Large language models (LLMs), exemplified by ChatGPT, have gained considerable attention for their excellent natural language processing capabilities. Nonetheless, these LLMs present many challenges, particularly in the realm of trustworthiness. Therefore, ensuring the trustworthiness of LLMs emerges as an important topic. This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions. Specifically, we first propose a set of principles for trustworthy LLMs that span eight different dimensions. Based on these principles, we further establish a benchmark across six dimensions including truthfulness, safety, fairness, robustness, privacy, and machine ethics. We then present a study evaluating 16 mainstream LLMs in TrustLLM, consisting of over 30 datasets. Our findings firstly show that in general trustworthiness and utility (i.e., functional effectiveness) are positively related. Secondly, our observations reveal that proprietary LLMs generally outperform most open-source counterparts in terms of trustworthiness, raising concerns about the potential risks of widely accessible open-source LLMs. However, a few open-source LLMs come very close to proprietary ones. Thirdly, it is important to note that some LLMs may be overly calibrated towards exhibiting trustworthiness, to the extent that they compromise their utility by mistakenly treating benign prompts as harmful and consequently not responding. Finally, we emphasize the importance of ensuring transparency not only in the models themselves but also in the technologies that underpin trustworthiness. Knowing the specific trustworthy technologies that have been employed is crucial for analyzing their effectiveness.

翻译：以ChatGPT为代表的大型语言模型因其卓越的自然语言处理能力而备受关注。然而，这些语言模型也带来诸多挑战，尤其在可信度领域。因此，确保语言模型的可信度成为重要议题。本文提出TrustLLM，一项关于大型语言模型可信度的全面研究，涵盖可信度不同维度的原则、主流语言模型可信度的基准构建、评估与分析，以及开放挑战与未来方向的讨论。具体而言，我们首先提出了涵盖八个不同维度的可信语言模型原则集。基于这些原则，我们进一步构建了包含真实性、安全性、公平性、鲁棒性、隐私性和机器伦理六个维度的基准测试。随后，我们呈现了一项评估TrustLLM中16个主流语言模型的研究，涉及超过30个数据集。研究结果首先表明，总体而言可信度与实用性（即功能有效性）呈正相关。其次，我们的观察揭示，专有语言模型在可信度方面通常优于大多数开源模型，这引发了对广泛可获取的开源语言模型潜在风险的担忧。然而，少数开源语言模型已非常接近专有模型。第三，值得注意的是，部分语言模型可能过度校准以表现可信度，以至于错误地将良性提示视为有害内容而拒绝响应，从而牺牲了实用性。最后，我们强调确保透明度的重要性——不仅在于模型本身，更在于支撑可信度的技术。了解已采用的具体可信技术对于分析其有效性至关重要。