世界模型 - arXivDaily 专题

2510.00855 2025-10-02 cs.CV cs.AI cs.CL cs.LG 94%

Can World Models Benefit VLMs for World Dynamics?

Kevin Zhang, Kuangzhi Ge, Xiaowei Chi, Renrui Zhang, Shaojun Shi, Zhen Dong, Sirui Han, Shanghang Zhang

机构 * Peking University（北京大学）； Hong Kong University of Science and Technology（香港科学与技术大学）； Chinese University of Hong Kong（香港中文大学）； University of California, Santa Barbara（加州大学圣芭芭拉分校）

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments Project page: https://dyva-worldlm.github.io

2509.24527 2025-09-30 cs.AI cs.LG cs.RO stat.ML 94%

Training Agents Inside of Scalable World Models

Danijar Hafner, Wilson Yan, Timothy Lillicrap

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments Website: https://danijar.com/dreamer4/

2507.12496 2025-07-18 cs.RO cs.AI cs.LG 94%

FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Making

Yucen Wang, Rui Yu, Shenghua Wan, Le Gan, De-Chuan Zhan

机构 * School of Artificial Intelligence, Nanjing University, China（人工智能学院，南京大学）； National Key Laboratory for Novel Software Technology, Nanjing University, China（新型软件技术国家重点实验室，南京大学）

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments Accepted by Forty-Second International Conference on Machine Learning (ICML 2025)

2506.01182 2025-07-10 cs.RO cs.AI 94%

Humanoid World Models: Open World Foundation Models for Humanoid Robotics

Muhammad Qasim Ali, Aditya Sridhar, Shahbuland Matiana, Alex Wong, Mohammad Al-Sharman

机构 * University of Waterloo（滑铁卢大学）

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2506.01600 2025-06-03 cs.RO cs.AI cs.CV 94%

WoMAP: World Models For Embodied Open-Vocabulary Object Localization

Tenny Yin, Zhiting Mei, Tao Sun, Lihan Zha, Emily Zhou, Jeremy Bao, Miyu Yamane, Ola Shorinwa, Anirudha Majumdar

机构 * Princeton University（普林斯顿大学）； McGill University（麦吉尔大学）

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2504.02792 2025-05-26 cs.RO cs.AI cs.LG 94%

Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets

Chuning Zhu, Raymond Yu, Siyuan Feng, Benjamin Burchfiel, Paarth Shah, Abhishek Gupta

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2503.15168 2025-03-20 cs.AI cs.CV cs.ET cs.LG 94%

World Models in Artificial Intelligence: Sensing, Learning, and Reasoning Like a Child

Javier Del Ser, Jesus L. Lobo, Heimo Müller, Andreas Holzinger

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments 11 pages, 1 figure

2503.08122 2025-03-12 cs.LG cs.AI cs.CV 94%

Toward Stable World Models: Measuring and Addressing World Instability in Generative Environments

Soonwoo Kwon, Jin-Young Kim, Hyojun Go, Kyungjune Baek

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments Preprint

2412.05337 2024-12-10 cs.CV cs.LG cs.RO 94%

ACT-Bench: Towards Action Controllable World Models for Autonomous Driving

Hidehisa Arai, Keishi Ishihara, Tsubasa Takahashi, Yu Yamaguchi

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2411.02914 2024-11-06 cs.AI cs.CV cs.RO 94%

Exploring the Interplay Between Video Generation and World Models in Autonomous Driving: A Survey

Ao Fu, Yi Zhou, Tao Zhou, Yi Yang, Bojun Gao, Qun Li, Guobin Wu, Ling Shao

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2411.02446 2024-11-06 cs.LG cs.AI cs.RO 94%

Learning World Models for Unconstrained Goal Navigation

Yuanlin Duan, Wensen Mao, He Zhu

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments NeurIPS2024 Poster. arXiv admin note: substantial text overlap with arXiv:2411.01396

2403.02622 2024-05-08 cs.LG cs.AI cs.RO 94%

World Models for Autonomous Driving: An Initial Survey

Yanchen Guan, Haicheng Liao, Zhenning Li, Jia Hu, Runze Yuan, Yunjian Li, Guohui Zhang, Chengzhong Xu

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2402.06665 2024-05-01 cs.AI cs.CL cs.LG cs.RO 94%

The Essential Role of Causality in Foundation World Models for Embodied AI

Tarun Gupta, Wenbo Gong, Chao Ma, Nick Pawlowski, Agrin Hilmkil, Meyer Scetbon, Marc Rigter, Ade Famoti, Ashley Juan Llorens, Jianfeng Gao, Stefan Bauer, Danica Kragic, Bernhard Schölkopf, Cheng Zhang

