We list detailed hyper-parameter settings here for reproducibility.

Implicit feedback is central to modern recommender systems but is inherently noisy, often impairing model training and degrading user experience. At scale, such noise can mislead learning processes, reducing both recommendation accuracy and platform value. Existing denoising strategies typically overlook the entity-specific nature of noise while introducing high computational costs and complex hyperparameter tuning. To address these challenges, we propose \textbf{EARD} (\textbf{E}ntity-\textbf{A}ware \textbf{R}eliability-\textbf{D}riven Denoising), a lightweight framework that shifts the focus from interaction-level signals to entity-level reliability. Motivated by the empirical observation that training loss correlates with noise, EARD quantifies user and item reliability via their average training losses as a proxy for reputation, and integrates these entity-level factors with interaction-level confidence. The framework is \textbf{model-agnostic}, \textbf{computationally efficient}, and requires \textbf{only two intuitive hyperparameters}. Extensive experiments across multiple datasets and backbone models demonstrate that EARD yields substantial improvements over state-of-the-art baselines (e.g., up to 27.01\% gain in NDCG@50), while incurring negligible additional computational cost. Comprehensive ablation studies and mechanism analyses further confirm EARD's robustness to hyperparameter choices and its practical scalability. These results highlight the importance of entity-aware reliability modeling for denoising implicit feedback and pave the way for more robust recommendation research.

--Federated recommender systems (FedRecSys) have emerged as a pivotal solution for privacy-aware recommendations, balancing growing demands for data security and personalized experiences. Current research efforts predominantly concentrate on adapting traditional recommendation architectures to federated environments, optimizing communication efficiency, and mitigating security vulnerabilities. However, user personalization modeling, which is essential for capturing heterogeneous preferences in this decentralized and non-IID data setting, remains underexplored. This survey addresses this gap by systematically exploring personalization in FedRecSys, charting its evolution from centralized paradigms to federated-specific innovations. We establish a foundational definition of person-alization in a federated setting, emphasizing personalized models as a critical solution for capturing fine-grained user preferences. The work critically examines the technical hurdles of building personalized FedRecSys and synthesizes promising methodologies to meet these challenges. As the first consolidated study in this domain, this survey serves as both a technical reference and a catalyst for advancing personalized FedRecSys research. A. Motivation Federated recommender systems (FedRecSys) [1]-[6] have burgeoned as a remarkable paradigm to promote privacy-preserving recommendation services. Besides, the distributed optimization pattern enables service providers to effectively harness the vast computational resources of various devices. This balance between performance and privacy protection makes FedRecSys an attractive research avenue with significant potential for edge AI development. Current research in FedRecSys primarily derives from the perspectives of RecSys and FL views. Chunxu Zhang, Zijian Zhang and Bo Y ang are with the College of Computer Science and Technology, Jilin University, Jilin, China (e-mail: zhangchunxu@jlu.edu.cn, Guodong Long and Zhiwei li are with the Australian AI Institute, Faculty of Engineering and IT, University of Technology Sydney, Sydney, Australia (e-mail: guodong.long@uts.edu.au, Honglei Zhang is with the School of Computer Science and Technology, Beijing Jiaotong University, Beijing, China (e-mail: hon-glei.zhang@bjtu.edu.cn). Qiang Y ang is Professor Emeritus at the Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Hong Kong, and the Chief AI Officer of WeBank, Shenzhen, China (e-mail: qyang@cse.ust.hk). Personalization technique comparison in centralized and federated RecSys. The colorful module denotes the user-specific parameters and the gray module represents the user-shared parameters. FL's ability to collaboratively train multiple models across different devices naturally supports the development of personalized models, making it easier to tailor recommendations to individual user needs.