中文
相关论文

相关论文: NoiseGate: Learning Per-Latent Timestep Schedules …

200 篇论文

A plausible scene evolution depends on the maneuver being considered, while a good maneuver depends on how the scene may evolve. Existing World Action Models (WAMs) largely miss this reciprocity, treating world prediction and action…

计算机视觉与模式识别 · 计算机科学 2026-05-13 Hongbo Lu , Liang Yao , Chenghao He , Haoyu Wang , Xiang Gu , Xianfei Li , Wenlong Liao , Tao He , Pai Peng

World-Action Models (WAM) initialized from pre-trained video generation backbones have demonstrated remarkable potential for robot policy learning. However, existing approaches face two critical bottlenecks that hinder performance and…

World Action Models (WAMs) have emerged as a promising paradigm for robot control by modeling physical dynamics. Current WAMs generally follow two paradigms: the "Imagine-then-Execute" approach, which uses video prediction to infer actions…

机器人学 · 计算机科学 2026-05-12 Qiuxuan Feng , Jiale Yu , Jiaming Liu , Yueru Jia , Zhuangzhe Wu , Hao Chen , Zezhong Qian , Shuo Gu , Peng Jia , Siwei Ma , Shanghang Zhang

Imitation learning has emerged as a promising approach towards building generalist robots. However, scaling imitation learning for large robot foundation models remains challenging due to its reliance on high-quality expert demonstrations.…

机器人学 · 计算机科学 2025-05-26 Chuning Zhu , Raymond Yu , Siyuan Feng , Benjamin Burchfiel , Paarth Shah , Abhishek Gupta

Latent action models (LAMs) aim to learn action-relevant changes from unlabeled videos by compressing changes between frames as latents. However, differences between video frames can be caused by controllable changes as well as exogenous…

机器学习 · 计算机科学 2025-11-13 Chuheng Zhang , Tim Pearce , Pushi Zhang , Kaixin Wang , Xiaoyu Chen , Wei Shen , Li Zhao , Jiang Bian

Latent Action Models (LAMs) enable the learning of world models from unlabeled video by inferring abstract actions between consecutive frames. However, LAMs face a fundamental trade-off between action abstraction and generation fidelity.…

计算机视觉与模式识别 · 计算机科学 2026-05-18 Tianqiu Zhang , Muyang Lyu , Yufan Zhang , Fang Fang , Si Wu

Pretrained foundation models have become an important basis for end-to-end autonomous driving. In contrast to vision-language models pretrained primarily on static image-text pairs, video generative models capture temporal dynamics and…

计算机视觉与模式识别 · 计算机科学 2026-05-28 Chen Shi , Jinrui Xu , Shaoshuai Shi , Kehua Sheng , Bo Zhang , Li Jiang

Learning latent actions from action-free video has emerged as a powerful paradigm for scaling up controllable world model learning. Latent actions provide a natural interface for users to iteratively generate and manipulate videos. However,…

机器学习 · 计算机科学 2026-05-26 Zizhao Wang , Chang Shi , Jiaheng Hu , Kevin Rohling , Roberto Martín-Martín , Amy Zhang , Peter Stone

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…

计算机视觉与模式识别 · 计算机科学 2025-12-29 Hongzhe Bi , Hengkai Tan , Shenghao Xie , Zeyuan Wang , Shuhe Huang , Haitian Liu , Ruowen Zhao , Yao Feng , Chendong Xiang , Yinze Rong , Hongyan Zhao , Hanyu Liu , Zhizhong Su , Lei Ma , Hang Su , Jun Zhu

We propose X-WAM, a Unified 4D World Model that unifies real-time robotic action execution and high-fidelity 4D world synthesis (video + 3D reconstruction) in a single framework, addressing the critical limitations of prior unified world…

机器人学 · 计算机科学 2026-05-08 Jun Guo , Qiwei Li , Peiyan Li , Zilong Chen , Nan Sun , Yifei Su , Heyun Wang , Yuan Zhang , Xinghang Li , Huaping Liu

Various world model frameworks are being developed today based on autoregressive frameworks that rely on discrete representations of actions and observations, and these frameworks are succeeding in constructing interactive generative models…

机器学习 · 计算机科学 2025-03-14 Kohei Hayashi , Masanori Koyama , Julian Jorge Andrade Guerreiro

World Action Models (WAMs) improve robot manipulation by using video-based future representations to condition action generation. In pixel-space WAMs, however, the best action condition is not necessarily the fully denoised video.…

机器人学 · 计算机科学 2026-05-28 Yirui Sun , Guangyu Zhuge , Keliang Liu , Jie Gu , Xinyu Bing , Zhongxue Gan , Chunxu Tian

Leveraging future observation modeling to facilitate action generation presents a promising avenue for enhancing the capabilities of Vision-Language-Action (VLA) models. However, existing approaches struggle to strike a balance between…

World models have become crucial for autonomous driving, as they learn how scenarios evolve over time to address the long-tail challenges of the real world. However, current approaches relegate world models to limited roles: they operate…

计算机视觉与模式识别 · 计算机科学 2026-04-20 Tianze Xia , Yongkang Li , Lijun Zhou , Jingfeng Yao , Kaixin Xiong , Haiyang Sun , Bing Wang , Kun Ma , Guang Chen , Hangjun Ye , Wenyu Liu , Xinggang Wang

Pretrained video generation models provide strong priors for robot control, but existing unified world action models still struggle to decode reliable actions without substantial robot-specific training. We attribute this limitation to a…

机器人学 · 计算机科学 2026-04-14 Liaoyuan Fan , Zetian Xu , Chen Cao , Wenyao Zhang , Mingqi Yuan , Jiayu Chen

Vision-Language-Action (VLA) models have achieved strong semantic generalization for embodied policy learning, yet they learn reactive observation-to-action mappings without explicitly modeling how the physical world evolves under…

Adapting pretrained video generation models into controllable world models via latent actions is a promising step towards creating generalist world models. The dominant paradigm adopts a two-stage approach that trains latent action model…

机器学习 · 计算机科学 2026-04-07 Yucen Wang , Fengming Zhang , De-Chuan Zhan , Li Zhao , Kaixin Wang , Jiang Bian

This paper presents the World-Action Model (WAM), an action-regularized world model that jointly reasons over future visual observations and the actions that drive state transitions. Unlike conventional world models trained solely via image…

人工智能 · 计算机科学 2026-04-01 Yuci Han , Alper Yilmaz

Anticipating future actions is inherently uncertain. Given an observed video segment containing ongoing actions, multiple subsequent actions can plausibly follow. This uncertainty becomes even larger when predicting far into the future.…

计算机视觉与模式识别 · 计算机科学 2023-11-28 Zeyun Zhong , Chengzhi Wu , Manuel Martin , Michael Voit , Juergen Gall , Jürgen Beyerer

World Action Models (WAMs) enhance Vision-Language-Action policies by jointly predicting scene evolution and robot actions, but existing methods usually represent the predicted world as holistic images, video tokens, or global latents.…

‹ 上一页 1 2 3 10 下一页 ›