Scaling the Authoring of AutoTutors with Large Language Models

The paper discusses the use of Large Language Models (LLMs) in education, particularly in the authoring of Intelligent Tutoring Systems. Although LLMs have potential, they often deviate from desired pedagogical strategies, like providing direct answers to students. The authors propose a tutoring system named MWPTutor, which uses LLMs to fill in the state space of a pre-defined finite state transducer. This system maintains the structure and pedagogy of traditional tutoring systems but adds the flexibility of LLM-based approaches.

Publication date: 15 Feb 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2402.09216