2 papers
cs.LG2026
Layer-wise Update Aggregation with Recycling for Communication-Efficient Federated Learning
Jisoo Kim, Sungmin Kang, Sunwoo Lee
Expensive communication cost is a common performance bottleneck in Federated Learning (FL), which makes it less appealing in real-world applications. Many communication-efficient F…
cs.LG2025
GEM: A Scale-Aware and Distribution-Sensitive Sparse Fine-Tuning Framework for Effective Downstream Adaptation
Sungmin Kang, Jisoo Kim, Salman Avestimehr +1
Parameter-efficient fine-tuning (PEFT) has become a popular way to adapt large pre-trained models to new tasks. Most PEFT methods update only a small subset of parameters while fre…