作者位次 #3/8ACL 2026Findings· Findings of ACL
FAITH : Factuality Alignment through Integrating Trustworthiness and Honestness
Xiaoning Dong · Chengyan Wu · Yajie Wen · Yu Chen · Yun Xue · Zhang Jing · Wei Xu · Bolei Ma
摘要
Large Language Models (LLMs) can generate factually inaccurate content even if they have corresponding knowledge, which critically undermines their reliability. Existing approaches attempt to mitigate this by incorporating uncertainty in QA prompt during training, but these numerical scores lack...