World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry 事件

World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry arXiv:2604.01985v2 Announce Type: replace-cross Abstract: General-purpose world models promise scalable policy evaluation, optimization, and planning, yet achieving the required level of robustness remains challenging. Unlike policy learning which primarily focuses on optimal actions, a world model needs to be reliable over a vast space of suboptimal actions, which are often underrepresented in action-labeled robot