The field of computational pathology is rapidly advancing, with a focus on developing innovative methods for analyzing medical images and improving diagnostic accuracy. Recent developments have centered around the use of deep learning techniques, such as self-supervised learning and knowledge distillation, to improve the performance of foundation models in pathology. These models have shown promise in capturing subspecialty-specific features and task adaptability, leading to significant advances in cervical pathology, breast cancer screening, and cancer prognosis analysis. Notably, the use of multi-magnification and prototype-aware architectures has improved the prediction of spatial gene expression, while anatomically aware self-supervision has enhanced the performance of mammographic image analysis. Furthermore, the development of causal learning frameworks has mitigated the impact of domain shift in histopathology image analysis, and progressive representation learning models have efficiently processed whole slide images for cancer prognosis. The introduction of vision foundation models for oral and maxillofacial radiology has also demonstrated robust generalization across diverse dental tasks. Some noteworthy papers in this area include: From Generic to Specialized: A Subspecialty Diagnostic System Powered by Self-Supervised Learning for Cervical Histopathology, which introduces a diagnostic system that surpasses prior foundation models in scope and clinical applicability. G2L:From Giga-Scale to Cancer-Specific Large-Scale Pathology Foundation Models via Knowledge Distillation, which presents a novel strategy to increase the performance of large-scale foundation models to a comparable level of giga-scale models. MMAP: A Multi-Magnification and Prototype-Aware Architecture for Predicting Spatial Gene Expression, which proposes a novel framework that addresses the challenges of predicting spatial gene expression from histological images. MammoDINO: Anatomically Aware Self-Supervision for Mammographic Images, which presents a novel SSL framework for mammography that achieves state-of-the-art performance on multiple breast cancer screening tasks. CLEAR: Causal Learning Framework For Robust Histopathology Tumor Detection Under Out-Of-Distribution Shifts, which proposes a novel causal-inference-based framework that leverages semantic features while mitigating the impact of confounders. DCMIL: A Progressive Representation Learning Model of Whole Slide Images for Cancer Prognosis Analysis, which proposes an easy-to-hard progressive representation learning model that efficiently processes WSIs for cancer prognosis. Towards Generalist Intelligence in Dentistry: Vision Foundation Models for Oral and Maxillofacial Radiology, which introduces the first family of vision foundation models designed for dentistry, demonstrating robust generalization to diverse dental tasks. WeCKD: Weakly-supervised Chained Distillation Network for Efficient Multimodal Medical Imaging, which presents the first-ever Weakly-supervised Chain-based KD network that redefines knowledge transfer through a structured sequence of interconnected models. Morphology-Aware Prognostic model for Five-Year Survival Prediction in Colorectal Cancer from H&E Whole Slide Images, which develops a novel, interpretable AI model that incorporates a continuous variability spectrum within each distinct morphology to characterize phenotypic diversity. A Multi-Task Deep Learning Framework for Skin Lesion Classification, ABCDE Feature Quantification, and Evolution Simulation, which proposes a deep learning framework that classifies skin lesions into categories and quantifies scores for each ABCD feature.