Teaching Language Models How to Code Like Learners: Conversational Serialization for Student Simulation

TL;DR

This paper introduces a framework for training open-weight models to simulate student programming behavior by serializing their problem-solving process into conversational data, improving alignment with real student debugging actions.

Contribution

It presents a novel training pipeline combining supervised fine-tuning and preference optimization on authentic student data, enabling models to better mimic student debugging behavior.

Findings

Models trained with environment feedback outperform prior approaches.

Fine-tuned models show improved functional alignment and code similarity.

The framework effectively simulates student debugging in programming assignments.

Abstract

Artificial students -- models that simulate how learners act and respond within educational systems -- are a promising tool for evaluating tutoring strategies and feedback mechanisms at scale. However, most existing approaches rely on prompting large, proprietary language models, limiting adaptability to specific courses and raising concerns around privacy, cost, and dependence. In this work, we propose a framework for training open-weight artificial programming learners directly from authentic student process data. Our approach serializes temporal log traces into a conversational format, representing each student's problem-solving process as a dialogue between the learner and their automated assessment system. Student code submissions and environment feedback, such as test outcomes, grades, and error traces, form alternating conversational turns, enabling models to learn from the iterative debugging process. We additionally introduce a training pipeline combining supervised fine-tuning with preference optimization to align models with authentic student debugging behavior. We evaluate our framework by training Qwen models at 4B and 8B scales on a large-scale dataset of real student submissions to Python programming assignments. Our results show that incorporating environment feedback strengthens models' ability to replicate student debugging behavior, improving over both prior code-only approaches and prompted large language models baselines in functional alignment and code similarity. We release our code to support reproducibility.