arXiv AI June 16, 2026 By Thinh T. H. Nguyen, Khoa D. Doan, Binh T. Nguyen, Danh Le-Phuoc, Kok-Seng Wong

When Generator Replay Degrades: Projected Rehearsal Orchestration for Heterogeneous Federated Class-Incremental Learning

arXiv:2606. 15695v1 Announce Type: cross Abstract: Federated class-incremental learning (FCIL) becomes substantially harder when clients observe different label subsets, progress through tasks at different stages, and provide uneven supervision for the same semantic concepts.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv Machine Learning

Jun 5

Revisiting Prototype Rehearsal for Exemplar-Free Continual Learning: Manifold-Aware Boundary Sampling with Adaptive Class-Balanced Loss

arXiv:2606. 05695v1 Announce Type: new Abstract: Exemplar-free class-incremental learning (EFCIL) aims to acquire new classes over time without storing raw data.

By Hongye Xu, Bartosz Krawczyk

benchmarks

arXiv Machine Learning

Jul 20

Rethinking Transfer in Continual Learning: A Replay-Based Realisation

arXiv:2607. 15587v1 Announce Type: new Abstract: Continual learning studies how deployed language models can continually acquire new tasks without expensive retraining from scratch.

By Yang Meng, Zhenya Liu, Zhuokai Zhao, Yuxin Chen

llms efficiency

arXiv AI

Jun 3

FlashbackCL: Mitigating Temporal Forgetting in Federated Learning

arXiv:2606. 03939v1 Announce Type: cross Abstract: Federated Learning (FL) of foundation and edge models increasingly targets deployments where client data distributions drift over time, yet existing forgetting-mitigation methods assume each client's distribution is stationary.

By Mubarak A. Ojewale, Adriana E. Chis, Jorge M. Cortes-Mendoza, Bernardo Pulido-Gaytan, Horacio Gonzalez-Velez

efficiency benchmarks

Hugging Face Trending Papers

Jun 2

FlashbackCL: Mitigating Temporal Forgetting in Federated Learning

Federated Learning (FL) of foundation and edge models increasingly targets deployments where client data distributions drift over time, yet existing forgetting-mitigation methods assume each client's distribution is stationary. Flashback, the strongest recent FL method against cross-client (spatial) forgetting, uses monotonically accumulating per-class label counts as a knowledge proxy; this proxy becomes miscalibrated under temporal distribution shift and anchors the global model to an outdated class balance.

efficiency benchmarks

arXiv Machine Learning

Jun 2

Dynamic Proxy-Mixing: Transferring Replay Controllers from Small to Large Models for Continual Instruction Tuning

arXiv:2606. 00400v1 Announce Type: new Abstract: Continual instruction tuning updates a language model through a sequence of new domains, yet each update can progressively erode previously learned capabilities and alignment behavior.

By Ibne Farabi Shihab, Fariya Afrin, Anuj Sharma

llms reinforcement-learning fine-tuning safety

arXiv Machine Learning

Jul 17

Knowledge-Aware Evolution for Task-Free Streaming Federated Continual Learning with Arbitrary Class Overlap

arXiv:2601. 19788v2 Announce Type: replace Abstract: Federated Continual Learning (FCL) leverages inter-client collaboration to better balance new knowledge acquisition and old knowledge retention on non-stationary data.

By Sixing Tan, Xianmin Liu

Related stories

Revisiting Prototype Rehearsal for Exemplar-Free Continual Learning: Manifold-Aware Boundary Sampling with Adaptive Class-Balanced Loss

Rethinking Transfer in Continual Learning: A Replay-Based Realisation

FlashbackCL: Mitigating Temporal Forgetting in Federated Learning

FlashbackCL: Mitigating Temporal Forgetting in Federated Learning

Dynamic Proxy-Mixing: Transferring Replay Controllers from Small to Large Models for Continual Instruction Tuning

Knowledge-Aware Evolution for Task-Free Streaming Federated Continual Learning with Arbitrary Class Overlap