1
LLM-FACETS: A Privacy-Preserving Framework for Evaluating LLM Transparency and Accountability
最新论文提出隐私保护框架LLM-FACETS，系统评估大模型透明度与可问责性，兼顾安全与合规。
arXiv:2605.31167v1 Announce Type: new Abstract: Assessing whether Large Language Models outputs are factually grounded, epistemically calibrated, and …