PWM: Policy Learning with Large World Models

Publication
ICLR 2025