Despite the many advances of Large Language Models (LLMs) and their unprecedented rapid evolution, their impact and integration into every facet of our daily lives is limited due to various reasons. One critical factor hindering their widespread adoption is the occurrence of hallucinations, where LLMs invent answers that sound realistic, yet drift away from factual truth. In this paper, we present a novel method for detecting hallucinations in large language models, which tackles a critical issue in the adoption of these models in various real-world scenarios. Through extensive evaluations across multiple datasets and LLMs, including Llama-2, we study the hallucination levels of various recent LLMs and demonstrate the effectiveness of our method to automatically detect them. Notably, we observe up to 62% hallucinations for Llama-2 in a specific experiment, where our method achieves a Balanced Accuracy (B-ACC) of 87%, all without relying on external knowledge.

翻译：尽管大型语言模型取得了诸多进展，且其进化速度之快前所未有，但由于多种原因，它们对我们日常生活方方面面的影响和整合依然受限。阻碍其广泛采用的关键因素之一是幻觉现象的出现，即大型语言模型会编造出听起来真实但偏离事实的答案。在本文中，我们提出了一种新颖的大型语言模型幻觉检测方法，这解决了这些模型在多种现实场景中应用的关键问题。通过在多个数据集和大型语言模型（包括Llama-2）上进行广泛评估，我们研究了近期多种大型语言模型的幻觉程度，并展示了我们方法在自动检测幻觉方面的有效性。值得注意的是，在特定实验中，我们观察到Llama-2的幻觉率高达62%，而我们的方法无需依赖外部知识即可达到87%的平衡准确率。