Abstract:Student simulation in online education is important to address dynamic learning behaviors of students with diverse backgrounds. Existing simulation models based on deep learning usually need massive training data, lacking prior knowledge in educational contexts. Large language models (LLMs) may contain such prior knowledge since they are pre-trained from a large corpus. However, because student behaviors are dynamic and multifaceted with individual differences, directly prompting LLMs is not robust nor accurate enough to capture fine-grained interactions among diverse student personas, learning behaviors, and learning outcomes. This work tackles this problem by presenting a newly annotated fine-grained large-scale dataset and proposing EduAgent, a novel generative agent framework incorporating cognitive prior knowledge (i.e., theoretical findings revealed in cognitive science) to guide LLMs to first reason correlations among various behaviors and then make simulations. Our two experiments show that EduAgent could not only mimic and predict learning behaviors of real students but also generate realistic learning behaviors of virtual students without real data.

What problem does this paper attempt to address?

The paper attempts to address the problem of simulating student learning behaviors in online education. Specifically, existing deep learning-based student behavior simulation models typically require a large amount of training data and lack prior knowledge in the educational context. Large Language Models (LLMs) may contain such prior knowledge as they are pre-trained from extensive corpora. However, since student behaviors are dynamic, multifaceted, and vary among individuals, directly prompting LLMs cannot robustly or accurately capture the nuanced interactions between different student roles, learning behaviors, and learning outcomes. To overcome these issues, the authors propose a new fine-grained large-scale dataset and a new generative agent framework named EduAgent. EduAgent combines cognitive prior knowledge (i.e., theoretical findings revealed in cognitive science) to guide LLMs to first reason about the correlations between various behaviors and then perform simulations. Through this approach, EduAgent can not only mimic and predict the learning behaviors of real students but also generate realistic learning behaviors of virtual students, even without real data. ### Main Contributions: 1. **Dataset**: Provides a new large-scale, fine-grained annotated learning behavior dataset, including data from 311 real students and 705 virtual students. 2. **Generative Agent Framework**: Developed an open-source generative agent framework, EduAgent, which follows cognitive prior knowledge to realistically simulate learning behaviors in online education. 3. **Experimental Validation**: Conducted comprehensive experiments to validate the effectiveness of the EduAgent framework and evaluated the capability of state-of-the-art LLMs in modeling fine-grained learning behaviors. ### Key Technologies: - **Cognitive Prior Knowledge**: Utilizes theoretical findings in cognitive science to guide LLMs in reasoning about the correlations between student behaviors. - **Fine-Grained Data**: The dataset includes multi-dimensional records such as students' gaze trajectories, mouse control behaviors, and six different cognitive states. - **Modular Architecture**: The EduAgent framework stores different behaviors in different memory layers and infers how these behaviors are modulated by student characteristics and course content to more accurately simulate the learning process. ### Experimental Results: - **Personalized Student Behavior Prediction**: EduAgent can accurately predict the learning behaviors and test results of real students, even with only a small amount of historical data. - **Virtual Student Generation**: The generated virtual students exhibit behavior patterns consistent with real students, even without real training data. In summary, this paper addresses the challenge of simulating student behaviors in online education by combining cognitive prior knowledge and a generative agent framework, providing new insights for the development of intelligent teaching systems.

EduAgent: Generative Student Agents in Learning

Leveraging generative artificial intelligence to simulate student learning behavior

Simulating Classroom Education with LLM-Empowered Agents

Students Rather Than Experts: A New AI For Education Pipeline To Model More Human-Like And Personalised Early Adolescences

User Behavior Simulation with Large Language Model based Agents

Experiential Co-Learning of Software-Developing Agents

Learning Interactive Real-World Simulators

Large Language Models for In-Context Student Modeling: Synthesizing Student's Behavior in Visual Programming

MetaAgents: Simulating Interactions of Human Behaviors for LLM-based Task-oriented Coordination via Collaborative Generative Agents

Community-organizing Agent: an Artificial Intelligent System for Building Learning Communities among Large Numbers of Learners

AI Agents and Education: Simulated Practice at Scale

Synergistic Simulations: Multi-Agent Problem Solving with Large Language Models

LMAgent: A Large-scale Multimodal Agents Society for Multi-user Simulation

Collaboration with Generative Artificial Intelligence: An Exploratory Study Based on Learning Analytics

Generative Co-Learners: Enhancing Cognitive and Social Presence of Students in Asynchronous Learning with Generative AI

Enhanced Classroom Dialogue Sequences Analysis with a Hybrid AI Agent: Merging Expert Rule-Base with Large Language Models

Affordable Generative Agents

ExpeL: LLM Agents Are Experiential Learners

EduChat: A Large-Scale Language Model-based Chatbot System for Intelligent Education

AgentGym: Evolving Large Language Model-based Agents across Diverse Environments