Orthogonalized Multimodal Contrastive Learning with Asymmetric Masking for Structured Representations
正交化多模态对比学习与不对称掩码用于结构化表示
机构 * Peter L. Reichertz Institute for Medical Informatics of TU Braunschweig and Hannover Medical School(图林根工业大学和汉诺威医学院医学信息学研究所) ; Lower Saxony Center for AI and Causal Methods in Medicine (CAIMed)(下萨克森人工智能与因果医学中心(CAIMed))
专题命中 跨模态检索 :multimodal(title,abstract);cross-modal(abstract)
AI总结 COrAL通过正交化多模态对比学习与不对称掩码,显式保留冗余、独特和协同信息,提升多模态表示的稳定性与全面性。