Motus: A Unified Latent Action World Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bi, Hongzhe, Tan, Hengkai, Xie, Shenghao, Wang, Zeyuan, Huang, Shuhe, Liu, Haitian, Zhao, Ruowen, Feng, Yao, Xiang, Chendong, Rong, Yinze, Zhao, Hongyan, Liu, Hanyu, Su, Zhizhong, Ma, Lei, Su, Hang, Zhu, Jun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917169270358016
author Bi, Hongzhe
Tan, Hengkai
Xie, Shenghao
Wang, Zeyuan
Huang, Shuhe
Liu, Haitian
Zhao, Ruowen
Feng, Yao
Xiang, Chendong
Rong, Yinze
Zhao, Hongyan
Liu, Hanyu
Su, Zhizhong
Ma, Lei
Su, Hang
Zhu, Jun
author_facet Bi, Hongzhe
Tan, Hengkai
Xie, Shenghao
Wang, Zeyuan
Huang, Shuhe
Liu, Haitian
Zhao, Ruowen
Feng, Yao
Xiang, Chendong
Rong, Yinze
Zhao, Hongyan
Liu, Hanyu
Su, Zhizhong
Ma, Lei
Su, Hang
Zhu, Jun
contents While a general embodied agent must function as a unified system, current methods are built on isolated models for understanding, world modeling, and control. This fragmentation prevents unifying multimodal generative capabilities and hinders learning from large-scale, heterogeneous data. In this paper, we propose Motus, a unified latent action world model that leverages existing general pretrained models and rich, sharable motion information. Motus introduces a Mixture-of-Transformer (MoT) architecture to integrate three experts (i.e., understanding, video generation, and action) and adopts a UniDiffuser-style scheduler to enable flexible switching between different modeling modes (i.e., world models, vision-language-action models, inverse dynamics models, video generation models, and video-action joint prediction models). Motus further leverages the optical flow to learn latent actions and adopts a recipe with three-phase training pipeline and six-layer data pyramid, thereby extracting pixel-level "delta action" and enabling large-scale action pretraining. Experiments show that Motus achieves superior performance against state-of-the-art methods in both simulation (a +15% improvement over X-VLA and a +45% improvement over Pi0.5) and real-world scenarios(improved by +11~48%), demonstrating unified modeling of all functionalities and priors significantly benefits downstream robotic tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13030
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Motus: A Unified Latent Action World Model
Bi, Hongzhe
Tan, Hengkai
Xie, Shenghao
Wang, Zeyuan
Huang, Shuhe
Liu, Haitian
Zhao, Ruowen
Feng, Yao
Xiang, Chendong
Rong, Yinze
Zhao, Hongyan
Liu, Hanyu
Su, Zhizhong
Ma, Lei
Su, Hang
Zhu, Jun
Computer Vision and Pattern Recognition
Machine Learning
Robotics
While a general embodied agent must function as a unified system, current methods are built on isolated models for understanding, world modeling, and control. This fragmentation prevents unifying multimodal generative capabilities and hinders learning from large-scale, heterogeneous data. In this paper, we propose Motus, a unified latent action world model that leverages existing general pretrained models and rich, sharable motion information. Motus introduces a Mixture-of-Transformer (MoT) architecture to integrate three experts (i.e., understanding, video generation, and action) and adopts a UniDiffuser-style scheduler to enable flexible switching between different modeling modes (i.e., world models, vision-language-action models, inverse dynamics models, video generation models, and video-action joint prediction models). Motus further leverages the optical flow to learn latent actions and adopts a recipe with three-phase training pipeline and six-layer data pyramid, thereby extracting pixel-level "delta action" and enabling large-scale action pretraining. Experiments show that Motus achieves superior performance against state-of-the-art methods in both simulation (a +15% improvement over X-VLA and a +45% improvement over Pi0.5) and real-world scenarios(improved by +11~48%), demonstrating unified modeling of all functionalities and priors significantly benefits downstream robotic tasks.
title Motus: A Unified Latent Action World Model
topic Computer Vision and Pattern Recognition
Machine Learning
Robotics
url https://arxiv.org/abs/2512.13030