Large language models for objective and fair assessment of coherence, argumentation, and writing style in English language teaching.

Abstract

In recent years, the rapid advancement of artificial intelligence (AI), particularly Large Language Models (LLMs), has created new opportunities in English language teaching, especially in the assessment of writing skills. Beyond identifying grammatical and spelling errors, LLMs such as GPT, Claude, Gemini, LLaMA, and ChatGPT are now capable of conducting in-depth analyses of textual structure, coherence, argumentation, and writing style. This study investigates the potential application of LLMs in evaluating English essays with the aim of enhancing fairness, objectivity, and efficiency in digital learning environments. By examining their analytical capabilities and pedagogical implications, the research contributes to ongoing discussions on integrating AI-driven tools into English language assessment practices.