FDA-Opt: Federated Fine-Tuning via Dynamic Update Schedules

Excerpt

arXiv:2505.04535v4 Announce Type: replace Abstract: Federated Learning (FL) enables the utilization of vast, previously inaccessible data sources. At the same time, pre-trained Language Models (LMs) have taken the world by storm and for good reason. They exhibit remarkable emergent abilities and are readily adapted to downstream tasks. This opens one of the most exciting frontiers in FL: fine-tuning LMs. Yet, a persistent challenge in FL is the frequent, rigid communication of parameters -- a pr