PGCL: Prompt guidance and self-supervised contrastive learning-based method for Visual Question Answering

Ling Gao, Hongda Zhang, Yiming Liu, Nan Sheng, Haotian Feng, Hao Xu

Abstract: Highlights•We propose a novel visual question answering model for science curricula.•This method guides the mining of information according to the constructed prompts.•This method also enhances the fusion of information by contrastive learning.•Extensive experiments demonstrate the superiority of our approach.