Saved in:
Bibliographic Details
Main Authors: Mehner, Luise, Fiedler, Lena Alicija Philine, Ammon, Sabine, Kolossa, Dorothea
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.20898
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913712667885568
author Mehner, Luise
Fiedler, Lena Alicija Philine
Ammon, Sabine
Kolossa, Dorothea
author_facet Mehner, Luise
Fiedler, Lena Alicija Philine
Ammon, Sabine
Kolossa, Dorothea
contents The widespread application of Large Language Models (LLMs) involves ethical risks for users and societies. A prominent ethical risk of LLMs is the generation of unfair language output that reinforces or exacerbates harm for members of disadvantaged social groups through gender biases (Weidinger et al., 2022; Bender et al., 2021; Kotek et al., 2023). Hence, the evaluation of the fairness of LLM outputs with respect to such biases is a topic of rising interest. To advance research in this field, promote discourse on suitable normative bases and evaluation methodologies, and enhance the reproducibility of related studies, we propose a novel approach to database construction. This approach enables the assessment of gender-related biases in LLM-generated language beyond merely evaluating their degree of neutralization.
format Preprint
id arxiv_https___arxiv_org_abs_2502_20898
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A database to support the evaluation of gender biases in GPT-4o output
Mehner, Luise
Fiedler, Lena Alicija Philine
Ammon, Sabine
Kolossa, Dorothea
Computation and Language
The widespread application of Large Language Models (LLMs) involves ethical risks for users and societies. A prominent ethical risk of LLMs is the generation of unfair language output that reinforces or exacerbates harm for members of disadvantaged social groups through gender biases (Weidinger et al., 2022; Bender et al., 2021; Kotek et al., 2023). Hence, the evaluation of the fairness of LLM outputs with respect to such biases is a topic of rising interest. To advance research in this field, promote discourse on suitable normative bases and evaluation methodologies, and enhance the reproducibility of related studies, we propose a novel approach to database construction. This approach enables the assessment of gender-related biases in LLM-generated language beyond merely evaluating their degree of neutralization.
title A database to support the evaluation of gender biases in GPT-4o output
topic Computation and Language
url https://arxiv.org/abs/2502.20898