1 paper
Bin Wang, Zhengyuan Liu, Xin Huang +4
We present SeaEval, a benchmark for multilingual foundation models. In addition to characterizing how these models understand and reason with natural language, we also investigate…