arXiv Machine Learning September 10, 2026 By Wenkai Huang, Siyuan Liang, Gaolei Li, Yiming Li, Tianhao Peng, Jianhua Li, Dacheng Tao

TrojanWorld: Backdooring World-Model Agents via Imagination Steering

Read the original on arXiv Machine Learning →

TrojanWorld is a backdoor framework that targets world-model agents by steering their internal imagination toward attacker-specified actions when a physical trigger is present. The attack uses Decision-Reflective Induction, Clean Behavior Anchoring, and Causal Propagation to maintain stealth, persistence, and high performance. Experiments on TD-MPC2, DreamerV3, and R2-Dreamer across several benchmarks show that the attack can induce target actions with minimal performance loss and can keep agents on a malicious trajectory even after the trigger is removed.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Hugging Face Trending Papers

Jul 30

Security of World-Model-Based Embodied AI: A Lifecycle of Threats, Defenses, and Evaluation

World models give embodied AI a predictive core: they compress observations into states, simulate action-conditioned futures, and enable planning beyond reactive control. This predictive layer, however, opens a new security boundary-compromise can propagate from data, sensors, prompts, or feedback into physical action.

robotics safety

arXiv Machine Learning

Jul 17

BadWAM: When World-Action Models Dream Right but Act Wrong

arXiv:2607. 15207v1 Announce Type: new Abstract: World-action models (WAMs) are emerging as a promising foundation for embodied control: rather than predicting actions alone, they learn representations that couple action generation with future world prediction.

By Qi Li, Xingyi Yang, Xinchao Wang

robotics safety

arXiv AI

Sep 15

When the World Lies: Backdoor Attacks on Latent World Models for Downstream Control

arXiv:2609.15781v1 Announce Type: cross Abstract: Pretrained world models, learned simulators that encode an observation into a latent state and predict how it evolves under actions, are beginning to...

By Roberto Ria\~no, Gorka Abad, Stjepan Picek, Aitor Urbieta

llms fine-tuning

arXiv AI

Jun 16

ARB4WM: An Adversarial Robustness Benchmark for World Models in Continuous Control

arXiv:2606. 16605v1 Announce Type: new Abstract: World models are widely used in robotic and agentic engineering control systems due to their ability to learn latent dynamics for planning and decision-making.

By Junjian Zhang, Hao Tan, Ruonan Li, Dong Zhu, Aiping Li, Zhaoquan Gu

agents robotics benchmarks safety

arXiv Machine Learning

Jul 30

AgentSnare: Learning to Delay, Divert, and Defuse Autonomous Penetration Agents

arXiv:2607. 26998v1 Announce Type: cross Abstract: Large language model (LLM) agents automate penetration testing through an observation-action loop, selecting actions based on observations returned by tools.

By Ruoyu Wang, Heng Zhao, Renjie Wu, Mengnan Zhao, Zhixuan Chu, Wanyu Lin, Tianhang Zheng

llms agents

Hugging Face Trending Papers

Jul 22

Defense Against LLM Backdoors using Critical Neuron Isolation Pruning

Large language models (LLMs) are vulnerable to backdoor attacks, where hidden triggers induce malicious outputs. Existing defenses generally fall into inference-time detection or training-time mitigation, but face two key limitations.

llms fine-tuning efficiency multimodal benchmarks