Abstract: Language-guided robotic grasping in cluttered environments presents significant challenges due to severe occlusions and complex scene structures, which often hinder accurate target ...