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2403.00504 2024-03-04 cs.CV cs.AI cs.LG 94%

Learning and Leveraging World Models in Visual Representation Learning

Quentin Garrido, Mahmoud Assran, Nicolas Ballas, Adrien Bardes, Laurent Najman, Yann LeCun

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments 23 pages, 16 figures

2402.03326 2024-02-07 cs.CV cs.LG 94%

Slot Structured World Models

Jonathan Collu, Riccardo Majellaro, Aske Plaat, Thomas M. Moerland

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2306.09179 2023-06-16 cs.CV cs.AI cs.RO 94%

Neural World Models for Computer Vision

Anthony Hu

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments PhD thesis

2305.14223 2023-05-24 cs.MA cs.AI cs.GT cs.LG 94%

Co-Learning Empirical Games and World Models

Max Olan Smith, Michael P. Wellman

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2209.00588 2023-03-02 cs.LG cs.AI cs.CV 94%

Transformers are Sample-Efficient World Models

Vincent Micheli, Eloi Alonso, François Fleuret

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments ICLR 2023 (notable top 5%)

2301.05832 2023-01-18 cs.RO cs.AI cs.LG 94%

World Models and Predictive Coding for Cognitive and Developmental Robotics: Frontiers and Challenges

Tadahiro Taniguchi, Shingo Murata, Masahiro Suzuki, Dimitri Ognibene, Pablo Lanillos, Emre Ugur, Lorenzo Jamone, Tomoaki Nakamura, Alejandra Ciria, Bruno Lara, Giovanni Pezzulo

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments 28 pages, 3 figures

2205.13817 2022-12-20 cs.LG cs.AI cs.RO 94%

Iso-Dream: Isolating and Leveraging Noncontrollable Visual Dynamics in World Models

Minting Pan, Xiangming Zhu, Yunbo Wang, Xiaokang Yang

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2206.11403 2022-11-29 cs.LG cs.AI cs.RO 94%

Curious Exploration via Structured World Models Yields Zero-Shot Object Manipulation

Cansu Sancaktar, Sebastian Blaes, Georg Martius

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments NeurIPS 2022 camera-ready version

2206.14176 2022-06-29 cs.RO cs.AI cs.LG 94%

DayDreamer: World Models for Physical Robot Learning

Philipp Wu, Alejandro Escontrela, Danijar Hafner, Ken Goldberg, Pieter Abbeel

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments Website: https://danijar.com/daydreamer

2201.09802 2022-02-08 cs.LG cs.AI cs.RO 94%

Constrained Policy Optimization via Bayesian World Models

Yarden As, Ilnura Usmanova, Sebastian Curi, Andreas Krause

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

2104.05632 2021-08-04 cs.LG cs.AI 94%

Augmented World Models Facilitate Zero-Shot Dynamics Generalization From a Single Offline Environment

Philip J. Ball, Cong Lu, Jack Parker-Holder, Stephen Roberts

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

Comments Accepted @ ICML 2021; Spotlight @ ICLR 2021 "Self-Supervision for Reinforcement Learning Workshop"

2608.19779 2026-08-21 quant-ph cs.AI cs.LG 新提交 94%

An Irreducible Quantum Advantage in Aligning World Models with Reality

世界模型与现实对齐的不可约量子优势

Josep Lumbreras, Hailan Ma, Jayne Thompson, Mile Gu

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

AI总结该研究证明经典世界模型无法通过增加记忆实现与真实世界的完美策略对齐，而量子世界模型仅用单个qutrit即可做到，揭示了世界模型对齐现实的不可约量子优势。

Comments 31 pages, 4 figures

2608.18746 2026-08-20 cs.LG cs.CV 新提交 94%

Decision-Metric Alignment in Latent World Models: Diagnostics and Action-Conditioned Objectives for MPC Planning

潜在世界模型中的决策度量对齐：MPC规划的诊断与动作条件目标

Jiawei Wang, Ke Rui, Yushen Zuo, Yichun Feng, Minglei Li

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

AI总结该研究针对潜在世界模型中MPC规划的决策度量对齐问题，提出两种秩一致性诊断指标，开发DA-LeWM模型，验证其可加快收敛并提升在线成功率，改善潜在MPC的几何结构。

2608.15156 2026-08-20 cs.RO cs.AI 版本更新 94%

Low-Rank Dynamics-Effective Latent Carriers for Counterfactual Rollout in Learned World Models

