Correlation Information Bottleneck: Towards Adapting Pretrained Multimodal Models for Robust Visual Question Answering

Re-assign community