审计大型语言模型中的专有对齐：一种无需真实标准的比较框架
Auditing Proprietary Alignment in Large Language Models: A Comparative Framework Without a Ground-Truth Standard
AI总结：
提出一种统计框架，通过比较目标模型与基线模型在共享语义空间中的响应偏差，检测黑盒语言模型中的专有对齐行为，无需真实标准即可实现外部审计。
摘要：
大型语言模型（LLMs）越来越多地通过不透明的开发和部署流程发布和部署，使得模型提供商能够在不正式宣布的情况下注入有意的、提供商特定的策略。因此，已有多种模型被报道生成反映专有规则和组织利益的响应，导致在有争议话题上的审查或错误信息。然而，系统性地识别这种对齐仍然是一个基本挑战，因为“专有”在不同语境中的含义模糊。在本文中，我们提出了一种统计框架，通过比较行为分析来检测黑盒语言模型中的专有对齐。我们的方法量化了目标模型与一组参考基线模型在共享语义空间中的响应之间的系统性偏差。通过评估相对行为差异而非绝对正确性，我们的框架能够在黑盒访问下进行有原则的审计。应用于几个广泛讨论但此前未量化的案例，它为外部评估大型语言模型中提供商特定的对齐行为提供了系统且可扩展的基础。
英文摘要：
Large language models (LLMs) are increasingly released and deployed through opaque development and deployment pipelines, enabling model providers to inject intentional, provider-specific policies without officially announcing them. As a result, various models have been reported to generate responses reflecting proprietary rules and organizational interests, leading to censorship or misinformation on controversial topics. However, systematic identification of such alignment remains a fundamental challenge, complicated by the ambiguity of what ``proprietary'' entails in different contexts. In this paper, we propose a statistical framework for detecting proprietary alignment in black-box language models via comparative behavioral analysis. Our approach quantifies systematic deviations between the responses of a target model and those of a reference set of baseline models in a shared semantic space. By evaluating relative behavioral divergence rather than absolute correctness, our framework enables principled auditing under black-box access. Applied to several widely discussed but previously unquantified cases, it provides a systematic and scalable basis for external assessment of provider-specific alignment behavior in large language models.
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。
