作者位次 #4/5ACL 2026Findings· Findings of ACL
J oint CQ : Improving Factual Hallucination Detection with Joint Claim and Query Generation
Fan Xu · Huixuan Zhang · Zhenliang Zhang · Jiahao Wang · Xiaojun Wan
摘要
Current large language models (LLMs) often suffer from hallucination issues, i,e, generating content that appears factual but is actually unreliable. A typical hallucination detection pipeline involves response decomposition (i.e., claim extraction), query generation, evidence collection (i.e., s...