Evaluating Large Language Models through Gender and Racial Stereotypes

TL;DR

This paper develops a framework to evaluate large language models for gender and racial biases in professional contexts, revealing progress in gender bias reduction but persistent racial bias.

Contribution

It introduces a novel evaluation framework specifically targeting gender and racial biases in language models within professional scenarios.

Findings

Gender bias has decreased in newer models.

Racial bias remains prevalent.

Framework can identify bias levels effectively.

Abstract

Language Models have ushered a new age of AI gaining traction within the NLP community as well as amongst the general population. AI's ability to make predictions, generations and its applications in sensitive decision-making scenarios, makes it even more important to study these models for possible biases that may exist and that can be exaggerated. We conduct a quality comparative study and establish a framework to evaluate language models under the premise of two kinds of biases: gender and race, in a professional setting. We find out that while gender bias has reduced immensely in newer models, as compared to older ones, racial bias still exists.