<video src=\"https://cdn-uploads.huggingface.co/production/uploads/6762165426fd6f7f087539b3/LAezQiubZ94-8zEzUgO3s.mp4\" controls=\"\" class=\"max-w-full!\"></video></p>","updatedAt":"2026-07-30T16:51:12.568Z","author":{"_id":"6762165426fd6f7f087539b3","avatarUrl":"/avatars/b08a8de8678c07db3eabf4c40467be5a.svg","fullname":"Sungjae Park","name":"sungj1026","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.3694959878921509},"editors":["sungj1026"],"editorAvatarUrls":["/avatars/b08a8de8678c07db3eabf4c40467be5a.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.26055","authors":[{"_id":"6a6b80ffb2106777884ab1d5","name":"Sungjae Park","hidden":false},{"_id":"6a6b80ffb2106777884ab1d6","name":"Shubham Tulsiani","hidden":false}],"publishedAt":"2026-07-28T00:00:00.000Z","submittedOnDailyAt":"2026-07-30T00:00:00.000Z","title":"πR^2: Reactive Real-time Flow Policies","submittedOnDailyBy":{"_id":"6762165426fd6f7f087539b3","avatarUrl":"/avatars/b08a8de8678c07db3eabf4c40467be5a.svg","isPro":false,"fullname":"Sungjae Park","user":"sungj1026","type":"user","name":"sungj1026"},"summary":"Generalist manipulation policies increasingly take the form of action-chunking flow policies built on large pretrained backbones. Such chunks run open-loop, so the policy cannot react to sensory input arriving mid-execution, sacrificing reactivity. Replanning more often would restore it, but the perception-to-action pipeline (a large backbone plus multiple denoising steps) is too slow: this latency forbids frequent replanning and leaves committed actions stale, making such policies ill-suited for dynamic, closed-loop control. We present πR^2, which makes these policies reactive and real-time while retaining large backbones, expressive multi-modal policies, and multi-action prediction. Built on the per-position noise schedule of diffusion forcing, πR^2 contributes two ideas. First, it splits conditioning into a fast channel (proprioception, fresh every tick) and an asynchronously updated slow channel (vision-language features), so the policy reacts to proprioception within a chunk while tolerating stale vision. Second, a latency-adaptive flow schedule treats in-flight actions as inpainting conditioning and emits actions in one denoising step per call, letting one trained model adapt to varying hardware latency. Requiring minimal modification to existing architectures, πR^2 can be finetuned from a pretrained policy: applied to GR00T-N1.7 on a real xArm6+XHand platform, it replans closed-loop roughly 4times faster than the base policy (~25Hz on an A5000 GPU), acting on a fresh observation every 40ms. Across simulation and real-world manipulation tasks, πR^2 improves the success rate by up to 23% in simulation and 30% in the real world over the strongest baseline. Project page: https://pi-r2-flow.github.io/","upvotes":2,"discussionId":"6a6b80ffb2106777884ab1d7","projectPage":"https://pi-r2-flow.github.io/","githubRepo":"https://github.com/pi-r2-flow/pi-r2-flow","githubRepoAddedBy":"user","githubStars":12,"organization":{"_id":"6362c6c180c1a705a6ed0d57","name":"CMU-SCS","fullname":"Carnegie Mellon University School of Computer Science","avatar":"https://www.gravatar.com/avatar/3178df24f569afc6497a1d64743d18ca?d=retro&size=100"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6762165426fd6f7f087539b3","avatarUrl":"/avatars/b08a8de8678c07db3eabf4c40467be5a.svg","isPro":false,"fullname":"Sungjae Park","user":"sungj1026","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6362c6c180c1a705a6ed0d57","name":"CMU-SCS","fullname":"Carnegie Mellon University School of Computer Science","avatar":"https://www.gravatar.com/avatar/3178df24f569afc6497a1d64743d18ca?d=retro&size=100"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.26055.md","query":{}}">
πR^2: Reactive Real-time Flow Policies
Abstract
Generalist manipulation policies increasingly take the form of action-chunking flow policies built on large pretrained backbones. Such chunks run open-loop, so the policy cannot react to sensory input arriving mid-execution, sacrificing reactivity. Replanning more often would restore it, but the perception-to-action pipeline (a large backbone plus multiple denoising steps) is too slow: this latency forbids frequent replanning and leaves committed actions stale, making such policies ill-suited for dynamic, closed-loop control. We present πR^2, which makes these policies reactive and real-time while retaining large backbones, expressive multi-modal policies, and multi-action prediction. Built on the per-position noise schedule of diffusion forcing, πR^2 contributes two ideas. First, it splits conditioning into a fast channel (proprioception, fresh every tick) and an asynchronously updated slow channel (vision-language features), so the policy reacts to proprioception within a chunk while tolerating stale vision. Second, a latency-adaptive flow schedule treats in-flight actions as inpainting conditioning and emits actions in one denoising step per call, letting one trained model adapt to varying hardware latency. Requiring minimal modification to existing architectures, πR^2 can be finetuned from a pretrained policy: applied to GR00T-N1.7 on a real xArm6+XHand platform, it replans closed-loop roughly 4times faster than the base policy (~25Hz on an A5000 GPU), acting on a fresh observation every 40ms. Across simulation and real-world manipulation tasks, πR^2 improves the success rate by up to 23% in simulation and 30% in the real world over the strongest baseline. Project page: https://pi-r2-flow.github.io/
Community
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/2607.26055 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.26055 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.26055 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.