arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.10539cs.CL
收藏

检测真实LLM-人类交互中的幻觉

Detecting Hallucinations in Authentic LLM-Human Interactions

Trustworthy AI Lab, University of Hamburg, Germany(可信AI实验室,汉堡大学,德国)

Yujie Ren, Niklas Gruhlke, Anne Lauscher

AI总结:

本文提出首个基于真实LLM-人类交互的幻觉检测基准,揭示了LLM在日常交互中产生幻觉的比例,并探讨了使用LLM自身作为检测器的潜力。

摘要:

随着大型语言模型(LLMs)在医学和法律等敏感领域中的应用日益增加,幻觉检测已成为一项关键任务。尽管已经提出了众多基准来推动该领域的研究,但大多数都是人工构造的——要么通过故意诱导幻觉或模拟交互来生成,而不是从真实的LLM-人类对话中衍生而来。因此,这些基准未能完全捕捉到实际使用中幻觉的特征。为了解决这一限制,我们引入了AuthenHallu,这是首个完全基于真实LLM-人类交互构建的幻觉检测基准。对于AuthenHallu,我们从真实的LLM-人类对话中选择并标注样本,从而提供了LLM在日常用户交互中产生幻觉的忠实反映。统计分析显示,在我们的基准中,31.4%的查询-响应对存在幻觉,而在数学与数字问题等挑战性领域中,这一比例急剧上升至60.0%。此外,我们探索了使用普通LLM本身作为幻觉检测器的潜力,并发现尽管有些前景,但其当前性能在实际应用场景中仍然不足。数据和代码已公开在https://github.com/TAI-HAMBURG/AuthenHallu。

英文摘要:

As large language models (LLMs) are increasingly applied in sensitive domains such as medicine and law, hallucination detection has become a critical task. Although numerous benchmarks have been proposed to advance research in this area, most of them are artificially constructed--either through deliberate hallucination induction or simulated interactions--rather than derived from genuine LLM-human dialogues. Consequently, these benchmarks fail to fully capture the characteristics of hallucinations that occur in real-world usage. To address this limitation, we introduce AuthenHallu, the first hallucination detection benchmark built entirely from authentic LLM-human interactions. For AuthenHallu, we select and annotate samples from genuine LLM-human dialogues, thereby providing a faithful reflection of how LLMs hallucinate in everyday user interactions. Statistical analysis shows that hallucinations occur in 31.4% of the query-response pairs in our benchmark, and this proportion increases dramatically to 60.0% in challenging domains such as Math & Number Problems. Furthermore, we explore the potential of using vanilla LLMs themselves as hallucination detectors and find that, despite some promise, their current performance remains insufficient in real-world scenarios. The data and code are publicly available at https://github.com/TAI-HAMBURG/AuthenHallu.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