Using Large Language Model for Assessing Business English Students’ Pragmatic Competence

Pragmatic competence is the the ability to say and do things appropriately in various contexts. Scholars have been studied this topic from different aspects and a large quantities of methods have been attest to assess students’ pragmatic competence. However, human rating is not only laborious but also less efficient. The emergency of Artificial Intelligence (AI), particularly Large Language Model shed a light on this matter. This essay put prompts into kimi—a LLM of China to see if it can evaluate Business English major students’ pragmatic competence and compare its scoring with human rater's. It is found that LLM's core shows consistency with human score and LLM can give instant and personalized feedback which is impossible for human. Therefore, the reliability of LLM evaluation on pragmatic competence is proved.