PRISM: Personalized Refinement of Imitation Skills for Manipulation via Human Instructions

ArXi:2603.05574v1 Announce Type: cross This paper presents PRISM: an instruction-conditioned refinement method for imitation policies in robotic manipulation. This approach bridges Imitation Learning (IL) and Reinforcement Learning (RL) frameworks into a seamless pipeline, such that an imitation policy on a broad generic task, generated from a set of user-guided nstrations, can be refined through reinforcement to generate new unseen fine-grain behaviours.