Three Questions Concerning the Use of Large Language Models to Facilitate Mathematics Learning
Submission Type: Regular Short Paper
Submission Track: Theme Track: Large Language Models and the Future of NLP
Keywords: Mathematics Learning, Adaptive Feedback, Large Language Models
TL;DR: This work formulates three research questions investigating the pedagogical ability of large language models to help students learn mathematics.
Abstract: Due to the remarkable language understanding and generation abilities of large language models (LLMs), their use in educational applications has been explored. However, little work has been done on investigating the pedagogical ability of LLMs in helping students to learn mathematics. In this position paper, we discuss the challenges associated with employing LLMs to enhance students' mathematical problem-solving skills by providing adaptive feedback. Apart from generating the wrong reasoning processes, LLMs can misinterpret the meaning of the question, and also exhibit difficulty in understanding the given questions' rationales when attempting to correct students' answers. Three research questions are formulated.
Submission Number: 889
Loading