作者位次 #6/8ACL 2026Poster· Natural Language Processing
Demystifying Uncertainty in LLM s: Active Calibration between Concepts and Human Evaluations
Pengqi Li · Lizhong Ding · Zhehao Zhou · Chunhui Zhang · Jiarun Fu · Hao Li · Ye Yuan · Guoren Wang
摘要
Hallucinations arise when large language models (LLMs) guess rather than acknowledge their underlying uncertainty. Existing static strategies for mitigating hallucinations have been only partially successful, largely because they do not explicitly model the information gain from interacting with...