Auditing Proprietary Alignment in Large Language Models: A Comparative Framework Without a Ground-Truth Standard 事件

Auditing Proprietary Alignment in Large Language Models: A Comparative Framework Without a Ground-Truth Standard arXiv:2606.08381v1 Announce Type: cross Abstract: 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 or