AI 中文总结
该研究针对基于大语言模型的群组推荐，审计聚合规则权限，发现权限委托会赋予模型聚合选择的裁量权，且不同条件下模型聚合结果分布存在差异。
AI 中文摘要
AI系统越来越多地为偏好存在冲突的用户做出联合推荐。然而，当合理的聚合规则支持不同的行动时，会出现一个进一步的问题：谁可以选择这些偏好的组合方式？我们将这一交互层面的问题称为聚合规则权限。我们使用合成偏好配置文件和从实证评分构建的配置文件，对三个大语言模型（LLM）在三种权限条件下进行受控行为审计：未指定、明确由用户保留、委托给模型。当两个见证规则支持不同行动时，在用户保留权限的情况下，模型几乎从不做出承诺，但在所有委托情况下都做出了承诺。所有三个模型在直接受指令时都完美执行了两个见证规则。然而，当权限未指定或被委托时，它们的聚合一致结果分布因模型和偏好设置而异。这些结果共同将规则执行能力与聚合规则权限区分开来：委托赋予模型解决聚合选择的自由裁量权，但并不确定后续的集体结果。
英文摘要
AI systems increasingly make joint recommendations for users with conflicting preferences. However, when reasonable aggregation rules support different actions, a further question arises: who may choose how those preferences are combined? We study this interaction-level problem as aggregation-rule authority. Using synthetic preference profiles and profiles constructed from empirical ratings, we conduct a controlled behavioral audit of three LLMs under three authority conditions: unspecified, explicitly retained by users, and delegated to the model. In cases where two witness rules supported different actions, models almost never committed when users retained authority, but committed in every delegated case. All three models executed both witness rules perfectly when directly instructed. Yet when authority was unspecified or delegated, their aggregation-consistent outcome distributions differed across models and preference settings. Together, these results separate rule-execution capability from aggregation-rule authority: delegation assigns the model discretion to resolve the aggregation choice, but does not determine which collective outcome follows.
