arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.06307cs.HC
收藏

基于LLM的教育模拟:评估跨ADHD类型的时间学生人格稳定性

LLM-Based Educational Simulation: Evaluating Temporal Student Persona Stability Across ADHD Profiles

Zuse Institute Berlin(齐泽研究所柏林);Weizenbaum Institute(魏岑鲍姆研究所);HU Berlin(柏林洪堡大学);TU Berlin(柏林工业大学)

Jana Gonnermann-Müller, Jennifer Haase, Nicolas Leins, Thomas Kosch, Sebastian Pokutta

AI总结:

本研究通过双重评估框架,测试大型语言模型在长时间交互中维持学生人格稳定性的能力,发现结构化交互设计可消除行为漂移,对教师培训和自适应辅导有重要意义。

摘要:

使用大型语言模型(LLM)进行学生模拟为教育研究和教师培训提供了一种可扩展的替代方案。然而,其有效性取决于模型是否能在长时间交互中维持稳定的人格。我们使用一个双重评估框架来测试这一前提,该框架测量自我报告特征和观察者评定的行为表达。通过两个实验,测试了四种临床基础的ADHD人格条件、五种LLM和三种提示设计,我们量化了对话间稳定性(N=4,968)和对话内稳定性(N=3,952,跨越9轮)。自我报告特征在高强度下保持稳定,构成了有效行为模拟的必要前提。观察者评定的行为表达显示出选择性不稳定性:在无脚本对话中,高和中度ADHD人格出现对话内漂移。使用明确任务提示的脚本交互完全消除了这种漂移。稳定、人格对齐的模拟学习者受益于结构化交互设计以维持行为一致性,这对教师培训、自适应辅导以及任何需要持续、路径依赖的学习者交互的应用具有重要意义。

英文摘要:

Large language model (LLM)-based student simulation offers a scalable alternative for educational research, teacher training, and learner practice. However, its validity depends on whether LLMs maintain stable personas across and within interactions. We test this using a dual-assessment framework measuring self-reported characteristics and observer-rated behavioral expressions. Across three ICD-11-informed ADHD-related intensity conditions and a default condition, five LLMs, and three prompt designs, we quantify between-conversation (Exp. I; N=$4,962$) and within-conversation stability (Exp. II; N=$3,952$). Experiment I shows that self-reports and observer ratings are more stable at high than moderate intensities across the tested models, prompts, and independent runs. Experiment II shows that self-reports remain stable throughout extended interactions, but observer-rated behavior drifts in unscripted dialog for high- and moderate-intensity personas. Scripted interactions with recurring task-relevant prompts eliminate this drift almost entirely (up to 97\% reduction). Structured interaction design helps simulated learners maintain behavioral stability in sustained, path-dependent interactions.

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。

↑