The field of federated learning is moving towards developing personalized and privacy-preserving models that can handle heterogeneous data distributions and varying computational capabilities. Recent works have focused on designing novel frameworks and algorithms that can adapt to different client-wise domain heterogeneity, including distribution diversity and class imbalance. These approaches aim to achieve stable and client-tailored adaptive modeling without sharing sensitive local data. Additionally, there is a growing interest in exploring the application of federated learning to real-world problems, such as energy forecasting and medical image segmentation. Noteworthy papers in this area include: Federated Client-tailored Adapter for Medical Image Segmentation, which proposes a novel framework for medical image segmentation that achieves stable and client-tailored adaptive segmentation without sharing sensitive local data. Privacy-Preserving Personalized Federated Learning for Distributed Photovoltaic Disaggregation under Statistical Heterogeneity, which proposes a two-level framework that combines local and global modeling for PV disaggregation. Robust Federated Personalised Mean Estimation for the Gaussian Mixture Model, which explores combining personalization for heterogeneous data with robustness to corrupted data. FedCCL: Federated Clustered Continual Learning Framework for Privacy-focused Energy Forecasting, which presents a framework specifically designed for environments with static organizational characteristics but dynamic client availability.