1 citations · 1 across the 4 of their papers we have counts for
1 paper · 1 filter
Huanyu Wang, Ziyu Xia, Zhuoming Chen +1
Large language model (LLM) services are mostly centralized, leading to scalability bottlenecks and underutilization of substantial scattered GPU resources. While decentralization o…