Streaming Flow Policy: Simplifying diffusion/flow-matching policies by treating action trajectories as flow trajectories

• URL: http://arxiv.org/abs/2505.21851v2
• Date: Wed, 24 Sep 2025 23:51:56 GMT
• Title: Streaming Flow Policy: Simplifying diffusion/flow-matching policies by treating action trajectories as flow trajectories
• Abstract summary: We simplify diffusion$/$flow policies by treating action trajectories as flow trajectories.<n>Our algorithm samples from a narrow Gaussian around the last action.<n>It incrementally integrates a velocity field learned via flow matching to produce a sequence of actions that constitute a single trajectory.
• Score: 28.01020980367346
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in diffusion$/$flow-matching policies have enabled imitation learning of complex, multi-modal action trajectories. However, they are computationally expensive because they sample a trajectory of trajectories: a diffusion$/$flow trajectory of action trajectories. They discard intermediate action trajectories, and must wait for the sampling process to complete before any actions can be executed on the robot. We simplify diffusion$/$flow policies by treating action trajectories as flow trajectories. Instead of starting from pure noise, our algorithm samples from a narrow Gaussian around the last action. Then, it incrementally integrates a velocity field learned via flow matching to produce a sequence of actions that constitute a single trajectory. This enables actions to be streamed to the robot on-the-fly during the flow sampling process, and is well-suited for receding horizon policy execution. Despite streaming, our method retains the ability to model multi-modal behavior. We train flows that stabilize around demonstration trajectories to reduce distribution shift and improve imitation learning performance. Streaming flow policy outperforms prior methods while enabling faster policy execution and tighter sensorimotor loops for learning-based robot control. Project website: https://streaming-flow-policy.github.io/

