作者位次 #2/10ACL 2026Poster· Natural Language Processing
Illusions of Confidence? Diagnosing LLM Truthfulness via Neighborhood Consistency
Haoming Xu · Ningyuan Zhao · Yunzhi Yao · Weihong Xu · Hongru WANG · Xinle Deng · Shumin Deng · Jeff Z. Pan · Huajun Chen · Ningyu Zhang
摘要
As Large Language Models (LLMs) are increasingly deployed in real-world settings, correctness alone is insufficient. Reliable deployment requires maintaining truthful beliefs under contextual perturbations. Existing evaluations largely rely on point-wise confidence like Self-Consistency, which ca...