Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Conversational Learning Diagnosis via Reasoning Multi-Turn Interactive Learning

Fangzhou Yao, Sheng Chang, Weibo Gao, Qi Liu

TL;DR

ParLD is introduced, a preview-analyze-reason framework for conversational learning diagnosis, which leverages multi-agent collaboration to diagnose students' cognitive state over multiple dialogue turns and conducts experiments to evaluate both performance prediction and tutoring support.

Abstract

Learning diagnosis is a critical task that monitors students' cognitive state during educational activities, with the goal of enhancing learning outcomes. With advancements in language models (LMs), many AI-driven educational studies have shifted towards conversational learning scenarios, where students engage in multi-turn interactive dialogues with tutors. However, conversational learning diagnosis remains underdeveloped, and most existing techniques acquire students' cognitive state through intuitive instructional prompts on LMs to analyze the dialogue text. This direct prompting approach lacks a solid psychological foundation and fails to ensure the reliability of the generated analytical text. In this study, we introduce ParLD, a preview-analyze-reason framework for conversational learning diagnosis, which leverages multi-agent collaboration to diagnose students' cognitive state over multiple dialogue turns. Specifically, ParLD comprises three main components: (1) Behavior Previewer, which generates a student behavior schema based on previous states and learning content; (2) State Analyzer, which diagnoses the tutor-student dialogue and behavior schema to update the cognitive state; and (3) Performance Reasoner, which predicts the student's future responses and provides verifiable feedback to support ParLD's self-reflection with the Chain Reflector. They operate sequentially and iteratively during each interaction turn to diagnose the student's cognitive state. We conduct experiments to evaluate both performance prediction and tutoring support, emphasizing the effectiveness of ParLD in providing reliable and insightful learning diagnosis.

Conversational Learning Diagnosis via Reasoning Multi-Turn Interactive Learning

TL;DR

ParLD is introduced, a preview-analyze-reason framework for conversational learning diagnosis, which leverages multi-agent collaboration to diagnose students' cognitive state over multiple dialogue turns and conducts experiments to evaluate both performance prediction and tutoring support.

Abstract

Learning diagnosis is a critical task that monitors students' cognitive state during educational activities, with the goal of enhancing learning outcomes. With advancements in language models (LMs), many AI-driven educational studies have shifted towards conversational learning scenarios, where students engage in multi-turn interactive dialogues with tutors. However, conversational learning diagnosis remains underdeveloped, and most existing techniques acquire students' cognitive state through intuitive instructional prompts on LMs to analyze the dialogue text. This direct prompting approach lacks a solid psychological foundation and fails to ensure the reliability of the generated analytical text. In this study, we introduce ParLD, a preview-analyze-reason framework for conversational learning diagnosis, which leverages multi-agent collaboration to diagnose students' cognitive state over multiple dialogue turns. Specifically, ParLD comprises three main components: (1) Behavior Previewer, which generates a student behavior schema based on previous states and learning content; (2) State Analyzer, which diagnoses the tutor-student dialogue and behavior schema to update the cognitive state; and (3) Performance Reasoner, which predicts the student's future responses and provides verifiable feedback to support ParLD's self-reflection with the Chain Reflector. They operate sequentially and iteratively during each interaction turn to diagnose the student's cognitive state. We conduct experiments to evaluate both performance prediction and tutoring support, emphasizing the effectiveness of ParLD in providing reliable and insightful learning diagnosis.
Paper Structure (31 sections, 5 equations, 3 figures, 2 tables)

This paper contains 31 sections, 5 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Problem Definition
  3. Notations.
  4. Methodology
  5. Overview
  6. Behavior Previewer
  7. State Analyzer
  8. Performance Reasoner
  9. Chain Reflector via Memory
  10. Conversation Memory.
  11. Chain Reflector.
  12. Experiments
  13. Dataset
  14. MathDial
  15. CoMTA
  16. ...and 16 more sections

Figures (3)

  • Figure 1: (a) A typical conversational learning scenario where the student interacts with a teacher or intelligent tutor in a turn-based manner. (b) The ParLD framework for diagnosing cognitive state through the preview-analyze-reason cycle, which can iteratively evolve with each interaction.
  • Figure 2: The ablation study of ParLD (GPT-4.1) on the MathDial dataset.
  • Figure 3: A specific case of multi-turn conversational learning diagnosis and tutoring from the MathDial dataset.

Theorems & Definitions (1)

  • Definition 1: Conversational Learning Diagnosis