Dialogic Learning in Child-Robot Interaction: A Hybrid Approach to Personalized Educational Content Generation
Elena Malnatsky, Shenghui Wang, Koen V. Hindriks, Mike E. U. Ligthart
TL;DR
The paper tackles the challenge of delivering safe, age-appropriate, and pedagogically aligned dialogic learning in child–robot interactions at scale. It proposes a hybrid framework that fuses rule-based scaffolding with selectively offline-generated, LLM-driven content, governed by a validation pipeline (LLM evaluator plus human moderation) to ensure quality and safety. The authors articulate design principles for educational CRIs, demonstrate the framework through the Robot-Bookworm project with 100 children, and report scalable, personalized dialogues that maintain developmental appropriateness. The work advances practical, scalable, and responsible dialogic learning in CRIs and outlines future directions involving knowledge graphs, retrieval-augmented generation, and enhanced real-time adaptability, aiming to deepen engagement and learning outcomes in classroom settings.
Abstract
Dialogic learning fosters motivation and deeper understanding in education through purposeful and structured dialogues. Foundational models offer a transformative potential for child-robot interactions, enabling the design of personalized, engaging, and scalable interactions. However, their integration into educational contexts presents challenges in terms of ensuring age-appropriate and safe content and alignment with pedagogical goals. We introduce a hybrid approach to designing personalized educational dialogues in child-robot interactions. By combining rule-based systems with LLMs for selective offline content generation and human validation, the framework ensures educational quality and developmental appropriateness. We illustrate this approach through a project aimed at enhancing reading motivation, in which a robot facilitated book-related dialogues.