Social Bias in Large Language Models For Bangla: An Empirical Study on Gender and Religious Bias

Jayanta Sadhu, Maneesha Rani Saha, Rifat Shahriyar
7/3/2024
cs.CL

No official code found. Best implementation: csebuetnlp/BanglaSocialBias (7 stars, last commit 2024-11-12).

Abstract

The rapid growth of Large Language Models (LLMs) has put forward the study of biases as a crucial field. It is important to assess the influence of different types of biases embedded in LLMs to ensure fair use in sensitive fields. Although there have been extensive works on bias assessment in English, such efforts are rare and scarce for a major language like Bangla. In this work, we examine two types of social biases in LLM generated outputs for Bangla language. Our main contributions in this work are: (1) bias studies on two different social biases for Bangla, (2) a curated dataset for bias measurement benchmarking and (3) testing two different probing techniques for bias detection in the context of Bangla. This is the first work of such kind involving bias assessment of LLMs for Bangla to the best of our knowledge. All our code and resources are publicly available for the progress of bias related research in Bangla NLP.

View on arXivView PDF

Code Implementations(1)

csebuetnlp/BanglaSocialBiasCommunity100%

This is the official repository containing all codes used to generate the results reported in the paper titled "Social Bias in Large Language Models For Bangla: An Empirical Study on Gender and Religious Bias"

73Shell, PythonJul 1, 20241 years ago
banglabangla-nlpsocial-bias

Similar Papers

An Empirical Study on the Characteristics of Bias upon Context Length Variation for Bangla

92% match1 repo32024

Cross-Lingual Probing and Community-Grounded Analysis of Gender Bias in Low-Resource Bengali

89% match2026

BanStereoSet: A Dataset to Measure Stereotypical Social Biases in LLMs for Bangla

89% match2024

Exploring Bengali Religious Dialect Biases in Large Language Models with Evaluation Perspectives

89% match2024

IndiBias: A Benchmark Dataset to Measure Social Biases in Language Models for Indian Context

89% match1 repo32024

Measuring South Asian Biases in Large Language Models

88% match2025

Cite this paper

@article{sadhu2024social,
  title  = {Social Bias in Large Language Models For Bangla: An Empirical Study on Gender and Religious Bias},
  author = {Jayanta Sadhu and Maneesha Rani Saha and Rifat Shahriyar},
  year   = {2024},
  eprint = {2407.03536},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2407.03536v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2407.03536.svg)](https://codeofpaper.com/paper/2407.03536)

Discussion