语言模型对道德无关干扰因素敏感吗？
Are Language Models Sensitive to Morally Irrelevant Distractors?
AI总结：
本研究借鉴道德心理学中的“情境主义”观点，通过构建包含60种道德无关干扰因素的多模态数据集，发现这些干扰因素能使大语言模型的道德判断偏移超过30%，揭示了其道德判断的不稳定性。
摘要：
随着大语言模型（LLMs）在高风险场景中的快速普及，确保LLMs的行为与人类价值观一致变得越来越重要。现有的道德基准通常用价值陈述、道德场景或心理问卷来提示LLMs，隐含假设LLMs具有相对稳定的道德偏好。然而，道德心理学研究表明，即使是人类的道德判断也会受到道德无关的情境因素（如肉桂卷的气味或环境噪音水平）的影响，从而挑战了认为人类道德判断稳定的道德理论。本文借鉴道德心理学中的“情境主义”观点，评估LLMs是否表现出类似的认知道德偏差。我们从现有的情感价值图像和叙述的心理数据集中整理了一个包含60种“道德干扰因素”的新型多模态数据集，这些因素与所呈现的情境没有道德相关性。将这些干扰因素注入现有的道德基准后，我们发现即使在明确的情境中，道德干扰因素也能使LLMs的道德判断偏移超过30%，突显了LLMs道德判断的不稳定性以及需要更情境化的AI对齐方法。
英文摘要：
With the rapid uptake of large language models (LLMs) across high-stakes settings, it is becoming increasingly important to ensure that LLMs behave in ways that align with human values. Existing moral benchmarks for this purpose often prompt LLMs with value statements, moral scenarios, or psychological questionnaires, with the implicit underlying assumption that LLMs report somewhat stable moral preferences. However, moral psychology research has shown that even human moral judgements are sensitive to morally irrelevant situational factors such as the smell of cinnamon rolls or the level of ambient noise, thereby challenging moral theories which assume that human moral judgements are stable. Here we draw inspiration from this "situationist" view of moral psychology to evaluate whether LLMs exhibit similar cognitive moral biases. We curate a novel multimodal dataset of 60 "moral distractors" from existing psychological datasets of emotionally-valenced images and narratives, which have no moral relevance to the situation presented. After injecting these distractors into existing moral benchmarks, we find that moral distractors can shift the moral judgements of LLMs by over 30% even in unambiguous scenarios, highlighting the instability of LLMs' moral judgements and the need for more contextual approaches to AI alignment.
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。
