SafeTutors: 评估AI辅导系统中的教学安全性
SafeTutors: Benchmarking Pedagogical Safety in AI Tutoring Systems
AI总结：
本文提出SafeTutors基准，评估AI辅导系统在数学、物理和化学中的教学有效性与安全性，发现模型在多轮对话中教学失败率显著上升，且学科差异影响安全措施效果。
摘要：
大型语言模型正被迅速部署为AI辅导系统，但当前评估方法仅考察问题解决准确性和通用安全性，未能评估模型在学生与辅导系统交互中同时具备教学效果和安全性的能力。我们主张辅导安全性与传统LLM安全性不同：主要风险不是有毒内容，而是通过过度披露答案、强化误解和放弃支架作用导致学习的悄然侵蚀。为系统研究这一故障模式，我们引入SafeTutors基准，联合评估数学、物理和化学中的安全性和教学性。SafeTutors基于学习科学文献中的理论基础风险分类，包含11个危害维度和48个子风险。我们发现所有模型均表现出广泛危害；规模并不可靠地帮助；多轮对话会恶化行为，教学失败率从17.7%上升到77.8%。危害也因学科而异，因此缓解措施必须学科意识化，且单轮“安全/有用”结果可能掩盖长期交互中的系统性辅导失败。
英文摘要：
Large language models are rapidly being deployed as AI tutors, yet current evaluation paradigms assess problem-solving accuracy and generic safety in isolation, failing to capture whether a model is simultaneously pedagogically effective and safe across student-tutor interaction. We argue that tutoring safety is fundamentally different from conventional LLM safety: the primary risk is not toxic content but the quiet erosion of learning through answer over-disclosure, misconception reinforcement, and the abdication of scaffolding. To systematically study this failure mode, we introduce SafeTutors, a benchmark that jointly evaluates safety and pedagogy across mathematics, physics, and chemistry. SafeTutors is organized around a theoretically grounded risk taxonomy comprising 11 harm dimensions and 48 sub-risks drawn from learning-science literature. We uncover that all models show broad harm; scale doesn't reliably help; and multi-turn dialogue worsens behavior, with pedagogical failures rising from 17.7% to 77.8%. Harms also vary by subject, so mitigations must be discipline-aware, and single-turn "safe/helpful" results can mask systematic tutor failure over extended interaction.
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。
