/M2IB

Code for the paper Visual Explanations of Image–Text Representations via Multi-Modal Information Bottleneck Attribution

Primary LanguageJupyter Notebook

Issues