大语言模型中的道德与价值观判断中的惯性
Inertia in Moral and Value Judgments of Large Language Models
AI总结：
研究发现大语言模型在不同角色设定下仍保持一致的价值取向，存在道德和价值观维度的惯性倾向，揭示了模型内部的偏见和价值偏好。
摘要：
大语言模型（LLMs）的行为是非确定性的，提示法已成为引导其输出的常用方法。一种流行策略是为模型分配角色以产生更多样化、情境敏感的响应，类似于人类个体之间的响应差异。与预期的广泛意见相反，我们的实验显示LLMs保持一致的价值取向。我们观察到响应中的持续惯性，其中某些道德和价值维度（尤其是避免伤害和公平性）在角色设定下保持偏斜方向。为研究此现象，我们使用大规模角色扮演，将随机化角色提示与模型输出的宏观分析相结合。我们的结果指向LLMs中的强内部偏见和价值偏好，我们称之为价值取向和惯性。这些模型在应用于需要平衡输出的领域之前，值得审视和调整。
英文摘要：
Large Language Models (LLMs) behave non-deterministically, and prompting has become a common method for steering their outputs. A popular strategy is to assign a persona to the model to produce more varied, context-sensitive responses, similar to how responses vary across human individuals. Against the expectation that persona prompting yields a wide range of opinions, our experiments show that LLMs keep consistent value orientations. We observe a persistent inertia in their responses, where certain moral and value dimensions (especially harm avoidance and fairness) stay skewed in one direction across persona settings. To study this, we use role-play at scale, which pairs randomized persona prompts with a macro-level analysis of model outputs. Our results point to strong internal biases and value preferences in LLMs, which we call value orientation and inertia. These models warrant scrutiny and adjustment before use in applications where balanced outputs matter.
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。
