RPO-PDT: Demonstrating Role-Play-Based Knowledge Adaptation for Student Support Dialogue (Demonstration System)

2026-06-08

AI Summary

This work addresses the challenge of delivering safe, compliant, and personalized academic tutoring conversations in higher education. We propose a dialogue system that integrates role-playing with retrieval-augmented generation to enable knowledge-grounded, adaptive tutoring through structured knowledge sources. A novel “reverse role-playing” mechanism is introduced, wherein unresolved student issues are replayed from the student’s perspective to generate reusable tutoring strategies, which are then stored in a strategy memory bank to establish a continuous improvement loop. The system embeds ethical constraints—including personality consistency, boundary enforcement, confidentiality, and safety—and supports both textual and Furhat embodied interaction, resulting in a deployable tutoring framework that adheres to institutional policies while maintaining personalization and grounding in domain knowledge.

Abstract

We present RPO-PDT: a retrieval-grounded, role-play-based dialogue system for adaptive student support in higher education. RPO-PDT is: (1) able to provide institution-specific Personal Development Tutor (PDT) guidance using structured knowledge sources; (2) constrained by explicit persona, boundary, confidentiality, and safety policies; and (3) designed around a reverse-roleplay loop where unresolved interactions are replayed from the student perspective, enabling alternative tutor strategies to be generated and stored as reusable strategy memory. RPO-PDT supports both text-based and Furhat-based embodied interaction for demonstrating grounded, safe, and adaptive student-support dialogue.

Problem

• student support
• dialogue system
• role-play
• adaptive tutoring
• higher education

Innovation

• role-play-based dialogue
• retrieval-grounded
• adaptive student support
• reverse-roleplay loop
• strategy memory