EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Retrieval-augmented generation for pedagogically aware educational AI: an expert-rated comparison of a prompt-only LLM tutor and an integrated, learner-state-aware RAG tutor

Jude A. Adenuga

Frontiers in Education · 17.08.2026 · 11

DOI: 10.3389/feduc.2026.1896839 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Education, online erschienen am 17.08.2026. Verfasst von Jude A. Adenuga. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Large language models can produce fluent tutoring dialogue, but their educational use remains limited by weak grounding, uneven pedagogical control, and the risk of unsupported content. This study compared two controlled algebra tutoring workflows built on the same foundation model, Gemini 2.5 Flash.

Methods

The first condition was a prompt-only tutor that used the immediate task and scripted follow-up only. The second was an integrated pedagogical RAG tutor that combined a curated algebra corpus, learner-state variables, and a pedagogical response policy. In total, 24 two-turn algebra episodes were evaluated. No students were recruited, no classroom intervention was conducted, and no learning outcomes were measured. Eight expert reviewers scored anonymized response pairs for solution accuracy, conceptual support, instructional quality, instructional helpfulness, and unsupported content.

Results

The pedagogical RAG condition received higher task-level expert ratings for conceptual support (M = 4.69 vs. 4.31), instructional quality (M = 4.61 vs. 4.29), and instructional helpfulness (M = 4.65 vs. 4.39). It also achieved a higher pedagogical-quality composite score (M = 4.65 vs. 4.33; Wilcoxon p = .00024; rank-biserial r = .86). Solution accuracy was high in both conditions and showed only a small descriptive difference (M = 4.84 vs. 4.74; p = .0769). However, unsupported content flags were more frequent in the RAG condition, occurring in 13 of 192 rating cells compared with one of 192 rating cells in the baseline condition. Fisher's exact test yielded p = .0015.

Discussion

These findings do not establish learning gains, classroom efficacy, or the separate causal value of retrieval alone. They indicate that an integrated learner-state-aware RAG architecture can receive stronger expert ratings for pedagogical text quality on a fixed task set while also introducing trace leakage and evidence-boundary risks that must be controlled before deployment

Zitieren

APA

Adenuga, J. A. (2026). Retrieval-augmented generation for pedagogically aware educational AI: an expert-rated comparison of a prompt-only LLM tutor and an integrated, learner-state-aware RAG tutor. Frontiers in Education, 11. https://doi.org/10.3389/feduc.2026.1896839

Vancouver

Adenuga JA. Retrieval-augmented generation for pedagogically aware educational AI: an expert-rated comparison of a prompt-only LLM tutor and an integrated, learner-state-aware RAG tutor. Frontiers in Education. 2026;11. doi:10.3389/feduc.2026.1896839

BibTeX
@article{Adenuga2026,
  author = {Adenuga, Jude A.},
  title = {Retrieval-augmented generation for pedagogically aware educational AI: an expert-rated comparison of a prompt-only LLM tutor and an integrated, learner-state-aware RAG tutor},
  journal = {Frontiers in Education},
  year = {2026},
  volume = {11},
  doi = {10.3389/feduc.2026.1896839}
}
Kostenlos starten →