LLM在日常困境中更倾向成长而非关怀，可能影响决策多样性。
Growth First, Care Second? Tracing the Landscape of LLM Value Preferences in Everyday Dilemmas
- 从Reddit帖子构建价值框架，分析不同情境下的价值冲突模式。
- 多模型测试显示，LLM普遍偏好探索与成长，弱化关怀与连接价值。
- 女性相关话题价值冲突最复杂，职业建议中安全与自我实现对立明显。
人们越来越向人类同伴和基于大语言模型（LLM）的聊天机器人寻求建议。这类建议通常不追求唯一正确答案，而是需要权衡相互竞争的价值。本文旨在刻画LLM在不同建议场景中如何处理价值权衡。首先，我们使用来自四个建议类Reddit子版块的精选数据集，通过自下而上的方法，将个体建议中的细粒度价值归纳为高层级价值类别，构建层级化价值框架。通过构建价值共现网络，分析价值在困境中的共现模式，发现不同建议场景中价值权衡结构存在显著异质性：以女性为主题的子版块网络密度最高，表明价值冲突最复杂；女性、男性及友谊相关子版块的价值冲突模式高度相关，集中于安全与尊重/连接/承诺之间的张力；而职业建议则形成独特结构，安全常与自我实现和成长对立。随后，我们将LLM在这些困境中的价值偏好进行评估，发现无论模型或情境如何，LLM始终系统性地优先选择与探索与成长相关的价值，而弱化利他与连接价值。这一系统性偏斜提示，在人工智能中介建议中可能存在价值同质化风险，引发对大规模决策与规范结果塑造的担忧。
原文摘要 · Abstract (English)
People increasingly seek advice online from both human peers and large language model (LLM)-based chatbots. Such advice rarely involves identifying a single correct answer; instead, it typically requires navigating trade-offs among competing values. We aim to characterize how LLMs navigate value trade-offs across different advice-seeking contexts. First, we examine the value trade-off structure underlying advice seeking using a curated dataset from four advice-oriented subreddits. Using a bottom-up approach, we inductively construct a hierarchical value framework by aggregating fine-grained values extracted from individual advice options into higher-level value categories. We construct value co-occurrence networks to characterize how values co-occur within dilemmas and find substantial heterogeneity in value trade-off structures across advice-seeking contexts: a women-focused subreddit exhibits the highest network density, indicating more complex value conflicts; women's, men's, and friendship-related subreddits exhibit highly correlated value-conflict patterns centered on security-related tensions (security vs. respect/connection/commitment); by contrast, career advice forms a distinct structure where security frequently clashes with self-actualization and growth. We then evaluate LLM value preferences against these dilemmas and find that, across models and contexts, LLMs consistently prioritize values related to Exploration & Growth over Benevolence & Connection. This systemically skewed value orientation highlights a potential risk of value homogenization in AI-mediated advice, raising concerns about how such systems may shape decision-making and normative outcomes at scale.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。