paper-with-me

홈 › Papers

OMP: One-step Meanflow Policy with Directional Alignment

2025-12-22 · Han Fang, Yize Huang, Yuheng Zhao, Paul Weng, Xiao Li, Yutong Ban arxiv

Robot manipulation has increasingly adopted data-driven generative policy frameworks, yet the field faces a persistent trade-off: diffusion models suffer from high inference latency, while flow-based methods often require complex architectural constraints. Although in image generation domain, the MeanFlow paradigm offers a path to single-step inference, its direct application to robotics is impeded by critical theoretical pathologies, specifically spectral bias and gradient starvation in low-velocity regimes. To overcome these limitations, we propose the One-step MeanFlow Policy (OMP), a novel framework designed for high-fidelity, real-time manipulation. We introduce a lightweight directional alignment mechanism to explicitly synchronize predicted velocities with true mean velocities. Furthermore, we implement a Differential Derivation Equation (DDE) to approximate the Jacobian-Vector Product (JVP) operator, which decouples forward and backward passes to significantly reduce memory complexity. Extensive experiments on the Adroit and Meta-World benchmarks demonstrate that OMP outperforms state-of-the-art methods in success rate and trajectory accuracy, particularly in high-precision tasks, while retaining the efficiency of single-step generation.

📄 PDF Abstract BibTeX arXiv:2512.19347

Code (0)

등록된 구현이 없습니다.

Tasks

Robot ManipulationImage Generation

Similar Papers 제목 키워드 기반

FlashRender: Few-Step Generative Rendering via Camera-Controlled Video MeanFlow

2026-09-03 · Byeongjun Park, Byung-Hoon Kim, Hyungjin Chung hf

We present FlashRender, a few-step generative rendering framework that retakes a source video along a target camera trajectory in seconds. We identify sampling-step-dependent camera control as a prominent manifestation o…

HybridFlow: A Two-Step Generative Policy for Robotic Manipulation

2026-02-14 · Zhenchen Dong, Jinna Fu, Jiaming Wu, Shengyuan Yu 외 arxiv

Limited by inference latency, existing robot manipulation policies lack sufficient real-time interaction capability with the environment. Although faster generation methods such as flow matching are gradually replacing d…

Robot ManipulationImage Generation

MeanFlowNFT: Bringing Forward-Process RL to Average-Velocity Generators

2026-07-16 · Yushi Huang, Xiangxin Zhou, Jun Zhang, Liefeng Bo 외 hf

MeanFlow generators achieve fast few-step sampling by predicting average velocities over time intervals, making them attractive for efficient generation. Reinforcement learning (RL) has become a powerful way to align dif…

Reinforcement LearningVideo Generation

Score-Based One-step MeanFlow Policy Optimization

2026-05-22 · Kyungyoon Kim, Donghyeon Ki, Hee-Jun Ahn, Byung-Jun Lee arxiv

Diffusion and flow matching have emerged as expressive policy classes in reinforcement learning, but their reliance on multi-step denoising imposes substantial computational overhead at inference time, which is particula…

Reinforcement Learning

Stochastic MeanFlow Policies: One-Step Generative Control with Entropic Mirror Descent

2026-05-20 · Zeyuan Wang, Da Li, Yulin Chen, Yuehu Gong 외 arxiv

Online off-policy reinforcement learning (RL) is shaped by two coupled choices: the policy class and the update rule. Gaussian policies are fast and have tractable entropy, but struggle with multimodal action distributio…

Reinforcement Learning