The widespread availability of large language models (LLMs), such as ChatGPT, has significantly impacted education, raising both opportunities and challenges. Students can frequently interact with LLM-powered, interactive learning tools, but their usage patterns need to be analyzed to ensure ethical usage of these tools. To better understand how students interact with LLMs in an academic setting, we introduce \textbf{StudyChat}, a publicly available dataset capturing real-world student interactions with an LLM-powered tutoring chatbot in a semester-long, university-level artificial intelligence (AI) course. We deploy a web application that replicates ChatGPT's core functionalities, and use it to log student interactions with the LLM while working on programming assignments. We collect 1,197 conversations, which we annotate using a dialogue act labeling schema inspired by observed interaction patterns and prior research. Additionally, we analyze these interactions, highlight behavioral trends, and analyze how specific usage patterns relate to course outcomes. \textbf{StudyChat} provides a rich resource for the learning sciences and AI in education communities, enabling further research into the evolving role of LLMs in education.

本研究针对学生与大语言模型（LLM）之间的互动模式进行分析，填补了教育领域对这种工具使用模式研究的空白。通过引入 StudyChat 数据集，该数据集记录了学生在大学人工智能课程中与 LLM 驱动的辅导聊天机器人进行的真实对话，并揭示了具体的使用模式对课程结果的影响。该研究为学习科学和教育中的人工智能相关领域提供了重要资源，推动了对 LLM 角色演变的进一步研究。

StudyChat 数据集：学生与 ChatGPT 在人工智能课程中的对话