MP1: MeanFlow Tames Policy Learning in 1-step for Robotic Manipulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sheng, Juyi, Wang, Ziyi, Li, Peiming, Liu, Mengyuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908689446731776
author Sheng, Juyi
Wang, Ziyi
Li, Peiming
Liu, Mengyuan
author_facet Sheng, Juyi
Wang, Ziyi
Li, Peiming
Liu, Mengyuan
contents In robot manipulation, robot learning has become a prevailing approach. However, generative models within this field face a fundamental trade-off between the slow, iterative sampling of diffusion models and the architectural constraints of faster Flow-based methods, which often rely on explicit consistency losses. To address these limitations, we introduce MP1, which pairs 3D point-cloud inputs with the MeanFlow paradigm to generate action trajectories in one network function evaluation (1-NFE). By directly learning the interval-averaged velocity via the "MeanFlow Identity", our policy avoids any additional consistency constraints. This formulation eliminates numerical ODE-solver errors during inference, yielding more precise trajectories. MP1 further incorporates CFG for improved trajectory controllability while retaining 1-NFE inference without reintroducing structural constraints. Because subtle scene-context variations are critical for robot learning, especially in few-shot learning, we introduce a lightweight Dispersive Loss that repels state embeddings during training, boosting generalization without slowing inference. We validate our method on the Adroit and Meta-World benchmarks, as well as in real-world scenarios. Experimental results show MP1 achieves superior average task success rates, outperforming DP3 by 10.2% and FlowPolicy by 7.3%. Its average inference time is only 6.8 ms-19x faster than DP3 and nearly 2x faster than FlowPolicy. Our project page is available at https://mp1-2254.github.io/, and the code can be accessed at https://github.com/LogSSim/MP1.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10543
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MP1: MeanFlow Tames Policy Learning in 1-step for Robotic Manipulation
Sheng, Juyi
Wang, Ziyi
Li, Peiming
Liu, Mengyuan
Robotics
In robot manipulation, robot learning has become a prevailing approach. However, generative models within this field face a fundamental trade-off between the slow, iterative sampling of diffusion models and the architectural constraints of faster Flow-based methods, which often rely on explicit consistency losses. To address these limitations, we introduce MP1, which pairs 3D point-cloud inputs with the MeanFlow paradigm to generate action trajectories in one network function evaluation (1-NFE). By directly learning the interval-averaged velocity via the "MeanFlow Identity", our policy avoids any additional consistency constraints. This formulation eliminates numerical ODE-solver errors during inference, yielding more precise trajectories. MP1 further incorporates CFG for improved trajectory controllability while retaining 1-NFE inference without reintroducing structural constraints. Because subtle scene-context variations are critical for robot learning, especially in few-shot learning, we introduce a lightweight Dispersive Loss that repels state embeddings during training, boosting generalization without slowing inference. We validate our method on the Adroit and Meta-World benchmarks, as well as in real-world scenarios. Experimental results show MP1 achieves superior average task success rates, outperforming DP3 by 10.2% and FlowPolicy by 7.3%. Its average inference time is only 6.8 ms-19x faster than DP3 and nearly 2x faster than FlowPolicy. Our project page is available at https://mp1-2254.github.io/, and the code can be accessed at https://github.com/LogSSim/MP1.
title MP1: MeanFlow Tames Policy Learning in 1-step for Robotic Manipulation
topic Robotics
url https://arxiv.org/abs/2507.10543