AI RESEARCH
HDMoE: A Hierarchical Decoupling-Fusion Mixture-of-Experts Framework for Multimodal Cancer Survival Prediction
arXiv CS.CV
•
ArXi:2605.20891v1 Announce Type: new Multimodal survival prediction, a crucial yet challenging task, demands the integration of multimodal medical data (\eg Whole Slide Images (WSIs) and Genomic Profiles) to achieve accurate prognostic modeling. Given the inherent heterogeneity across modalities, the feature decoupling-fusion paradigm has emerged as a dominant approach.