Next-Generation Educational AI Using Retrieval-Augmented Generation: Research Trends and Open Problems

DOI

10.2991/978-94-6239-791-0_6How to use a DOI?

Keywords

Retrieval-Augmented Generation (RAG); Artificial Intelligence in Education (AIEd); Multimodal Learning Systems; Publication Trends

Abstract

The present study is a bibliometric evaluation and an extensive analysis of the literature that has emerged concerning retrieval augmented generation as an educational tool in AI as published in Scopus from 2021 to 2025. The purpose of the analysis is to assess the researchers’ potential in determining the flaws of today’s educational system and improve learners’ outcomes. We therefore analyzed more than 4500 research papers published in Scopus and identified more than 3400 keywords relevant to the topic of education. Furthermore, we attempted to find out how education is changing in relation to AI based on frequently used terms like LLM (1,338), learning (5,216), education AI (3,995), and RAG (324) and reached the conclusion that specific terms used in such applications as grading and tutoring do not receive much attention nowadays. The introduction of RAG-based Measurable Multimodal Educational Framework (MMEF) aimed at combining ethical principles, relation to teaching methods, and incorporated retrieval systems helped us clearly define the fields of study that have not been exhausted. The main part of the research incorporates three sections dealing with publishing trends, surveys, and keywords analysis. The research into retrieval augmented generation allows for educational interpretation of the ideas of LLM and refers to the basic methods of vector database technology and multimodal LLM system to demonstrate AI concepts used in educational domain. The research mainly focuses on revolutionary trends in “AI in education”; technology based learning represents the primary area of study.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ankita Amburle
AU  - Ashok Kanthe
PY  - 2026
DA  - 2026/10/09
TI  - Next-Generation Educational AI Using Retrieval-Augmented Generation: Research Trends and Open Problems
BT  - Proceedings of the International Conference on Artificial Intelligence and Machine Learning for Agro-Ecosystems (ICAIML-AE 2026)
PB  - Atlantis Press
SP  - 50
EP  - 63
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-791-0_6
DO  - 10.2991/978-94-6239-791-0_6
ID  - Amburle2026
ER  -