A Two-stage Fine-tuning Strategy for Generalizable Manipulation Skill of Embodied AI

Open in new window