用于学习世界模型中反事实滚动的低秩动力学有效潜在载体

Yang Liu, Yuming Chen

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

AI总结本文针对学习世界模型的反事实滚动问题，提出基于低秩动力学有效潜在载体的方法，在两物体碰撞环境中验证秩4修补块可实现稳定的12步自主反事实滚动，且效果可复现。

Comments Revised version: removed an inconclusive development-only event-relative phase analysis; the main rank-4 carrier, fresh-checkpoint replication, B1/B2 temporal reuse, position-edit, and joint-edit conclusions are unchanged

2608.17542 2026-08-19 cs.LG cs.AI 新提交 94%

No Gaussian Required: Contrastive Inverse Dynamics for JEPA World Models

无需高斯：用于JEPA世界模型的对比逆动力学

Jack Boylan, Chris Hokamp

机构 * Quantexa（昆泰克萨公司）

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

AI总结该研究提出AC-MTM，用对比逆动力学替代JEPA世界模型的高斯型抗坍塌机制，在多目标视觉任务上性能优于SIGReg，且训练稳定无额外复杂组件

Comments 17 pages, 5 figures. Code: https://github.com/jackboyla/action-contrastive-jepa

2608.12959 2026-08-14 cs.LG cs.AI 新提交 94%

The Objective Is the Bottleneck: Latent World Models Encode What Their Planners Cannot Use

目标是瓶颈：潜在世界模型编码了其规划器无法使用的内容

Joyjeet Singh

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

AI总结该研究发现潜在世界模型的规划瓶颈在于规划器目标而非预测器，通过替换目标无需额外训练即可大幅提升长视野规划性能，揭示模型编码了规划器未利用的可达性信息。

Comments Follow-up to arXiv:2608.10145. All experiments run on a laptop CPU; no model was trained or fine-tuned. Code, checkpoints and every measurement: github.com/joyjeet-singh/tinylab

2607.00836 2026-08-13 cs.RO cs.AI cs.SY eess.SY 版本更新 94%

From World Models to World Action Models: A Concise Tutorial for Robotics

从世界模型到世界动作模型：面向机器人学的简明教程

Xiaoxiong Zhang, Xiong Zeng, Wei Zhang

专题命中通用世界模型：world model(title,abstract)；world models(title,abstract)；world model(title,abstract)；world models(title,abstract)

AI总结本教程提出世界模型作为动作条件预测模型的设计空间视图，分类为观测空间和状态空间模型，并引入世界动作模型，总结四种代表性范式，为具身预测与控制提供结构化分类。

Comments Project page: https://clearlab-sustech.github.io/WorldModelSurvey/

1. 通用世界模型 4308 篇

Can World Models Benefit VLMs for World Dynamics?

Training Agents Inside of Scalable World Models

FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Making

Humanoid World Models: Open World Foundation Models for Humanoid Robotics

WoMAP: World Models For Embodied Open-Vocabulary Object Localization

Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets

World Models in Artificial Intelligence: Sensing, Learning, and Reasoning Like a Child

Toward Stable World Models: Measuring and Addressing World Instability in Generative Environments

ACT-Bench: Towards Action Controllable World Models for Autonomous Driving

Exploring the Interplay Between Video Generation and World Models in Autonomous Driving: A Survey

Learning World Models for Unconstrained Goal Navigation

World Models for Autonomous Driving: An Initial Survey

The Essential Role of Causality in Foundation World Models for Embodied AI

Learning and Leveraging World Models in Visual Representation Learning

Slot Structured World Models

Neural World Models for Computer Vision

Co-Learning Empirical Games and World Models

Transformers are Sample-Efficient World Models

World Models and Predictive Coding for Cognitive and Developmental Robotics: Frontiers and Challenges

Iso-Dream: Isolating and Leveraging Noncontrollable Visual Dynamics in World Models

Curious Exploration via Structured World Models Yields Zero-Shot Object Manipulation

DayDreamer: World Models for Physical Robot Learning

Constrained Policy Optimization via Bayesian World Models

Augmented World Models Facilitate Zero-Shot Dynamics Generalization From a Single Offline Environment

An Irreducible Quantum Advantage in Aligning World Models with Reality

Decision-Metric Alignment in Latent World Models: Diagnostics and Action-Conditioned Objectives for MPC Planning

Low-Rank Dynamics-Effective Latent Carriers for Counterfactual Rollout in Learned World Models

No Gaussian Required: Contrastive Inverse Dynamics for JEPA World Models

The Objective Is the Bottleneck: Latent World Models Encode What Their Planners Cannot Use

From World Models to World Action Models: A Concise Tutorial for Robotics