πR^2: Reactive Real-time Flow Policies

Hugging Face Daily Papers Papers
robot-manipulation flow-policies real-time-control reactive-policies diffusion-forcing closed-loop-control

Summary

πR^2 introduces a reactive real-time flow policy for robot manipulation that splits conditioning into fast and slow channels and uses a latency-adaptive flow schedule, enabling closed-loop replanning ~4x faster than baseline policies and improving success rates by up to 30%.

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/
Original Article
View Cached Full Text

Cached at: 07/30/26, 05:49 PM

Paper page - πR^2: Reactive Real-time Flow Policies

Source: https://huggingface.co/papers/2607.26055

Abstract

Generalistmanipulationpoliciesincreasinglytaketheformofaction-chunkingflowpoliciesbuiltonlargepretrainedbackbones.Suchchunksrunopen-loop,sothepolicycannotreacttosensoryinputarrivingmid-execution,sacrificingreactivity.Replanningmoreoftenwouldrestoreit,buttheperception-to-actionpipeline(alargebackboneplusmultipledenoisingsteps)istooslow:thislatencyforbidsfrequentreplanningandleavescommittedactionsstale,makingsuchpoliciesill-suitedfordynamic,closed-loopcontrol.WepresentπR^2,whichmakesthesepoliciesreactiveandreal-timewhileretaininglargebackbones,expressivemulti-modalpolicies,andmulti-actionprediction.Builtontheper-positionnoisescheduleofdiffusionforcing,πR^2contributestwoideas.First,itsplitsconditioningintoafastchannel(proprioception,fresheverytick)andanasynchronouslyupdatedslowchannel(vision-languagefeatures),sothepolicyreactstoproprioceptionwithinachunkwhiletoleratingstalevision.Second,alatency-adaptiveflowscheduletreatsin-flightactionsasinpaintingconditioningandemitsactionsinonedenoisingsteppercall,lettingonetrainedmodeladapttovaryinghardwarelatency.Requiringminimalmodificationtoexistingarchitectures,πR^2canbefinetunedfromapretrainedpolicy:appliedtoGR00T-N1.7onarealxArm6+XHandplatform,itreplansclosed-looproughly4timesfasterthanthebasepolicy(~25HzonanA5000GPU),actingonafreshobservationevery40ms.Acrosssimulationandreal-worldmanipulationtasks,πR^2improvesthesuccessratebyupto23%insimulationand30%intherealworldoverthestrongestbaseline.Projectpage:https://pi-r2-flow.github.io/

View arXiv pageView PDFProject pageGitHub12Add to collection

Get this paper in your agent:

hf papers read 2607\.26055

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.26055 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.26055 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.26055 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Warp RL: Reshaping Base Policy Distributions for Dynamics Adaptation

arXiv cs.LG

Warp RL replaces additive residual corrections in reinforcement learning with an invertible, state-conditioned transformation of the base policy's action distribution using monotonic rational-quadratic spline flows, enabling adaptation of distribution shape, scale, and geometry under dynamics shifts. It matches or outperforms residual correction in ManiSkill3 manipulation tasks and achieves 30% faster task completion in a real robot peg-insertion task.