A Survey of Large Language Models Attribution

TL;DR

This survey reviews attribution mechanisms in large language models used in open-domain generative AI, highlighting challenges like biases and ambiguous knowledge, and aims to guide future improvements in attribution methods.

Contribution

It provides a comprehensive overview of attribution techniques in large language models and offers a repository to track ongoing research in this emerging field.

Findings

Attribution improves factuality and verifiability of AI responses.

Challenges include biases and ambiguous knowledge reservoirs.

The field is still in early development stages.

Abstract

Open-domain generative systems have gained significant attention in the field of conversational AI (e.g., generative search engines). This paper presents a comprehensive review of the attribution mechanisms employed by these systems, particularly large language models. Though attribution or citation improve the factuality and verifiability, issues like ambiguous knowledge reservoirs, inherent biases, and the drawbacks of excessive attribution can hinder the effectiveness of these systems. The aim of this survey is to provide valuable insights for researchers, aiding in the refinement of attribution methodologies to enhance the reliability and veracity of responses generated by open-domain generative systems. We believe that this field is still in its early stages; hence, we maintain a repository to keep track of ongoing studies at https://github.com/HITsz-TMG/awesome-llm-attributions.