Enhancing Multimodal Entity and Relation Extraction with Variational Information Bottleneck

Shiyao Cui, Jiangxia Cao, Xin Cong +4

This paper studies the multimodal named entity recognition (MNER) and multimodal relation extraction (MRE), which are important for multimedia social platform analysis. The core of…