Sep 2026· Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing· Vol abs/2609.29804· 0 citations· 39 references
Computer Science
TL;DR
The proposed REAT framework couples experience distillation from historical dialogues with real-time adaptive retrieval, which significantly outperforms both prompt-only and supervised fine-tuning (SFT) baselines, particularly in improving complex, low-scoring tutoring scenarios.
Abstract
Current Large Language Models (LLMs) excel at solving complex mathematical problems, yet this proficiency does not inherently translate into effective tutoring. While advanced LLM tutors may leverage multi-agent frameworks or fine-tuning, most still lack a mechanism to systematically accumulate and reuse pedagogical experience over time, limiting their adaptability to diverse student needs during fluid, multi-turn interactions. To bridge this gap, we propose the Reflective Experience-Augmented Tutoring (REAT) framework, which couples experience distillation from historical dialogues with real-time adaptive retrieval. Driven by a multi-agent Observer-Critic-Mentor (OCM) distillation pipeline, REAT reviews past conversational trajectories and distills raw interactions into structured, problem-agnostic pedagogical experiences. During live tutoring, a state-aware retrieval module injects these curated experiences to provide adaptive scaffolding based on the student's cognitive state. Experiments demonstrate that the proposed framework significantly outperforms both prompt-only and supervised fine-tuning (SFT) baselines, particularly in improving complex, low-scoring tutoring scenarios. Crucially, the distilled experiences exhibit robust generalization across diverse model architectures and mathematical datasets.
It is indicated 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.
Jude A. Adenuga· Frontiers in Education· 1 citation
Retrieval-augmented generation (RAG) is increasingly used to ground large language model (LLM) tutors, but comparing tutoring architectures is expensive, because every design decision ordinarily requires a fresh study with students. We present a reproducible simulation harness that evaluates tutoring architectures agai...
It is argued that for embodied educational CRI the central design problem is not maximizing generative flexibility but architecting interaction that earns and uses what it learns about the child.
Elena Malnatsky, Sheng-Hui Wang, Kuhu Sinha et al.· IEEE International Symposium...· 0 citations
PLAI is presented, a web-based prototype that generates explanations from lecture slides, audio transcripts, and a structured student profile through a chat-based interface and suggests that profile-based multimodal prompts may mainly support motivational engagement rather than immediate test performance.
Furkan Ali Yurdakul, Yi-Man Wu, Maria Torres Vega et al.· Message Understanding Confer...· 0 citations
An engaging and cost-effective digital tool for developing subject-matter knowledge (SMK) could mitigate educational inequities in resource-constrained contexts. This case study explores the effectiveness of free generative artificial intelligence (GAI) platforms, prompted to function as a Flipped-Interaction Intellige...
A. Stott, Stefan Scheepers· Canadian Journal of Science,...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.