Adversarial Attacks on World Models

Latest papers

17 papers2 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Across Viewpoints and Time: Trajectory-Generalizable Physical Attacks on World Models and Video Understanding Models
Oct 11, 2026Song Xia, Meiwen Ding, Chenqi Kong +2Adversarial Attacks Physical Adversarial Attacks
Nanyang Technological University · National University of Singapore · Pengcheng Lab
TAPDreamer: Transferable Adversarial Patches for World Action Models
Oct 5, 2026Xuanyu Lu, Fengqing Jiang, Kaiyuan Zheng +8Adversarial Patch Attacks Adversarial Attacks
University of Washington · University of Georgia · Western Washington University +2
When the World Lies: Backdoor Attacks on Latent World Models for Downstream Control
Sep 14, 2026Roberto Riaño, Gorka Abad, Stjepan Picek +1Backdoor Attacks Latent World Models
Radboud University, The Netherlands · IKERLAN Technology Research Centre, Spain · University of Bergen, Norway +1
TrojanWorld: Backdooring World-Model Agents via Imagination Steering
Sep 7, 2026Wenkai Huang, Siyuan Liang, Gaolei Li +4Backdoor Attacks Adversarial Attacks on World Models
Shanghai Jiao Tong University · Nanyang Technological University
Where World Models Break: Natural-Input Failure Discovery
Aug 23, 2026Zhanpeng Shi, Zi Liang, Rong Feng +3Adversarial Attacks on World Models Uncertainty-Aware Exploration
School of Computer Science, Northwestern Polytechnical University · The Hong Kong Polytechnic University · Department of Data Science, City University of Hong Kong +3
Security of World-Model-Based Embodied AI: A Lifecycle of Threats, Defenses, and Evaluation
Jul 30, 2026Fazhong Liu, Zhuoyan Chen, Haozhen Tan +3AI Agent Security World Model Evaluation
False Prophets: On the Security of World Models in Agentic Systems
Jul 25, 2026Erik Imgrund, Anna Wimbauer, Klim Kireev +1Adversarial Attacks on World Models Adversarial Attacks
BIFOLD & TU Berlin Berlin, Germany
BadWAM: When World-Action Models Dream Right but Act Wrong
Jul 16, 2026Qi Li, Xingyi Yang, Xinchao WangWorld Models for Robotics Adversarial Attacks
National University of Singapore · The Hong Kong Polytechnic University
Attacking the Trusted Imagination: Oracle-Level Integrity Attacks on Imagine-then-Act World Models
Jun 22, 2026Linghan Chen, Kaiyan Ji, Minyu GuoAdversarial Attacks on World Models Adversarial Attacks
Adelaide University
Stealthy World Model Manipulation via Data Poisoning
Jun 17, 2026Yibin Hu, Xiaolin Sun, Zizhan ZhengWorld Model Learning Adversarial Attacks on World Models
Department of Computer Science Tulane University New Orleans, LA, USA
ARB4WM: An Adversarial Robustness Benchmark for World Models in Continuous Control
Jun 15, 2026Junjian Zhang, Hao Tan, Ruonan Li +3World Models for Robotics Adversarial Robustness
College of Computer Science, National University of Defense Technology, Changsha, 410073, China · National Key Laboratory of Advanced Communication Networks, Shijiazhuang, 050081, Hebei, China · College of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, 518055, China +1
BadWorld: Adversarial Attacks on World Models
Jun 15, 2026Linghui Shen, Mingyue Cui, Xingyi YangAdversarial Attacks on World Models Adversarial Attacks
The Hong Kong Polytechnic University
Targeting World Models to Compromise Robot Learning Pipelines
Jun 8, 2026Ethan Rathbun, Ahmed Agha, Saaduddin Mahmud +3Robot Policy Learning World Models for Robotics
Northeastern University · University of Massachusetts Amherst
WMAttack: Automated Attack Search for Adversarial Evaluation of World-Model Agents
May 22, 2026Zhixiang Guo, Siyuan Liang, Shi Fu +4Adversarial Attacks on World Models Adversarial Attacks
Nanyang Technological University · University of Szeged
TRAP: Tail-aware Ranking Attack for World-Model Planning
May 3, 2026Siyuan Duan, Ke Zhang, Xizhao LuoBackdoor Attacks World Model-Based Planning
Department of Computer Science and Technology, Soochow University Suzhou, China
Safety, Security, and Cognitive Risks in World Models
Apr 1, 2026Manoj ParmarAdversarial Attacks on World Models World Models
SovereignAI Security Labs, Bengaluru, India
When World Models Dream Wrong: Physical-Conditioned Adversarial Attacks against World Models
Feb 21, 2026Zhixiang Guo, Siyuan Liang, Andras Balogh +4Adversarial Attacks on World Models Adversarial Attacks
Nanyang Technological University · University of Szeged

Oct 11, 2026J/K move · Enter open · S save

Across Viewpoints and Time: Trajectory-Generalizable Physical Attacks on World Models and Video Understanding Models

Song Xia, Meiwen Ding, Chenqi Kong, Wenhan Yang+1

Nanyang Technological University · National University of Singapore · Pengcheng Lab

Adversarial Attacks Physical Adversarial Attacks Adversarial Attacks on World Models Consistent Adversarial Attacks Video World Models

Open paper AskPDF arXiv

World models and video understanding models are increasingly deployed in dynamic visual environments, yet their robustness to physical adversarial threats remains largely underexplored. We study whether a single physical adversarial texture can mislead the victim model across different real-world scenarios where observing viewpoints vary along uncontrolled object--camera trajectories. We identify that heterogeneous video semantics and trajectories induce poorly aligned gradients for optimizing the adversarial texture. To address that, we propose the Trajectory-aware adversarial Alignment and Refinement Attack (TARA), which incorporates trajectory-aware optimization to capture the evolving geometric variations of a printed texture under real-world scenarios. To improve cross-instance optimization compatibility, TARA incorporates trajectory-aware contrastive alignment to encourage attacked representations from diverse video--trajectory instances to approach a shared adversarial region and then performs trajectory-aware task refinement to translate the aligned representation into task-level prediction errors. For reproducible evaluation, we develop a geometry-grounded tracking pipeline that provides physically plausible adversarial regions with temporally consistent trajectories on EPIC-KITCHENS-100 and Something-Something V2. We further collect real-world print-and-recapture videos to validate attack effectiveness under physical deployment conditions. Together, the annotated benchmarks and real-world captured data form a reproducible testbed for future evaluation. Notably, our results reveal that a single printable adversarial texture can persistently disrupt world models and video understanding models under physical conditions, highlighting the fundamental security challenge.