AI RESEARCH

TRAP: Tail-aware Ranking Attack for World-Model Planning

arXiv CS.LG

ArXi:2605.01950v1 Announce Type: new World models enable long-horizon planning by internally generating and evaluating imagined trajectories, making them a promising foundation for generalist agents. However, this imagination-driven decision process also To exploit this vulnerability, we propose TRAP, a backdoor attack framework for world models that targets imagined trajectory ranking.