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 bootstraps 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.</p>\n<p><video src=\"https://cdn-uploads.huggingface.co/production/uploads/6875264f22da0869fb9e41ec/chZa6DbxDC7cyJ23EUAzO.mp4\" controls=\"\" class=\"max-w-full!\"></video></p>","updatedAt":"2026-07-09T18:37:18.942Z","author":{"_id":"6875264f22da0869fb9e41ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/T3RuRxAaVwAI7TsIwAPb5.png","fullname":"Kelin Yu","name":"KelinYu1","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8264310359954834},"editors":["KelinYu1"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/T3RuRxAaVwAI7TsIwAPb5.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.03723","authors":[{"_id":"6a4e9cc048d70828b718dd18","user":{"_id":"6875264f22da0869fb9e41ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/T3RuRxAaVwAI7TsIwAPb5.png","isPro":false,"fullname":"Kelin Yu","user":"KelinYu1","type":"user","name":"KelinYu1"},"name":"Kelin Yu","status":"claimed_verified","statusLastChangedAt":"2026-07-09T11:53:47.262Z","hidden":false},{"_id":"6a4e9cc048d70828b718dd19","name":"Haode Zhang","hidden":false},{"_id":"6a4e9cc048d70828b718dd1a","name":"Harish Ravichandar","hidden":false},{"_id":"6a4e9cc048d70828b718dd1b","name":"Yunhai Han","hidden":false},{"_id":"6a4e9cc048d70828b718dd1c","name":"Ruohan Gao","hidden":false}],"publishedAt":"2026-07-04T00:00:00.000Z","submittedOnDailyAt":"2026-07-09T00:00:00.000Z","title":"OmniTacTune: Policy-Agnostic Real-World RL for Tactile Residual Adaptation of Visual Policies","submittedOnDailyBy":{"_id":"6875264f22da0869fb9e41ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/T3RuRxAaVwAI7TsIwAPb5.png","isPro":false,"fullname":"Kelin Yu","user":"KelinYu1","type":"user","name":"KelinYu1"},"summary":"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 bootstraps 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. Project page: https://colinyu1.github.io/omnitactune-site/","upvotes":3,"discussionId":"6a4e9cc048d70828b718dd1d","projectPage":"https://colinyu1.github.io/omnitactune-site/","ai_summary":"OmniTacTune enables efficient adaptation of tactile feedback to visual robot policies through a two-stage reinforcement learning approach that improves success rates in contact-rich manipulation tasks.","ai_keywords":["visual policies","tactile sensing","contact-rich manipulation","policy-agnostic","real-world RL","tactile feedback","residual correction","tactile-aware learning","base-policy rollouts","tactile residual policy","reinforcement learning"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","organization":{"_id":"68b3c3bbc375e05b059370b2","name":"UMCP","fullname":"University of Maryland College Park","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68b3c2c3a4ea236d1a97871a/bji3nI5ZWm2r4JX_-HLo0.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6875264f22da0869fb9e41ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/T3RuRxAaVwAI7TsIwAPb5.png","isPro":false,"fullname":"Kelin Yu","user":"KelinYu1","type":"user"},{"_id":"65037565da2d88e201f63b7a","avatarUrl":"/avatars/d1b6ce17236360e9583b8bb4cb87e506.svg","isPro":false,"fullname":"Runpeng Dai","user":"Leo-Dai","type":"user"},{"_id":"6623ea65b642e29cdf90a1b4","avatarUrl":"/avatars/e32e90574c1162b2be87ed78604e3e4d.svg","isPro":false,"fullname":"TongZheng","user":"TongZheng1999","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"68b3c3bbc375e05b059370b2","name":"UMCP","fullname":"University of Maryland College Park","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68b3c2c3a4ea236d1a97871a/bji3nI5ZWm2r4JX_-HLo0.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.03723.md","query":{}}">
OmniTacTune: Policy-Agnostic Real-World RL for Tactile Residual Adaptation of Visual Policies
Abstract
OmniTacTune enables efficient adaptation of tactile feedback to visual robot policies through a two-stage reinforcement learning approach that improves success rates in contact-rich manipulation tasks.
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 bootstraps 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. Project page: https://colinyu1.github.io/omnitactune-site/
Community
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 bootstraps 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.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2607.03723 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.03723 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.03723 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.