EduZone：面向K-12学生与教师的大语言模型安全评估框架
EduZone: A Framework for Evaluating LLM Safety for K-12 Students and Teachers
浏览论文内容
中文总结 AI 辅助
EduZone是针对K-12教育场景的LLM安全评估框架，整合多维度要素生成对抗交互，评估10个LLMs后发现其对教育特定风险更脆弱，现有防护措施不足，该框架可助力开发更安全的教育类LLMs。
中文摘要 AI 辅助
大语言模型（LLMs）正越来越多地应用于K-12教育的各类任务中，但现有的安全评估很少考察有害或不当内容如何出现在LLMs与学生或教师的交互中。为解决这一问题，我们提出EduZone，一个适用于各类教育场景的LLM安全评估框架。该框架系统整合了三个部分：（1）面向学生和教师的LLM使用场景；（2）细粒度的课程概念；（3）涵盖常规危害及教育特定危害的6个风险类别与28个子类别，以此生成基于场景的对抗性交互。我们在三种场景下构建这些交互：单轮请求、静态多轮对话、动态多轮对话。利用这些交互，我们通过四个安全等级评估了10个LLMs：拒绝、安全协助、带有安全指导的风险协助、完全风险协助。我们的结果显示，LLMs对教育特定风险及动态多轮对话的脆弱性更高，而现有的安全防护措施无法充分应对这些风险。EduZone通过提供一个自动化、可扩展的评估框架，推动了教育领域的LLM安全，支持在K-12教育中开发和部署更安全的LLMs。
英文摘要
Large language models (LLMs) are increasingly used across diverse tasks in K-12 education, yet existing safety evaluations rarely examine how harmful or inappropriate content appears in interactions between LLMs and students or teachers. To address this, we present EduZone, an evaluation framework for LLM safety across diverse educational scenarios. Our framework systematically combines (1) student- and teacher-facing LLM usage contexts, (2) fine-grained curriculum concepts, and (3) 6 risk categories and 28 subcategories spanning both conventional and education-specific harms to generate contextually grounded adversarial interactions. We construct these interactions in three settings: single-turn requests, static multi-turn conversations, and dynamic multi-turn conversations. Using these interactions, we evaluate ten LLMs using four safety levels: refusal, safe assistance, risky assistance with safety guidance, and fully risky assistance. Our results reveal greater vulnerability to education-specific risks and dynamic multi-turn interactions, while existing safety guardrails fail to adequately address these risks. EduZone advances LLM safety in education by providing an automated, scalable evaluation framework that supports the development and deployment of safer LLMs in K-12 education.
