Explainability for Large Language Models: A Survey

doi:10.1145/3639372

	Explainability for Large Language Models: A Survey
	Zhao, Haiyan 1; Chen, Hanjie 2; Yang, Fan 3; Liu, Ninghao 4; Deng, Huiqi 5; Cai, Hengyi 6; Wang, Shuaiqiang 7; Yin, Dawei 7; Du, Mengnan 1
	2024-04-01
发表期刊	ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
ISSN	2157-6904
卷号	15 期号:2 页码:38
摘要	Large language models (LLMs) have demonstrated impressive capabilities in natural language processing. However, their internal mechanisms are still unclear and this lack of transparency poses unwanted risks for downstream applications. Therefore, understanding and explaining these models is crucial for elucidating their behaviors, limitations, and social impacts. In this article, we introduce a taxonomy of explainability techniques and provide a structured overview ofmethods for explaining Transformer-based language models. We categorize techniques based on the training paradigms of LLMs: traditional fine-tuning-based paradigm and prompting-based paradigm. For each paradigm, we summarize the goals and dominant approaches for generating local explanations of individual predictions and global explanations of overall model knowledge. We also discuss metrics for evaluating generated explanations and discuss how explanations can be leveraged to debug models and improve performance. Lastly, we examine key challenges and emerging opportunities for explanation techniques in the era of LLMs in comparison to conventional deep learning models.
关键词	Explainability interpretability large language models
DOI	10.1145/3639372
收录类别	SCI
语种	英语
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Information Systems
WOS记录号	WOS:001208775700001
出版者	ASSOC COMPUTING MACHINERY
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/39001
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhao, Haiyan
作者单位	1.New Jersey Inst Technol, 323 Dr Martin Luther King Jr Blvd, Newark, NJ 07102 USA 2.Johns Hopkins Univ, 3400 N Charles St, Baltimore, MD 21218 USA 3.Wake Forest Univ, 1834 Wake Forest Rd, Winston Salem, NC 27109 USA 4.Univ Georgia, Herty Dr, Athens, GA 30602 USA 5.Shanghai Jiao Tong Univ, 800 Dongchuan RD, Shanghai 200240, Peoples R China 6.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 7.10 Shangdi 10th St, Beijing 100085, Peoples R China
推荐引用方式 GB/T 7714	Zhao, Haiyan,Chen, Hanjie,Yang, Fan,et al. Explainability for Large Language Models: A Survey[J]. ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY,2024,15(2):38.
APA	Zhao, Haiyan.,Chen, Hanjie.,Yang, Fan.,Liu, Ninghao.,Deng, Huiqi.,...&Du, Mengnan.(2024).Explainability for Large Language Models: A Survey.ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY,15(2),38.
MLA	Zhao, Haiyan,et al."Explainability for Large Language Models: A Survey".ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY 15.2(2024):38.