ARB4WM: An Adversarial Robustness Benchmark for World Models in Continuous Control
2026-06-15 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors created a testing system called ARB4WM to check how well world-model agents used in robots can handle tricky, unexpected visual changes. They looked at different ways these changes can mess with the agents' decisions, internal understanding, and predictions over time. Their tests on several agents showed that attacks on parts like value estimation or hidden states can be just as harmful as attacks that directly confuse actions. They also found that early or frequent attacks cause more problems and simple defenses don’t work well against smart attacks. The authors suggest that safety checks should test many parts of the agent, not just its actions.
world modelsrobotics controllatent dynamicsvisual perturbationsadversarial attackspolicy disruptionvalue estimationDreamer agentsMetaWorldDeepMind Control Suite
Authors
Junjian Zhang, Hao Tan, Ruonan Li, Dong Zhu, Aiping Li, Zhaoquan Gu
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. As these systems are increasingly deployed in safety-critical settings, understanding their robustness under adversarial conditions has become essential. However, existing evaluations lack a unified benchmark for testing adversarial threats across the policy, value, and latent-dynamics levels of world-model agents. To fill this gap, we present ARB4WM, a unified evaluation framework for pre-deployment robustness and risk assessment of world-model agents under visual perturbations. ARB4WM defines five white-box loss objectives across these three levels and studies their effects when combined with single-step or multi-step perturbation strategies and temporal attack modes, including full-frame, half-sequence, and sparse-frame exposure. Specifically, we evaluate four Dreamer-style agents across 20 tasks from MetaWorld and the DeepMind Control Suite under different loss objectives, perturbation strategies, and temporal attack modes. Results show that attacks targeting value estimation, latent representations, and RSSM dynamics can be as damaging as direct policy disruption, and that early or frequent perturbations are especially harmful, while input-level defenses provide limited recovery under adaptive attacks. These findings suggest that safety, risk, and reliability assessment for world models should cover multiple component-oriented attack objectives and temporal exposure protocols rather than relying solely on action-space robustness. Source code is available at https://github.com/zaoanguai/ARB4WM.