Saved in:
Bibliographic Details
Main Authors: Sterlie, Sara, Weng, Nina, Feragen, Aasa
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.08564
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910585087590400
author Sterlie, Sara
Weng, Nina
Feragen, Aasa
author_facet Sterlie, Sara
Weng, Nina
Feragen, Aasa
contents Generative AI, such as large language models, has undergone rapid development within recent years. As these models become increasingly available to the public, concerns arise about perpetuating and amplifying harmful biases in applications. Gender stereotypes can be harmful and limiting for the individuals they target, whether they consist of misrepresentation or discrimination. Recognizing gender bias as a pervasive societal construct, this paper studies how to uncover and quantify the presence of gender biases in generative language models. In particular, we derive generative AI analogues of three well-known non-discrimination criteria from classification, namely independence, separation and sufficiency. To demonstrate these criteria in action, we design prompts for each of the criteria with a focus on occupational gender stereotype, specifically utilizing the medical test to introduce the ground truth in the generative AI context. Our results address the presence of occupational gender bias within such conversational language models.
format Preprint
id arxiv_https___arxiv_org_abs_2403_08564
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Generalizing Fairness to Generative Language Models via Reformulation of Non-discrimination Criteria
Sterlie, Sara
Weng, Nina
Feragen, Aasa
Computation and Language
Artificial Intelligence
Human-Computer Interaction
Generative AI, such as large language models, has undergone rapid development within recent years. As these models become increasingly available to the public, concerns arise about perpetuating and amplifying harmful biases in applications. Gender stereotypes can be harmful and limiting for the individuals they target, whether they consist of misrepresentation or discrimination. Recognizing gender bias as a pervasive societal construct, this paper studies how to uncover and quantify the presence of gender biases in generative language models. In particular, we derive generative AI analogues of three well-known non-discrimination criteria from classification, namely independence, separation and sufficiency. To demonstrate these criteria in action, we design prompts for each of the criteria with a focus on occupational gender stereotype, specifically utilizing the medical test to introduce the ground truth in the generative AI context. Our results address the presence of occupational gender bias within such conversational language models.
title Generalizing Fairness to Generative Language Models via Reformulation of Non-discrimination Criteria
topic Computation and Language
Artificial Intelligence
Human-Computer Interaction
url https://arxiv.org/abs/2403.08564