TrustLLM: Trustworthiness in Large Language Models

TrustLLM: Надёжность больших языковых моделей
Yue Huang, Lichao Sun, Haoran Wang, Siyuan Wu, Qihui Zhang, Yuan Li, Chang Gao, Yixin Huang, Wenhan Lyu, Yixuan Zhang, Xiner Li, Yixin Liu, Yixin Liu, Yijue Wang, Zhikun Zhang, Bertie Vidgen, Bhavya Kailkhura, C Xiong, Chaowei Xiao, Chunyuan Li, Eric Xing, F T Huang, Hao Liu, H Q Ji, Hongyi Wang, Huan Zhang, Huaxiu Yao, M Kellis, Marinka Žitnik, May S. Jiang, Mohit Bansal, James Zou, Jian Pei, Jian Liu, Gao, Jianfeng, Jiawei Han, Jieyu Zhao, Jiliang Tang, Jindong Wang, Joaquin Vanschoren, John C. Mitchell, Shu K, Kaidi Xu, Kai-Wei Chang, He, Lifang, Lifu Huang, Michael Backes A, Neil Zhenqiang Gong, Philip S. Yu, Pin-Yu Chen, Quanquan Gu, Xu Rd, Rex Ying, Shuiwang Ji, Suman Jana, Tianlong Chen, Tianming Liu, Tianyi Zhou, William Wang, Xing Xie, Xiangliang Zhang, Wang, Xiao, X J Xie, Xun Chen, Xuyu Wang, Yan Liu, Yanfang Ye, Yinzhi Cao, Yong Chen, Yue Zhao
2024-01-01

LLM trustworthinessTrustLLM benchmarkfairness and robustnessprivacy and machine ethicstruthfulness and safety
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.
1
Evaluation of 16 mainstream LLMs finds that trustworthiness and functional utility are generally positively related.
2
Proprietary LLMs generally outperform most open-source models in trustworthiness, although several open-source models approach proprietary performance.
3
Some models over-calibrate toward trustworthiness, incorrectly treating benign prompts as harmful and reducing utility by refusing to respond.
4
The benchmark covers six trustworthiness dimensions—truthfulness, safety, fairness, robustness, privacy, and machine ethics—and includes over 30 datasets.
5
The study emphasizes transparency about both model behavior and the specific technologies used to achieve trustworthiness.
6
TrustLLM introduces principles for trustworthy large language models spanning eight dimensions and uses them to structure systematic evaluation.

large language models (LLMs)

trustworthiness of LLMs across truthfulness, safety, fairness, robustness, privacy, and machine ethics, including its relationship with utility

Publication Details
Publication Date
2024-01-01
Journal
Publisher
ISSN
Access Type
Author Information
Authors
Yue Huang
Lichao Sun
Haoran Wang
Siyuan Wu
Qihui Zhang
Yuan Li
Chang Gao
Yixin Huang
Wenhan Lyu
Yixuan Zhang
Xiner Li
Yixin Liu
Yixin Liu
Yijue Wang
Zhikun Zhang
Bertie Vidgen
Bhavya Kailkhura
C Xiong
Chaowei Xiao
Chunyuan Li
Eric Xing
F T Huang
Hao Liu
H Q Ji
Hongyi Wang
Huan Zhang
Huaxiu Yao
M Kellis
Marinka Žitnik
May S. Jiang
Mohit Bansal
James Zou
Jian Pei
Jian Liu
Gao, Jianfeng
Jiawei Han
Jieyu Zhao
Jiliang Tang
Jindong Wang
Joaquin Vanschoren
John C. Mitchell
Shu K
Kaidi Xu
Kai-Wei Chang
He, Lifang
Lifu Huang
Michael Backes A
Neil Zhenqiang Gong
Philip S. Yu
Pin-Yu Chen
Quanquan Gu
Xu Rd
Rex Ying
Shuiwang Ji
Suman Jana
Tianlong Chen
Tianming Liu
Tianyi Zhou
William Wang
Xing Xie
Xiangliang Zhang
Wang, Xiao
X J Xie
Xun Chen
Xuyu Wang
Yan Liu
Yanfang Ye
Yinzhi Cao
Yong Chen
Yue Zhao
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%