1 paper
Xiaochen Kev Gao, Feng Yao, Kewen Zhao +4
Model scaling is becoming the default choice for many language tasks due to the success of large language models (LLMs). However, it can fall short in specific scenarios where simp…