Enhancing Multimodal Entity and Relation Extraction with Variational Information Bottleneck

Open in new window