Academics | The Hong Kong University of Science and Technology

O ur research is focused on constructing a comprehensive benchmark to evaluate the performance of multimodal large language models (MLLMs) in breast cancer tasks. This benchmark is designed to assess the model's ability to handle a variety of critical tasks, including but not limited to breast-cancer histological subtyping, breast-cancer histological grading, necrosis recognition, and calcification recognition. By establishing this benchmark, we aim to provide a standardized framework for evaluating how well advanced multimodal models can interpret pathology images and textual data in the context of breast cancer diagnosis and treatment.