Fig. 1From: Dual modality prompt learning for visual question-grounded answering in robotic surgeryOverall architecture of the proposed DMPL. a Dual modality prompt learning (DMPL) network; b VCP; c TCPBack to article page