上一条： CISum: Learning Cross-modality Interaction to Enhance Multimodal Semantic Coverage for Multimodal Summarization

下一条： Multilevel Attention Networks and Policy Reinforcement Learning for Image Caption Generation