The field of medical image segmentation is moving towards more generalizable and adaptable models, with a focus on reducing the need for extensive annotations and improving performance across diverse tasks. This is driven by the development of universal medical image segmentation models, which have shown strong potential for a wide range of clinical applications. Noteworthy papers in this regard include MedSAMix, which proposes a training-free model merging approach that integrates the strengths of generalist and specialist models, and SelfAdapt, which enables the adaptation of pre-trained cell segmentation models without the need for labels. Additionally, CoFi introduces a fast and efficient coarse-to-fine few-shot segmentation pipeline for glomerular basement membrane segmentation, while LGMSNet proposes a novel lightweight framework that achieves state-of-the-art performance with minimal computational overhead. The molecular-empowered All-in-SAM Model also advances computational pathology by leveraging the capabilities of vision foundation models. These innovative approaches are expected to have a significant impact on the field, enabling more accurate and reliable medical image analysis and diagnosis.