Biases in Large Language Models: Origins, Inventory, and Discussion
Journal of Data and Information Quality · 2023-06-30 · 15 (2) · 1-21
DOI: 10.1145/3597307 ↗At a glance
Journal article in Journal of Data and Information Quality, published online on 2023-06-22. Written by Roberto Navigli and 2 co-authors. Cited 368 times so far (Crossref); 132 references.
Abstract (original language)
In this article, we introduce and discuss the pervasive issue of bias in the large language models that are currently at the core of mainstream approaches to Natural Language Processing (NLP). We first introduce data selection bias, that is, the bias caused by the choice of texts that make up a training corpus. Then, we survey the different types of social bias evidenced in the text generated by language models trained on such corpora, ranging from gender to age, from sexual orientation to ethnicity, and from religion to culture. We conclude with directions focused on measuring, reducing, and tackling the aforementioned types of bias.
Cite
APA
Navigli, R., Conia, S., & Ross, B. (2023). Biases in Large Language Models: Origins, Inventory, and Discussion. Journal of Data and Information Quality, 15(2), 1-21. https://doi.org/10.1145/3597307
Vancouver
Navigli R, Conia S, Ross B. Biases in Large Language Models: Origins, Inventory, and Discussion. Journal of Data and Information Quality. 2023;15(2):1-21. doi:10.1145/3597307
BibTeX
@article{Navigli2023,
author = {Navigli, Roberto and Conia, Simone and Ross, Björn},
title = {Biases in Large Language Models: Origins, Inventory, and Discussion},
journal = {Journal of Data and Information Quality},
year = {2023},
volume = {15},
number = {2},
pages = {1--21},
doi = {10.1145/3597307}
}
EUVIMED