field is moving toward increasingly scalable data sources, from teleoperated robot demonstrations [1, 2] to more recently large-scale human videos [3–5], offering a promising path toward general visual policies across diverse tasks, environments, and embodiments. However, scaling visual data alone does not resolve the challenges of contact-rich manipulation. Cameras provide rich spatial observations for planning, but they cannot directly measure contact or forces during physical interaction, such as insertion and tool use. As a result, many failures occur precisely at contact, where unobserved physical interactions determine whether a task succeeds or fails.  \nTouch, on the other hand, provides direct access to local interaction signals critical for contactrich manipulation, including normal force, shear force, and contact geometry [6, 7] . Prior work has  \nshown that tactile sensing improves tasks requiring precise contact reasoning [7–10], but these methods often rely on task-specific tactile data collected with a particular sensor or setup [11–14], limiting cross-platform transfer. Despite recent efforts to scale tactile data [12, 14, 15], tactile datasets remain orders of magnitude smaller than visual datasets [1, 4, 16] . This creates a fundamental scale gap, as tactile observations are rarely paired with the large-scale visual data from which general robot behaviors are increasingly learned.  \nOur key idea is to leverage the best of both worlds, rather than scale tactile data to the level of vision. Vision and touch play different roles in manipulation, much like in human learning. By watching others, humans can infer the global structure of a task: what objects are involved, how they move, and what action sequence is likely to succeed. Yet reliable execution often comes only through hands-on practice, where touch reveals whether parts are aligned, resistance is too high, and small adjustments are needed. This distinction suggests a scalabl","cbCaiqPRgDMHQLkv","https://ap.wps.com/l/cbCaiqPRgDMHQLkv","pdf",17885785,29,"English","# Introduction\n## Scalable visual policy learning and contact-rich limitations\n## Tactile sensing as local residual supervision\n## OMNITACTUNE: two-stage real-world RL pipeline","[{\"question\":\"What problem does OMNITACTUNE address in contact-rich manipulation?\",\"answer\":\"Visual policies often fail at contact because they cannot directly measure forces or contact geometry. OMNITACTUNE uses tactile feedback to refine execution during contact to reduce these failures.\"},{\"question\":\"How does OMNITACTUNE adapt tactile feedback to visual policies?\",\"answer\":\"It uses a policy-agnostic RL pipeline with residual correction. The method warm-starts tactile-aware learning from base-policy rollouts and then learns a lightweight tactile residual policy via online interaction.\"},{\"question\":\"What performance improvements does OMNITACTUNE achieve in real-world tasks?\",\"answer\":\"Across four real-world contact-rich tasks, OMNITACTUNE improves success rates of visual base policies from 5–40% to 85–100% within 40–80 minutes.\"}]","OMNITACTUNE - Policy-Agnostic Real-World RL for Tactile Residual Adaptation of Visual Policies | PDF",1789978057,73]