Capturing Bias Diversity in LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gosavi, Purva Prasad, Kulkarni, Vaishnavi Murlidhar, Smeaton, Alan F.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908476151693312
author Gosavi, Purva Prasad
Kulkarni, Vaishnavi Murlidhar
Smeaton, Alan F.
author_facet Gosavi, Purva Prasad
Kulkarni, Vaishnavi Murlidhar
Smeaton, Alan F.
contents This paper presents research on enhancements to Large Language Models (LLMs) through the addition of diversity in its generated outputs. Our study introduces a configuration of multiple LLMs which demonstrates the diversities capable with a single LLM. By developing multiple customised instances of a GPT model, each reflecting biases in specific demographic characteristics including gender, age, and race, we propose, develop and evaluate a framework for a more nuanced and representative AI dialogue which we call BiasGPT. The customised GPT models will ultimately collaborate, merging their diverse perspectives on a topic into an integrated response that captures a broad spectrum of human experiences and viewpoints. In this paper, through experiments, we demonstrate the capabilities of a GPT model to embed different biases which, when combined, can open the possibilities of more inclusive AI technologies.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12839
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Capturing Bias Diversity in LLMs
Gosavi, Purva Prasad
Kulkarni, Vaishnavi Murlidhar
Smeaton, Alan F.
Computation and Language
Artificial Intelligence
This paper presents research on enhancements to Large Language Models (LLMs) through the addition of diversity in its generated outputs. Our study introduces a configuration of multiple LLMs which demonstrates the diversities capable with a single LLM. By developing multiple customised instances of a GPT model, each reflecting biases in specific demographic characteristics including gender, age, and race, we propose, develop and evaluate a framework for a more nuanced and representative AI dialogue which we call BiasGPT. The customised GPT models will ultimately collaborate, merging their diverse perspectives on a topic into an integrated response that captures a broad spectrum of human experiences and viewpoints. In this paper, through experiments, we demonstrate the capabilities of a GPT model to embed different biases which, when combined, can open the possibilities of more inclusive AI technologies.
title Capturing Bias Diversity in LLMs
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2410.12839