Author name: Kang Liu, Zhuoqi Ma, Siyu Liang, Yunan Li, Xiyue Gao, Chao Liang, Kun Xie, Qiguang Miao

arXiv:2603.26049v1 Announce Type: new
Abstract: Despite recent advances in medical vision-language pretraining, existing models still struggle to capture the diagnostic workflow: radiographs are typically treated as context-agnostic images, while radi…