ArXiv TLDR

Ahmed Arafa

2 papers · Latest:

Machine Learning

FedSIR: Spectral Client Identification and Relabeling for Federated Learning with Noisy Labels

FedSIR is a federated learning framework that uses spectral analysis to identify noisy clients and relabel corrupted samples, improving robustness.

2604.20825
Machine Learning

FB-NLL: A Feature-Based Approach to Tackle Noisy Labels in Personalized Federated Learning

FB-NLL is a feature-based personalized federated learning framework that robustly clusters users and corrects noisy labels by leveraging feature space geometry.

2604.19729

📬 Weekly AI Paper Digest

Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.