paper
FlashbackCL: Mitigating Temporal Forgetting in Federated Learning
paperactiveprovisional
flashbackcl-mitigating-temporal-forgetting-in-federated-learning-df2db443·1 events·first seen 14d agoAliases: FlashbackCL: Mitigating Temporal Forgetting in Federated Learning
Co-occurring entities
More like this (12)
Attention Amnesia in Hybrid LLMs: When CoT Fine-Tuning Breaks Long-Range Recall, and How to Fix ItFederated LearningYour Privacy My Cloak: Backdoor Attacks on Differentially Private Federated LearningFlashbackCLRecalling Too Well: Sycophancy Evaluation and Mitigation in Memory-Augmented ModelsHeterogeneous Differential Privacy Federated LearningRange Penalization: Theoretical Insights with Applications in Federated LearningSparse Subspace-to-Expert Sharing for Task-Agnostic Continual LearningLanguage Models Need Sleep: Learning to Self-Modify and Consolidate MemoriesTailLoR: Protecting Principal Components in Parameter-Efficient Continual LearningLearning from the Self-future: On-policy Self-distillation for dLLMsSupervised Memory Training
Recent events (1)
FlashbackCL extends federated learning to mitigate temporal distribution shift and forgetting
FlashbackCL is a proposed extension to the Flashback federated learning method that addresses temporal forgetting — the degradation caused by client data distributions drifting over time, a scenario existing FL methods do not handle. The approach introduces temporally-decayed label counts, a device-aware replay buffer with Class-Balanced Reservoir Sampling, and server-side coreset curation. On CIFAR-10 with 50 clients, FlashbackCL achieves 6.9–10.0% relative improvement over Flashback while reducing temporal forgetting by up to 68%, with CBRS replay identified as the critical component.