How (Un)Fair is Text Summarization?

Hannah Brown, Reza Shokri

Keywords: Natural language processing, Summarization, Fairness

Abstract: Creating a good summary requires carefully choosing details from the original text to accurately represent it in a limited space. If a summary contains biased information about a group, it risks passing this bias off to readers as fact. These risks increase if we consider not just one biased summary, but rather a biased summarization algorithm. Despite this, little work has measured whether these summarizers demonstrate biased performance. Rather, most work in summarization focuses on improving performance, ignoring questions of bias. In this paper we demonstrate that automatic summarizers both amplify and introduce bias towards information about under-represented groups. Additionally, we show that summarizers are highly sensitive to document structure, making the summaries they generate unstable under changes that are semantically meaningless to humans, which poses a further fairness risk. Given these results, and the large scale potential for harm presented by biased summarization, we recommend that bias analysis be performed and reported on summarizers to ensure that new automatic summarization methods do not introduce bias to the summaries they generate.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)

TL;DR: We show that machine learning based summarizers exhibit bias toward different groups and are very sensitive to document structure.