Preliminary Systematic Review of Open-Source Large Language Models in Education

This work-in-progress study aims to explore and analyze the growing impact of large language models (LLMs) in the fields of education and industry. We preliminarily review how LLMs can be integrated into educational contexts with their technical features, open-source nature, and applicability. Through a systematic search, we have identified a selection of open-source LLMs that have been released or significantly updated post-2021. This initial search indicates a thriving field with immense potential for both academic and industry applications. While LLMs hold promise for education, some challenges need to be addressed. These include limited application of open-source LLMs, concerns regarding data privacy, content accuracy, and potential biases. It is critical to carefully consider these factors before deploying LLMs in educational settings. However, our preliminary research highlights the versatility of LLMs in generating educational content and supporting diverse instructional strategies. This suggests a shift towards more adaptive and personalized learning environments. By assessing the suitability of these models for educational purposes, our study lays the foundation for future research aimed at fully maximizing the potential of open-source LLMs to transform teaching and learning practices. As our work progresses, we plan to expand our investigation to explore the broader implications of LLMs on educational outcomes and pedagogical contexts. Ultimately, our goal is to facilitate dynamic, inclusive, and effective learning experiences across various educational environments.