The field is witnessing a significant shift towards incorporating temporal modeling and attention mechanisms to effectively capture complex patterns in electronic health records (EHRs) and time-series data. This is evident in the development of novel frameworks that integrate heterogeneous clinical notes, chest X-ray imaging, and high-frequency clinical data to predict patient outcomes and trajectories. Noteworthy papers include DENSE, which leverages a clinically informed retrieval strategy to generate temporally aware progress notes, and CXR-TFT, which predicts chest X-ray trajectories in critically ill patients. TALE-EHR and TIDSIT also demonstrate innovative approaches to modeling EHRs and estimating battery state of health using time-aware attention mechanisms and continuous time embeddings.