Saved in:
Bibliographic Details
Main Authors: Li, Haoxuan, Yu, Jifan, Cong, Xin, Dang, Yang, Zhang-li, Daniel, Mi, Lu, Zhan, Yisi, Liu, Huiqin, Liu, Zhiyuan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.11678
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914179116433408
author Li, Haoxuan
Yu, Jifan
Cong, Xin
Dang, Yang
Zhang-li, Daniel
Mi, Lu
Zhan, Yisi
Liu, Huiqin
Liu, Zhiyuan
author_facet Li, Haoxuan
Yu, Jifan
Cong, Xin
Dang, Yang
Zhang-li, Daniel
Mi, Lu
Zhan, Yisi
Liu, Huiqin
Liu, Zhiyuan
contents While rapid advances in large language models (LLMs) are reshaping data-driven intelligent education, accurately simulating students remains an important but challenging bottleneck for scalable educational data collection, evaluation, and intervention design. However, current works are limited by scarce real interaction data, costly expert evaluation for realism, and a lack of large-scale, systematic analyses of LLMs ability in simulating students. We address this gap by presenting a three-stage LLM-human collaborative pipeline to automatically generate and filter high-quality student agents. We leverage a two-round automated scoring validated by human experts and deploy a score propagation module to obtain more consistent scores across the student similarity graph. Experiments show that combining automated scoring, expert calibration, and graph-based propagation yields simulated student that more closely track authentication by human judgments. We then analyze which profiles and behaviors are simulated more faithfully, supporting subsequent studies on personalized learning and educational assessment.
format Preprint
id arxiv_https___arxiv_org_abs_2502_11678
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Which Type of Students can LLMs Act? Investigating Authentic Simulation with Graph-based Human-AI Collaborative System
Li, Haoxuan
Yu, Jifan
Cong, Xin
Dang, Yang
Zhang-li, Daniel
Mi, Lu
Zhan, Yisi
Liu, Huiqin
Liu, Zhiyuan
Computers and Society
Computation and Language
While rapid advances in large language models (LLMs) are reshaping data-driven intelligent education, accurately simulating students remains an important but challenging bottleneck for scalable educational data collection, evaluation, and intervention design. However, current works are limited by scarce real interaction data, costly expert evaluation for realism, and a lack of large-scale, systematic analyses of LLMs ability in simulating students. We address this gap by presenting a three-stage LLM-human collaborative pipeline to automatically generate and filter high-quality student agents. We leverage a two-round automated scoring validated by human experts and deploy a score propagation module to obtain more consistent scores across the student similarity graph. Experiments show that combining automated scoring, expert calibration, and graph-based propagation yields simulated student that more closely track authentication by human judgments. We then analyze which profiles and behaviors are simulated more faithfully, supporting subsequent studies on personalized learning and educational assessment.
title Which Type of Students can LLMs Act? Investigating Authentic Simulation with Graph-based Human-AI Collaborative System
topic Computers and Society
Computation and Language
url https://arxiv.org/abs/2502.11678