Language Model Evaluation

C-Eval is a Chinese foundational model evaluation suite jointly developed by Shanghai Jiao Tong University, Tsinghua University, and the University of Edinburgh. It comprises 13,948 multiple-choice questions across 52 disciplines and four difficulty levels, aiming to comprehensively assess large language models' Chinese comprehension and reasoning abilities.

MMLU (Massive Multitask Language Understanding) is a benchmark test launched by the University of California, Berkeley in September 2020, aiming to comprehensively evaluate large language models' multitask understanding across 57 different domains.

HELM (Holistic Evaluation of Language Models) is a comprehensive evaluation system for language models introduced by Stanford University, aiming to assess the performance and characteristics of language models through standardized datasets, unified model interfaces, and multidimensional evaluation metrics.