A Survey of Large Language Models Attribution
Abstract: Open-domain generative systems have gained significant attention in the field of conversational AI (e.g., generative search engines). In this paper, we present a comprehensive review of the attribution mechanisms employed by these systems, particularly with large language models. While attribution or citation improves factuality and verifiability, issues like ambiguous knowledge reservoirs, inherent biases, and the drawbacks of excessive attribution can hinder the effectiveness of these systems. The purpose of this survey is to provide valuable implications for researchers, helping in the refinement of attribution methodologies to improve the reliability and veracity of responses generated by open-domain generative systems. We believe that this field is still in its early stages; therefore, we maintain a repository to keep track of ongoing studies at \url{AnonymousURL}.
Paper Type: Long
Research Area: NLP Applications
Research Area Keywords: Verifiable Text Generation, Large Language Models, Attribution and Groundedness
Contribution Types: Surveys
Languages Studied: not specific
Submission Number: 573
Loading