Cross-modal Representation Flattening for Multi-modal Domain Generalization | Read Paper on Bytez