Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cao, Jie, Nguyen, Ha, Yavuz, Selim, Yu, Boran, Wang, Shuguang, Bharaj, Pavneet Kaur, Francis, Dionne Cross
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2604.04361
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910105555959808
author Cao, Jie
Nguyen, Ha
Yavuz, Selim
Yu, Boran
Wang, Shuguang
Bharaj, Pavneet Kaur
Francis, Dionne Cross
author_facet Cao, Jie
Nguyen, Ha
Yavuz, Selim
Yu, Boran
Wang, Shuguang
Bharaj, Pavneet Kaur
Francis, Dionne Cross
contents Large Language Model (LLM) simulations, where LLMs act as students with varying approaches to learning tasks, can support teachers' noticing of student thinking. However, simulations using zero- or few-shot prompting often yield inauthentic knowledge and language, directing teachers to unrealistic reasoning. We evaluate three approaches (Fine-tuning, Multi-agent, and Direct Preference Optimization; DPO) to improve the authenticity and pedagogical utility of simulated students. All approaches improve cognitive and linguistic authenticity, compared with few-shot prompts. Interviews with elementary mathematics pre-service teachers and researchers (\textit{n} = 8) reveal distinct pedagogical affordances. The fine-tuned model produces realistic, brief responses but limits opportunities to extend students' thinking. Meanwhile, the multi-agent and DPO approaches generate explicit reasoning behind student strategies. We discuss implications for designing LLM simulations that balance authenticity with instructional utility for teacher learning.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04361
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Developing Authentic Simulated Learners for Mathematics Teacher Learning: Insights from Three Approaches with Large Language Models
Cao, Jie
Nguyen, Ha
Yavuz, Selim
Yu, Boran
Wang, Shuguang
Bharaj, Pavneet Kaur
Francis, Dionne Cross
Human-Computer Interaction
Large Language Model (LLM) simulations, where LLMs act as students with varying approaches to learning tasks, can support teachers' noticing of student thinking. However, simulations using zero- or few-shot prompting often yield inauthentic knowledge and language, directing teachers to unrealistic reasoning. We evaluate three approaches (Fine-tuning, Multi-agent, and Direct Preference Optimization; DPO) to improve the authenticity and pedagogical utility of simulated students. All approaches improve cognitive and linguistic authenticity, compared with few-shot prompts. Interviews with elementary mathematics pre-service teachers and researchers (\textit{n} = 8) reveal distinct pedagogical affordances. The fine-tuned model produces realistic, brief responses but limits opportunities to extend students' thinking. Meanwhile, the multi-agent and DPO approaches generate explicit reasoning behind student strategies. We discuss implications for designing LLM simulations that balance authenticity with instructional utility for teacher learning.
title Developing Authentic Simulated Learners for Mathematics Teacher Learning: Insights from Three Approaches with Large Language Models
topic Human-Computer Interaction
url https://arxiv.org/abs/2604.04361