TAPDreamer: Transferable Adversarial Patches for World Action Models
[Widzenie komputerowe] Autorzy: Xuanyu Lu, Fengqing Jiang, Kaiyuan Zheng i 8 więcej
World models learn to predict how their environment will evolve, making them an important foundation for general-purpose robotic control. Yet world action models depend on camera inputs whose manipulation can corrupt the visual representations used across tasks and action policies. Existing attacks on these models optimize against the victim's actions or predicted futures and therefore require access to target-model outputs. In this paper, we propose an attack, TAPDreamer, against world action models that instead uses a public encoder alone to construct a fixed local perturbation that transfers across tasks and action architectures. TAPDreamer requires no target-policy queries. Our key insight is that interactions between patch-induced changes in attention weights and value vectors broadcast a nearly identical representation shift far beyond the patch footprint, and this shift remains s…
- Kategoria arxiv
- cs.CV
- Kategoria
- Widzenie komputerowe