HiFlow: Tokenization-Free Scale-Wise Autoregressive Policy Learning via Flow Matching

• URL: http://arxiv.org/abs/2603.27281v1
• Date: Sat, 28 Mar 2026 14:22:58 GMT
• Title: HiFlow: Tokenization-Free Scale-Wise Autoregressive Policy Learning via Flow Matching
• Abstract summary: 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.
• Score: 11.405254375363818
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Coarse-to-fine autoregressive modeling has recently shown strong promise for visuomotor policy learning, combining the inference efficiency of autoregressive methods with the global trajectory coherence of diffusion-based policies. However, existing approaches rely on discrete action tokenizers that map continuous action sequences to codebook indices, a design inherited from image generation where learned compression is necessary for high-dimensional pixel data. We observe that robot actions are inherently low-dimensional continuous vectors, for which such tokenization introduces unnecessary quantization error and a multi-stage training pipeline. In this work, we propose Hierarchical Flow Policy (HiFlow), a tokenization-free coarse-to-fine autoregressive policy that operates directly on raw continuous actions. HiFlow constructs multi-scale continuous action targets from each action chunk via simple temporal pooling. Specifically, it averages contiguous action windows to produce coarse summaries that are refined at finer temporal resolutions. The entire model is trained end-to-end in a single stage, eliminating the need for a separate tokenizer. Experiments on MimicGen, RoboTwin 2.0, and real-world environments demonstrate that HiFlow consistently outperforms existing methods including diffusion-based and tokenization-based autoregressive policies.

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)
• Primary-Fine Decoupling for Action Generation in Robotic Imitation [91.2899765310853]
  Multi-modal distribution in robotic manipulation action sequences poses critical challenges for imitation learning.<n>We propose Primary-Fine Decoupling for Action Generation (PF-DAG), a two-stage framework that decouples coarse action consistency from fine-grained variations.<n>PF-DAG outperforms state-of-the-art baselines across 56 tasks from Adroit, DexArt, and MetaWorld benchmarks.
  arXiv  Detail & Related papers  (2026-02-25T08:36:45Z)
• FlowSteer: Interactive Agentic Workflow Orchestration via End-to-End Reinforcement Learning [49.369614288007334]
  FlowSteer is an end-to-end reinforcement learning framework that takes a lightweight policy model as the agent and an executable canvas environment.<n>We show that FlowSteer significantly outperforms baselines across various tasks.
  arXiv  Detail & Related papers  (2026-02-02T05:30:42Z)
• CoLA-Flow Policy: Temporally Coherent Imitation Learning via Continuous Latent Action Flow Matching for Robotic Manipulation [4.726851899243877]
  LG-Flow Policy is a trajectory-level imitation learning framework that performs flow matching in a continuous latent action space.<n>By encoding action sequences into temporally regularized latent trajectories and learning an explicit latent-space flow, the proposed approach decouples global motion structure from low-level control noise.
  arXiv  Detail & Related papers  (2026-01-30T15:36:43Z)
• Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach [78.4812458793128]
  We propose textbfTACO, a test-time-scaling framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks.<n>Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits.
  arXiv  Detail & Related papers  (2025-12-02T14:42:54Z)
• 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)
• Unified Flow Matching for Long Horizon Event Forecasting [3.0639815065447036]
  We propose a unified flow matching framework for marked temporal point processes.<n>By learning continuous-time flows for both components, our method generates coherent long horizon event trajectories without sequential decoding.<n>We evaluate our model on six real-world benchmarks and demonstrate significant improvements over autoregressive and diffusion-based baselines in both accuracy and generation efficiency.
  arXiv  Detail & Related papers  (2025-08-06T19:42:49Z)
• Dense Policy: Bidirectional Autoregressive Learning of Actions [51.60428100831717]
  This paper introduces a bidirectionally expanded learning approach, termed Dense Policy, to establish a new paradigm for autoregressive policies in action prediction.<n>It employs a lightweight encoder-only architecture to iteratively unfold the action sequence from an initial single frame into the target sequence in a coarse-to-fine manner.<n>Experiments validate that our dense policy has superior autoregressive learning capabilities and can surpass existing holistic generative policies.
  arXiv  Detail & Related papers  (2025-03-17T14:28:08Z)
• CARP: Visuomotor Policy Learning via Coarse-to-Fine Autoregressive Prediction [28.761494362934087]
  Coarse-to-Fine AutoRegressive Policy (CARP) is a novel paradigm for visuomotor policy learning.<n>It redefines the autoregressive action generation process as a coarse-to-fine, next-scale approach.<n>CARP achieves competitive success rates, with up to a 10% improvement, and delivers 10x faster inference compared to state-of-the-art policies.
  arXiv  Detail & Related papers  (2024-12-09T18:59:18Z)
• AdaFlow: Imitation Learning with Variance-Adaptive Flow-Based Policies [21.024480978703288]
  We propose AdaFlow, an imitation learning framework based on flow-based generative modeling. AdaFlow represents the policy with state-conditioned ordinary differential equations (ODEs) We show that AdaFlow achieves high performance with fast inference speed.
  arXiv  Detail & Related papers  (2024-02-06T10:15:38Z)

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.