Related papers

• ContextFlow: In-Context Flow Matching for Robot Manipulation [91.60992853468242]
  We introduce ContextFlow, a conditional flow-matching model that learns continuous action distributions for in-context imitation learning.<n>On LIBERO, ContextFlow outperforms ICRT by 35 percentage points in average success rate on unseen task configurations.<n>On real robots, it generalizes to unseen configurations of both single-arm and bimanual tasks, achieving 40% success on a new pen-uncapping configuration.
  arXiv  Detail & Related papers  (2026-09-06T21:53:58Z)
• FLASH: Efficient Visuomotor Policy via Sparse Sampling [25.25550374813956]
  We present Fast Legendre-polynomial Action policy via Sparse History-anchored flow.<n>FLASH Policy replaces discrete action-chunk generation with continuous Legendre trajectory representation.
  arXiv  Detail & Related papers  (2026-05-15T00:15:24Z)
• Flow Motion Policy: Manipulator Motion Planning with Flow Matching Models [3.4672098154671995]
  Flow Motion Policy is an open-loop, end-to-end neural motion planner for robotic manipulators.<n>It generates multiple end-to-end candidate paths, evaluates their collision status after planning, and executes the first collision-free solution.
  arXiv  Detail & Related papers  (2026-04-08T13:38:58Z)
• HiFlow: Tokenization-Free Scale-Wise Autoregressive Policy Learning via Flow Matching [11.405254375363818]
  Hierarchical Flow Policy (HiFlow) is a tokenization-free coarse-to-fine autoregressive policy.<n>HiFlow constructs multi-scale continuous action targets from each action chunk via simple temporal pooling.<n>Experiments on MimicGen, RoboTwin 2.0, and real-world environments demonstrate that HiFlow consistently outperforms existing methods.
  arXiv  Detail & Related papers  (2026-03-28T14:22:58Z)
• FlowConsist: Make Your Flow Consistent with Real Trajectory [99.22869983378062]
  We argue that current fast-flow training paradigms suffer from two fundamental issues.<n> conditional velocities constructed from randomly paired noise-data samples introduce systematic trajectory drift.<n>We propose FlowConsist, a training framework designed to enforce trajectory consistency in fast flows.
  arXiv  Detail & Related papers  (2026-02-06T03:24:23Z)
• Flow Policy Gradients for Robot Control [67.61978635211048]
  Flow matching policy gradients can be made effective for training and fine-tuning more expressive policies.<n>We show how policies can exploit the flow representation for exploration when training from scratch, as well as improved fine-tuning robustness over baselines.
  arXiv  Detail & Related papers  (2026-02-02T18:56:49Z)
• Flow Straighter and Faster: Efficient One-Step Generative Modeling via MeanFlow on Rectified Trajectories [14.36205662558203]
  Rectified MeanFlow is a framework that models the mean velocity field along the rectified trajectory using only a single reflow step.<n>Experiments on ImageNet at 64, 256, and 512 resolutions show that Re-MeanFlow consistently outperforms prior one-step flow distillation and Rectified Flow methods in both sample quality and training efficiency.
  arXiv  Detail & Related papers  (2025-11-28T16:50:08Z)
• One-Step Generative Policies with Q-Learning: A Reformulation of MeanFlow [56.13949180229929]
  We introduce a one-step generative policy for offline reinforcement learning that maps noise directly to actions via a residual reformulation of MeanFlow.<n>Our method achieves strong performance in both offline and offline-to-online reinforcement learning settings.
  arXiv  Detail & Related papers  (2025-11-17T06:34:17Z)
• SAC Flow: Sample-Efficient Reinforcement Learning of Flow-Based Policies via Velocity-Reparameterized Sequential Modeling [9.936731043466699]
  Training expressive flow-based policies with off-policy reinforcement learning is notoriously unstable due to gradient pathologies in the multi-step action sampling process.<n>We trace this instability to a fundamental connection: the flow rollout is algebraically equivalent to a residual recurrent computation, making it susceptible to the same vanishing and exploding gradients as RNNs.<n>We develop a practical SAC-based algorithm, enabled by a noise-augmented rollout, that facilitates direct end-to-end training of these policies.
  arXiv  Detail & Related papers  (2025-09-30T04:21:20Z)
• Self-Guided Action Diffusion [53.38661283705301]
  Self-guided action diffusion is a more efficient variant of bidirectional decoding tailored for diffusion-based policies.<n>Our method achieves up to 70% higher success rates than existing counterparts on challenging dynamic tasks.
  arXiv  Detail & Related papers  (2025-08-17T00:39:15Z)
• ActionSink: Toward Precise Robot Manipulation with Dynamic Integration of Action Flow [93.00917887667234]
  This paper introduces a novel robot manipulation framework, i.e., ActionSink, to pave the way toward precise action estimations.<n>As the name suggests, ActionSink reformulates the actions of robots as action-caused optical flows from videos, called "action flow"<n>Our framework outperformed prior SOTA on the LIBERO benchmark by a 7.9% success rate, and obtained nearly an 8% accuracy gain on the challenging long-horizon visual task LIBERO-Long.
  arXiv  Detail & Related papers  (2025-08-05T08:46:17Z)
• GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving [16.620086368657834]
  GoalFlow is an end-to-end autonomous driving method for generating high-quality multimodal trajectories.<n>It constrains the generative process to produce high-quality, multimodal trajectories.<n>GoalFlow achieves state-of-the-art performance, delivering robust multimodal trajectories for autonomous driving.
  arXiv  Detail & Related papers  (2025-03-07T18:52:08Z)
• FlowTS: Time Series Generation via Rectified Flow [67.41208519939626]
  FlowTS is an ODE-based model that leverages rectified flow with straight-line transport in probability space.<n>For unconditional setting, FlowTS achieves state-of-the-art performance, with context FID scores of 0.019 and 0.011 on Stock and ETTh datasets.<n>For conditional setting, we have achieved superior performance in solar forecasting.
  arXiv  Detail & Related papers  (2024-11-12T03:03:23Z)
• Affordance-based Robot Manipulation with Flow Matching [6.863932324631107]
  We present a framework for assistive robot manipulation.<n>We tackle two challenges: first, efficiently adapting large-scale models to downstream scene affordance understanding tasks, and second, effectively learning robot action trajectories by grounding the visual affordance model.<n>We learn robot action trajectories guided by affordances in a supervised flow matching method.
  arXiv  Detail & Related papers  (2024-09-02T09:11:28Z)
• Streaming Diffusion Policy: Fast Policy Synthesis with Variable Noise Diffusion Models [24.34842113104745]
  Diffusion models have seen rapid adoption in robotic imitation learning, enabling autonomous execution of complex tasks. Recent works have explored how the distillation of the diffusion process can be used to accelerate policy synthesis. We propose SDP (Streaming Diffusion Policy), an alternative method to accelerate policy synthesis.
  arXiv  Detail & Related papers  (2024-06-07T10:13:44Z)
• PeRFlow: Piecewise Rectified Flow as Universal Plug-and-Play Accelerator [73.80050807279461]
  Piecewise Rectified Flow (PeRFlow) is a flow-based method for accelerating diffusion models. PeRFlow achieves superior performance in a few-step generation.
  arXiv  Detail & Related papers  (2024-05-13T07:10:53Z)
• Optimal Flow Matching: Learning Straight Trajectories in Just One Step [89.37027530300617]
  We develop and theoretically justify the novel textbf Optimal Flow Matching (OFM) approach. It allows recovering the straight OT displacement for the quadratic transport in just one FM step. The main idea of our approach is the employment of vector field for FM which are parameterized by convex functions.
  arXiv  Detail & Related papers  (2024-03-19T19:44:54Z)
• Towards Understanding and Improving GFlowNet Training [71.85707593318297]
  We introduce an efficient evaluation strategy to compare the learned sampling distribution to the target reward distribution. We propose prioritized replay training of high-reward $x$, relative edge flow policy parametrization, and a novel guided trajectory balance objective.
  arXiv  Detail & Related papers  (2023-05-11T22:50:41Z)
• Learning Control by Iterative Inversion [21.127717602247454]
  We propose an algorithm for learning an inverse function without input-output pairs, but only with samples from the desired output distribution. We prove that iterative inversion can steer the learning correctly, under rather strict conditions on the function. We report an improved performance on imitating diverse behaviors compared to reward based methods.
  arXiv  Detail & Related papers  (2022-11-03T11:25:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.