output
20122026
most citedAssessing Student Errors in Experimentation Using Artificial Intelligence and Large Language Models: A Comparative Study with Human Raters

70 citations

Showing cs.AIShow all

2 papers · 1 filter