用亚里士多德美德伦理学分析大模型在道德困境中的价值偏好。
Aristotelian Virtue Profiling of LLMs through Ethical Dilemmas
- 通过人类对五种回应的排序，构建美德优先级框架。
- 九个大模型族平均排名一致性达90.3%，勇气、节制与正义差异最大。
- 可本地运行的交互网站，对比人类与模型的美德画像。
大型语言模型（LLMs）在面对伦理权衡时，常需在公平、诚实、勇敢、克制等不同价值间抉择。本文提出VirtueMap框架，以亚里士多德美德伦理学为视角，分析这些选择模式。不寻求单一正确答案，而是让人类或模型对七类非致命、非政治、非宗教的通用道德困境中五个响应进行排序。针对每个困境和美德，我们先定义五种响应从最到最少体现该美德的参考顺序，并收集每条顺序超100次评估，仅当至少95%确认时才作为操作性真实基准。使用归一化布道对齐法对排名打分，生成实用智慧、正义、诚实、勇气、节制五个维度的美德画像。在九个大模型家族的重复实验中，平均排名一致性达90.3%，最大差异出现在勇气、节制与正义维度。同时发布交互式网页，可在浏览器本地计算并比较人类与模型的美德画像。
原文摘要 · Abstract (English)
Large Language Models (LLMs) often face ethical tradeoffs in which several responses may be defensible but express different priorities, such as fairness, honesty, courage, or restraint. We introduce VirtueMap, a framework for describing these patterns through an Aristotelian virtue-ethics lens. Instead of asking for a single correct answer, VirtueMap asks humans or LLMs to rank all five responses to each of seven general, non-lethal, non-political, and non-religious ethical dilemmas. To define the reference orderings used for scoring, we first proposed, for each dilemma and virtue, an ordering of the five responses from most to least expressive of that virtue. We then collected more than 100 respondent evaluations per ordering and retained it as operational ground truth only when at least 95% confirmed it. Rankings are scored against these retained orderings using normalized Borda alignment, yielding profiles over Practical Wisdom, Justice, Truthfulness, Courage, and Temperance. We apply VirtueMap to nine LLM families in a repeated-run evaluation and find high mean rank consistency (90.3%), with the largest differences appearing on Courage, Temperance, and Justice. We also release an interactive website that computes profiles locally in the browser and compares respondents with measured LLM profiles.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。