幻觉检测：在大语言模型中稳健地辨别可靠答案
Hallucination Detection: Robustly Discerning Reliable Answers in Large Language Models
AI总结：
本文提出稳健判别器 RelD，并在双语问答数据集 RelQA 上训练，用于有效检测大语言模型生成答案中的幻觉，实验证明其在分布内和分布外数据上均表现良好。
摘要：
大语言模型（LLMs）已在多种自然语言处理任务中得到广泛应用，包括问答和对话系统。然而，LLMs 的一个主要缺陷是幻觉问题，即它们生成偏离输入来源的不忠实或不一致内容，从而导致严重后果。在本文中，我们提出了一种名为 RelD 的稳健判别器，用于有效检测 LLMs 生成答案中的幻觉。RelD 在构建的 RelQA 上进行训练，RelQA 是一个双语问答对话数据集，同时包含由 LLMs 生成的答案以及一套全面的度量指标。我们的实验结果表明，所提出的 RelD 能够成功检测多种 LLMs 生成答案中的幻觉。此外，它在区分分布内和分布外数据集中 LLMs 生成答案的幻觉方面也表现良好。另外，我们还对出现的幻觉类型进行了深入分析，并提出了有价值的见解。本研究对检测 LLMs 生成的可靠答案做出了重要贡献，并对未来工作中缓解幻觉问题具有重要意义。
英文摘要：
Large Language Models (LLMs) have gained widespread adoption in various natural language processing tasks, including question answering and dialogue systems. However, a major drawback of LLMs is the issue of hallucination, where they generate unfaithful or inconsistent content that deviates from the input source, leading to severe consequences. In this paper, we propose a robust discriminator named RelD to effectively detect hallucination in LLMs' generated answers. RelD is trained on the constructed RelQA, a bilingual question-answering dialogue dataset along with answers generated by LLMs and a comprehensive set of metrics. Our experimental results demonstrate that the proposed RelD successfully detects hallucination in the answers generated by diverse LLMs. Moreover, it performs well in distinguishing hallucination in LLMs' generated answers from both in-distribution and out-of-distribution datasets. Additionally, we also conduct a thorough analysis of the types of hallucinations that occur and present valuable insights. This research significantly contributes to the detection of reliable answers generated by LLMs and holds noteworthy implications for mitigating hallucination in the future work.
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。
