1 paper
Sihan Chen, Dan Zhao, Jongwoo Ko +5
The growing computational demands of large language models (LLMs) make efficient inference and activation strategies increasingly critical. While recent approaches, such as Mixture…