Language models (LMs) are known to represent the perspectives of some social
groups better than others, which may impact their performance, especially on
subjective tasks such as content moderation and hate speech detection. To
explore how LMs represent different perspectives, existing research focused on
positional alignment, i.e., how closely the models mimic the opinions and
stances of different groups, e.g., liberals or conservatives. However, human
communication also encompasses emotional and moral dimensions. We define the
problem of affective alignment, which measures how LMs' emotional and moral
tone represents those of different groups. By comparing the affect of responses
generated by 36 LMs to the affect of Twitter messages, we observe significant
misalignment of LMs with both ideological groups. This misalignment is larger
than the partisan divide in the U.S. Even after steering the LMs towards
specific ideological perspectives, the misalignment and liberal tendencies of
the model persist, suggesting a systemic bias within LMs.

语言模型（LMs）代表某些社会群体的观点较好，可能在内容管理和仇恨言论检测等主观任务上发挥作用。本研究旨在探讨 LMs 如何代表不同观点，现有研究主要关注定位对齐，即模型模拟不同群体（如自由派或保守派）的观点和立场的接近程度，而人类交流还包含情感和道德维度。我们定义了情感对齐问题，度量了 LMs 情感和道德色彩在代表不同群体方面的表现。通过比较 36 个 LMs 生成的回应与 Twitter 消息的情感对比，我们观察到 LMs 与意识形态群体都存在显著的不对齐问题。这种不对齐超过了美国的党派分歧。即使将 LMs 定向于特定的意识形态观点，不对齐问题和模型的自由倾向仍然存在，暗示 LMs 内存在系统偏见。