Large language models as grading assistants in public health education: a method-comparison study of essay-style exam assessment
Some LLMs, particularly ChatGPT and Kimi, showed moderate alignment with human examiners, but the alignemnet was not consistently stable across repeated sessions or operational settings.
Asgeir Brevik, H. Jerpseth, S. Lafontan
· Frontiers in Education · 0 citations