70 citations · 205 across the 31 of their papers we have counts for
1 paper · 2 filters
Zhenheng Tang, Yuxin Wang, Xin He +8
The rapid growth of memory and computation requirements of large language models (LLMs) has outpaced the development of hardware, hindering people who lack large-scale high-end GPU…