# Strategize Before Teaching: A Conversational Tutoring System with Pedagogy Self-Distillation

## TL;DR

This paper introduces a conversational tutoring system that jointly predicts teaching strategies and generates responses, using a self-distillation approach to improve engagement and pedagogical effectiveness in educational dialogues.

## Contribution

It proposes a unified framework combining strategy prediction and response generation with self-distillation, advancing realistic and effective conversational tutoring.

## Key findings

- Joint strategy prediction and response generation improves tutoring quality
- Self-distillation enhances pedagogical strategy learning
- Benchmark results demonstrate effectiveness across datasets

## Abstract

Conversational tutoring systems (CTSs) aim to help students master educational material with natural language interaction in the form of a dialog. CTSs have become a key pillar in educational data mining research. A key challenge in CTSs is to engage the student in the conversation while exposing them to a diverse set of teaching strategies, akin to a human teacher, thereby, helping them learn in the process. Different from previous work that generates responses given the strategies as input, we propose to jointly predict teaching strategies and generate tutor responses accordingly, which fits a more realistic application scenario. We benchmark several competitive models on three dialog tutoring datasets and propose a unified framework that combines teaching response generation and pedagogical strategy prediction, where a self-distillation mechanism is adopted to guide the teaching strategy learning and facilitate tutor response generation. Our experiments and analyses shed light on how teaching strategies affect dialog tutoring.