Casteist but Not Racist? Quantifying Disparities in Large Language Model
Bias between India and the West
- URL: http://arxiv.org/abs/2309.08573v1
- Date: Fri, 15 Sep 2023 17:38:41 GMT
- Title: Casteist but Not Racist? Quantifying Disparities in Large Language Model
Bias between India and the West
- Authors: Khyati Khandelwal, Manuel Tonneau, Andrew M. Bean, Hannah Rose Kirk,
Scott A. Hale
- Abstract summary: Large Language Models (LLMs) can encode societal biases, exposing their users to representational harms.
We quantify stereotypical bias in popular LLMs according to an Indian-centric frame and compare bias levels between the Indian and Western contexts.
We find that the majority of LLMs tested are strongly biased towards stereotypes in the Indian context, especially as compared to the Western context.
- Score: 19.286414041202818
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Large Language Models (LLMs), now used daily by millions of users, can encode
societal biases, exposing their users to representational harms. A large body
of scholarship on LLM bias exists but it predominantly adopts a Western-centric
frame and attends comparatively less to bias levels and potential harms in the
Global South. In this paper, we quantify stereotypical bias in popular LLMs
according to an Indian-centric frame and compare bias levels between the Indian
and Western contexts. To do this, we develop a novel dataset which we call
Indian-BhED (Indian Bias Evaluation Dataset), containing stereotypical and
anti-stereotypical examples for caste and religion contexts. We find that the
majority of LLMs tested are strongly biased towards stereotypes in the Indian
context, especially as compared to the Western context. We finally investigate
Instruction Prompting as a simple intervention to mitigate such bias and find
that it significantly reduces both stereotypical and anti-stereotypical biases
in the majority of cases for GPT-3.5. The findings of this work highlight the
need for including more diverse voices when evaluating LLMs.
Related papers
- White Men Lead, Black Women Help? Benchmarking Language Agency Social Biases in LLMs [58.27353205269664]
Language agency is an important aspect of evaluating social biases in texts.
Previous research often relies on string-matching techniques to identify agentic and communal words.
We introduce the novel Language Agency Bias Evaluation benchmark.
arXiv Detail & Related papers (2024-04-16T12:27:54Z) - The Impact of Unstated Norms in Bias Analysis of Language Models [0.03495246564946556]
Large language models (LLMs) can carry biases that manifest in various forms, from overt discrimination to implicit stereotypes.
One facet of bias is performance disparities in LLMs, often harming underprivileged groups, such as racial minorities.
A common approach to quantifying bias is to use template-based bias probes, which explicitly state group membership.
arXiv Detail & Related papers (2024-04-04T14:24:06Z) - Exploring Value Biases: How LLMs Deviate Towards the Ideal [57.99044181599786]
Large-Language-Models (LLMs) are deployed in a wide range of applications, and their response has an increasing social impact.
We show that value bias is strong in LLMs across different categories, similar to the results found in human studies.
arXiv Detail & Related papers (2024-02-16T18:28:43Z) - Large Language Models are Geographically Biased [51.37609528538606]
We study what Large Language Models (LLMs) know about the world we live in through the lens of geography.
We show various problematic geographic biases, which we define as systemic errors in geospatial predictions.
arXiv Detail & Related papers (2024-02-05T02:32:09Z) - GPTBIAS: A Comprehensive Framework for Evaluating Bias in Large Language
Models [83.30078426829627]
Large language models (LLMs) have gained popularity and are being widely adopted by a large user community.
The existing evaluation methods have many constraints, and their results exhibit a limited degree of interpretability.
We propose a bias evaluation framework named GPTBIAS that leverages the high performance of LLMs to assess bias in models.
arXiv Detail & Related papers (2023-12-11T12:02:14Z) - What Do Llamas Really Think? Revealing Preference Biases in Language
Model Representations [62.91799637259657]
Do large language models (LLMs) exhibit sociodemographic biases, even when they decline to respond?
We study this research question by probing contextualized embeddings and exploring whether this bias is encoded in its latent representations.
We propose a logistic Bradley-Terry probe which predicts word pair preferences of LLMs from the words' hidden vectors.
arXiv Detail & Related papers (2023-11-30T18:53:13Z) - Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs [67.51906565969227]
We study the unintended side-effects of persona assignment on the ability of LLMs to perform basic reasoning tasks.
Our study covers 24 reasoning datasets, 4 LLMs, and 19 diverse personas (e.g. an Asian person) spanning 5 socio-demographic groups.
arXiv Detail & Related papers (2023-11-08T18:52:17Z) - Global Voices, Local Biases: Socio-Cultural Prejudices across Languages [22.92083941222383]
Human biases are ubiquitous but not uniform; disparities exist across linguistic, cultural, and societal borders.
In this work, we scale the Word Embedding Association Test (WEAT) to 24 languages, enabling broader studies.
To encompass more widely prevalent societal biases, we examine new bias dimensions across toxicity, ableism, and more.
arXiv Detail & Related papers (2023-10-26T17:07:50Z) - Gender bias and stereotypes in Large Language Models [0.6882042556551611]
This paper investigates Large Language Models' behavior with respect to gender stereotypes.
We use a simple paradigm to test the presence of gender bias, building on but differing from WinoBias.
Our contributions in this paper are as follows: (a) LLMs are 3-6 times more likely to choose an occupation that stereotypically aligns with a person's gender; (b) these choices align with people's perceptions better than with the ground truth as reflected in official job statistics; (d) LLMs ignore crucial ambiguities in sentence structure 95% of the time in our study items, but when explicitly prompted, they recognize
arXiv Detail & Related papers (2023-08-28T22:32:05Z) - CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked
Language Models [30.582132471411263]
We introduce the Crowd Stereotype Pairs benchmark (CrowS-Pairs)
CrowS-Pairs has 1508 examples that cover stereotypes dealing with nine types of bias, like race, religion, and age.
We find that all three of the widely-used sentences we evaluate substantially favor stereotypes in every category in CrowS-Pairs.
arXiv Detail & Related papers (2020-09-30T22:38:40Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.