Claim-Level Confidence Calibration for Reliable Decision Making with Large Language Models
面向大型语言模型可靠决策的声明级置信度校准
机构 * Tsinghua University(清华大学) ; Vulcan Research(伏尔坎研究院) ; AIFT ; Keio Global Research Institute (KGRI)(庆应全球研究所(KGRI)) ; China Mobile Research Institute(中国移动研究院) ; Zhongguancun Laboratory(中关村实验室)
AI总结 该研究针对大型语言模型的幻觉及置信度与事实不匹配问题,提出黑箱场景下的声明级置信度校准框架,在TriviaQA等数据集上降低了事实问题的预期校准误差。
Comments In Proceedings of The 5th Workshop on Uncertainty Reasoning and Quantification in Decision Making (held in conjunction with ACM SIGKDD 2026), Jeju, Korea