کنفرانس بین المللی یادگیری و یاددهی الکترونیکی ایران

صفحه اصلی / نوزدهمین کنفرانس ملی و سیزدهمین کنفرانس بین المللی یادگیری و یاددهی الکترونیکی ایران

Applications of Large Language Models in Collaborative Learning: A Systematic review

نویسندگان :

Soraya Shokry pour dehghan¹ Maedeh Mosharraf²

1- دانشگاه شهید بهشتی تهران 2- دانشگاه شهید بهشتی تهران

کلمات کلیدی :

Collaborative Learning،Generative AI،ollaborative problem-solving،LLM،Education

چکیده :

Abstract—The integration of Large Language Models (LLMs) into collaborative learning presents a transformative potential to move beyond the limitations of traditional digital tools by acting as intelligent, context-aware facilitators. This paper presents a systematic review of recent applications of LLMs in collaborative learning environments. Following PRISMA guidelines, 14 studies were identified from IEEE Xplore and categorized into five domains: idea generation, interaction analysis, Artificial Intelligence (AI) moderation, feedback assistance, and role simulation. Results show that LLMs can enhance creativity, support interaction monitoring, provide adaptive facilitation, refine feedback, and simulate expert roles. The synthesis further reveals critical challenges, such as risks of over-reliance, contextual misinterpretation, and limited generalizability. Consequently, this review provides a foundational synthesis to guide future research and design towards realizing the reliable and scalable potential of LLMs in collaborative learning.