发表机构
University College London; Ben Gurion University of the Negev（伦敦大学学院; 内盖夫本古里安大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究评估7款LLMs对新闻标题情感框架的理解与人类的对齐度，发现不同模型相关性差异大，领先模型总体对齐但存在人口统计组差异，凸显AI对齐的非普遍性。
AI 中文摘要
大型语言模型（LLMs）正日益塑造我们消费信息和形成世界观的方式，这引发了超越AI偏见的担忧：LLMs是否能理解文本框架传达的情感细微差别？本研究中，我们实证评估一系列LLMs与人类情感感知的对齐程度。针对涵盖政治和地缘政治冲突的新闻标题，人类参与者（n=3011，通过YouGov调查获取的英国成年人口代表性样本）和7个LLMs均回答标题是否唤起对冲突中指定方的共情。我们发现，AI与人类评估的相关性因模型而异，范围从极高（0.789，GPT-5.2）到中等（0.4，Mistral Large 2512）。关键的是，领先模型在所有人口统计亚组（包括年龄、性别、教育水平、先前地缘政治知识及参与者对冲突的倾向）中与人类判断大致对齐，尽管组间存在统计学显著差异。本研究凭借其严谨设计和庞大的人口统计多样化数据集，提供了迄今为止对LLMs新闻框架理解最全面的评估。研究结果凸显了一个重要且常被忽视的差异对齐方面：即使总体性能很高，AI对齐也并非普遍存在——它可能与人口统计特征和文化规范对应不同。因此，考虑或忽视差异对齐的需求，可能对开发符合伦理且有用的AI系统产生重大影响。
英文摘要
Large Language Models (LLMs) are increasingly shaping how we consume information and form our worldview. This raises concerns beyond bias in AI: do LLMs grasp the emotional nuances conveyed via textual framing? In this work, we empirically evaluate how well an array of LLMs aligns with human emotional perception. Considering news headlines covering political and geopolitical conflicts, both human participants (n = 3011, a representative sample of the U.K. adult population, via a YouGov survey) and seven LLMs answered whether headlines evoked sympathy for a specified side in a conflict. We find that the correlation between AI and human evaluations varies across models, ranging from very high (0.789, GPT-5.2) to medium (0.4 ,Mistral Large 2512). Crucially, the leading models are broadly aligned with human judgments across all demographic subgroups, including age, gender, level of education, prior geopolitical knowledge, and participants' predispositions regarding the conflict, although there are statistically significant differences between groups. This research, with its robust design and large, demographically diverse dataset, offers the most comprehensive evaluation of LLMs' comprehension of news framing to date. Findings highlight an important, often-ignored aspect of differential alignment: even when aggregate performance is high, AI alignment is not universal -- it may correspond differently with demographic features and cultural norms. Considering or ignoring the need for differential alignment may therefore have significant implications for the development of ethical and useful AI systems.
