arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.03546cs.CLcs.AIcs.HCcs.LG
收藏

大语言模型在隐私与亲社会冲突下的价值-行动对齐

Value-Action Alignment in Large Language Models under Privacy-Prosocial Conflict

Arizona State University(亚利桑那州立大学);University of Southern California(南加州大学)

Guanyu Chen, Chenxiao Yu, Xiyang Hu

AI总结:

研究探讨大语言模型在隐私与亲社会冲突中的价值-行动对齐,通过多组结构方程模型分析隐私关注与亲社会性对数据共享的影响,提出价值-行动对齐率(VAAR)作为评估指标。

摘要:

大语言模型(LLMs)日益被用于模拟涉及个人数据共享的决策任务,其中隐私担忧和亲社会动机可能导向相反的选择。现有评估通常孤立测量隐私相关态度或共享意图,难以确定模型表达的价值是否共同预测其下游数据共享行为,如同真实人类行为。我们引入基于情境的评估协议,依次在有限且具有历史记录的会话中施行政策标准化问卷,以评估在竞争态度下的价值-行动对齐。我们使用多组结构方程模型(MGSEM)识别隐私关注和亲社会性到数据共享的关系。我们提出价值-行动对齐率(VAAR),一种以人类为参考的方向一致性度量,聚合路径层面的证据以确定预期符号。在多个LLMs上,我们观察到稳定的但模型特定的隐私-PSA-AoDS轮廓,以及价值-行动对齐的显著异质性。

英文摘要:

Large language models (LLMs) are increasingly used to simulate decision-making tasks involving personal data sharing, where privacy concerns and prosocial motivations can push choices in opposite directions. Existing evaluations often measure privacy-related attitudes or sharing intentions in isolation, which makes it difficult to determine whether a model's expressed values jointly predict its downstream data-sharing actions as in real human behaviors. We introduce a context-based assessment protocol that sequentially administers standardized questionnaires for privacy attitudes, prosocialness, and acceptance of data sharing within a bounded, history-carrying session. To evaluate value-action alignments under competing attitudes, we use multi-group structural equation modeling (MGSEM) to identify relations from privacy concerns and prosocialness to data sharing. We propose Value-Action Alignment Rate (VAAR), a human-referenced directional agreement metric that aggregates path-level evidence for expected signs. Across multiple LLMs, we observe stable but model-specific Privacy-PSA-AoDS profiles, and substantial heterogeneity in value-action 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。

↑