China
Inner Mongolia University of Science and Technology
PET/CT lung cancer lesion images, dual encoder, semisupervised learning, SwinUNet segmentation model
Person re-identification · Modality entanglement · Disentanglement representation · Separate learning
Medical visual question answering, large language models, multimodal feature alignment, dynamic knowledge routing, knowledge-infused dual-attention
person re-identification, cross-modality, meta-learning, heterogeneous generalization
