Jul 2026
FedWeave: Rethinking the Unit of Specialization in Heterogeneous Federated MoE-LoRA
The theoretical analysis explains why asymmetric aggregation is advantageous: it controls expert convergence in stationarity through off-pattern contamination, identifies the consensus error induced by fragmented router trajectories, and bounds sparse-inference risk.
Donghang Duan, Xu Zheng, Li-Zong Zhang et al.
· arXiv.org · 0 citations