PATS：基于大语言模型导师的个性感知教学策略
PATS: Personality-Aware Teaching Strategies with Large Language Model Tutors
AI总结：
PATS通过结合学生性格特征优化大语言模型的教学策略，提升教学效果和学生参与度。
摘要：
近年来，大语言模型（LLMs）在教育辅导中的应用展现出巨大潜力。然而，不同的教学策略对不同学生性格特征具有不同效果，匹配不当可能会对学生成绩产生负面影响。尽管如此，当前的LLM辅导系统并未考虑学生性格特征。为了解决这一问题，我们首先构建了一个将教学方法与性格特征联系起来的分类体系，基于教学文献。我们模拟了学生与教师的对话，并利用我们的框架让LLM导师根据模拟学生的性格特征调整其策略。我们通过与人类教师的评估发现，他们一致认为我们的方法优于两个基线方法。此外，我们的方法还增加了使用较少但影响较大的策略，如角色扮演，这被人类和LLM标注者显著偏好。我们的发现为开发更个性化和有效的LLM在教育应用中的使用铺平了道路。
英文摘要：
Recent advances in large language models (LLMs) demonstrate their potential as educational tutors. However, different tutoring strategies benefit different student personalities, and mismatches can be counterproductive to student outcomes. Despite this, current LLM tutoring systems do not take into account student personality traits. To address this problem, we first construct a taxonomy that links pedagogical methods to personality profiles, based on pedagogical literature. We simulate student-teacher conversations and use our framework to let the LLM tutor adjust its strategy to the simulated student personality. We evaluate the scenario with human teachers and find that they consistently prefer our approach over two baselines. Our method also increases the use of less common, high-impact strategies such as role-playing, which human and LLM annotators prefer significantly. Our findings pave the way for developing more personalized and effective LLM use in educational applications.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
