From Knowledge Augmentation to Pedagogical Decision-Making: A Review of Large Language Model-Based Intelligent Tutoring Systems

Rapid advances in large language models (LLMs) have enabled more flexible pedagogical interactions in intelligent tutoring systems. However, the reliability of generated content and its impact on actual learning outcomes remain contested. A central research question is how to integrate reliable knowledge with learner states to support tutoring decisions aligned with instructional objectives. Drawing on studies published between 2020 and 2026, this review examines advances in knowledge augmentation and learner modelling, analyses approaches to pedagogical strategy selection and feedback regulation, and assesses existing evidence of educational effectiveness. The literature indicates that retrieval-augmented generation and knowledge graphs can provide a knowledge basis for tutoring responses, while knowledge tracing and cognitive diagnosis help identify learners' knowledge states. Multi-turn planning and feedback mechanisms also provide greater control over instructional behaviour. Nevertheless, these capabilities have not yet been adequately validated in combination, and improvements in tutoring response quality do not necessarily translate into genuine learning gains. Empirical studies have reported substantially different learning outcomes, suggesting that instructional design and evaluation methods may influence judgements of educational effectiveness. Future research should strengthen the dynamic connections among supporting evidence, learner states, and pedagogical decisions. Independent assessments, knowledge transfer tasks, and longitudinal follow-up are also needed to verify learning outcomes and provide a more robust empirical basis for the design and application of LLM-based intelligent tutoring systems.

Authors

Institutions

Publication Details

Journal
International Educational Research
Published
2026-11-20
DOI
https://doi.org/10.30560/ier.v9n6p12
Primary Topic
Intelligent Tutoring Systems and Adaptive Learning
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

From Knowledge Augmentation to Pedagogical Decision-Making: A Review of Large Language Model-Based Intelligent Tutoring Systems

Kaiyue Zhao, Xiaochan Wang, Xiaolei Zhang, Hao Lin
3 links
International Educational Research
Intelligent Tutoring Systems and Adaptive Learning
article

From Knowledge Augmentation to Pedagogical Decision-Making: A Review of Large Language Model-Based Intelligent Tutoring Systems

Kaiyue Zhao, Xiaochan Wang, Xiaolei Zhang, Hao Lin
article en

Abstract

Rapid advances in large language models (LLMs) have enabled more flexible pedagogical interactions in intelligent tutoring systems. However, the reliability of generated content and its impact on actual learning outcomes remain contested. A central research question is how to integrate reliable knowledge with learner states to support tutoring decisions aligned with instructional objectives. Drawing on studies published between 2020 and 2026, this review examines advances in knowledge augmentation and learner modelling, analyses approaches to pedagogical strategy selection and feedback regulation, and assesses existing evidence of educational effectiveness. The literature indicates that retrieval-augmented generation and knowledge graphs can provide a knowledge basis for tutoring responses, while knowledge tracing and cognitive diagnosis help identify learners' knowledge states. Multi-turn planning and feedback mechanisms also provide greater control over instructional behaviour. Nevertheless, these capabilities have not yet been adequately validated in combination, and improvements in tutoring response quality do not necessarily translate into genuine learning gains. Empirical studies have reported substantially different learning outcomes, suggesting that instructional design and evaluation methods may influence judgements of educational effectiveness. Future research should strengthen the dynamic connections among supporting evidence, learner states, and pedagogical decisions. Independent assessments, knowledge transfer tasks, and longitudinal follow-up are also needed to verify learning outcomes and provide a more robust empirical basis for the design and application of LLM-based intelligent tutoring systems.

International Educational ResearchVol. 9(6)
Nanjing Agricultural University (CN), University of Hong Kong (HK)
Nanjing Agricultural University
Quality education
Intelligent Tutoring Systems and Adaptive Learning
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
From Knowledge Augmentation to Pedagogical Decision-Making: A Review of Large Language Model-Based Intelligent Tutoring Systems — Kaiyue Zhao, Xiaochan Wang, et al. · International Educational Research (2026) | TGRS Research Map | TGRS