Bias Beyond English: Evaluating Social Bias and Debiasing Methods in a Low-Resource Setting

• URL: http://arxiv.org/abs/2504.11183v1
• Date: Tue, 15 Apr 2025 13:40:22 GMT
• Title: Bias Beyond English: Evaluating Social Bias and Debiasing Methods in a Low-Resource Setting
• Authors: Ej Zhou, Weiming Lu,
• Abstract summary: Social bias in language models can potentially exacerbate social inequalities.<n>This study aims to leverage high-resource language corpora to evaluate bias and experiment with debiasing methods in low-resource languages.
• Score: 8.478711218359532
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Social bias in language models can potentially exacerbate social inequalities. Despite it having garnered wide attention, most research focuses on English data. In a low-resource scenario, the models often perform worse due to insufficient training data. This study aims to leverage high-resource language corpora to evaluate bias and experiment with debiasing methods in low-resource languages. We evaluated the performance of recent multilingual models in five languages: English (\textsc{eng}), Chinese (\textsc{zho}), Russian (\textsc{rus}), Indonesian (\textsc{ind}) and Thai (\textsc{tha}), and analyzed four bias dimensions: \textit{gender}, \textit{religion}, \textit{nationality}, and \textit{race-color}. By constructing multilingual bias evaluation datasets, this study allows fair comparisons between models across languages. We have further investigated three debiasing methods-\texttt{CDA}, \texttt{Dropout}, \texttt{SenDeb}-and demonstrated that debiasing methods from high-resource languages can be effectively transferred to low-resource ones, providing actionable insights for fairness research in multilingual NLP.

Related papers

• Zero-shot Sentiment Analysis in Low-Resource Languages Using a Multilingual Sentiment Lexicon [78.12363425794214]
  We focus on zero-shot sentiment analysis tasks across 34 languages, including 6 high/medium-resource languages, 25 low-resource languages, and 3 code-switching datasets. We demonstrate that pretraining using multilingual lexicons, without using any sentence-level sentiment data, achieves superior zero-shot performance compared to models fine-tuned on English sentiment datasets.
  arXiv  Detail & Related papers  (2024-02-03T10:41:05Z)
• Multilingual Text-to-Image Generation Magnifies Gender Stereotypes and Prompt Engineering May Not Help You [64.74707085021858]
  We show that multilingual models suffer from significant gender biases just as monolingual models do. We propose a novel benchmark, MAGBIG, intended to foster research on gender bias in multilingual models. Our results show that not only do models exhibit strong gender biases but they also behave differently across languages.
  arXiv  Detail & Related papers  (2024-01-29T12:02:28Z)
• Investigating Bias in Multilingual Language Models: Cross-Lingual Transfer of Debiasing Techniques [3.9673530817103333]
  Cross-lingual transfer of debiasing techniques is not only feasible but also yields promising results. Using translations of the CrowS-Pairs dataset, our analysis identifies SentenceDebias as the best technique across different languages.
  arXiv  Detail & Related papers  (2023-10-16T11:43:30Z)
• NusaWrites: Constructing High-Quality Corpora for Underrepresented and Extremely Low-Resource Languages [54.808217147579036]
  We conduct a case study on Indonesian local languages. We compare the effectiveness of online scraping, human translation, and paragraph writing by native speakers in constructing datasets. Our findings demonstrate that datasets generated through paragraph writing by native speakers exhibit superior quality in terms of lexical diversity and cultural content.
  arXiv  Detail & Related papers  (2023-09-19T14:42:33Z)
• On Evaluating and Mitigating Gender Biases in Multilingual Settings [5.248564173595024]
  We investigate some of the challenges with evaluating and mitigating biases in multilingual settings. We first create a benchmark for evaluating gender biases in pre-trained masked language models. We extend various debiasing methods to work beyond English and evaluate their effectiveness for SOTA massively multilingual models.
  arXiv  Detail & Related papers  (2023-07-04T06:23:04Z)
• CBBQ: A Chinese Bias Benchmark Dataset Curated with Human-AI Collaboration for Large Language Models [52.25049362267279]
  We present a Chinese Bias Benchmark dataset that consists of over 100K questions jointly constructed by human experts and generative language models. The testing instances in the dataset are automatically derived from 3K+ high-quality templates manually authored with stringent quality control. Extensive experiments demonstrate the effectiveness of the dataset in detecting model bias, with all 10 publicly available Chinese large language models exhibiting strong bias in certain categories.
  arXiv  Detail & Related papers  (2023-06-28T14:14:44Z)
• Average Is Not Enough: Caveats of Multilingual Evaluation [0.0]
  We argue that a qualitative analysis informed by comparative linguistics is needed for multilingual results to detect this kind of bias. We show in our case study that results in published works can indeed be linguistically biased and we demonstrate that visualization based on onEL typological database can detect it.
  arXiv  Detail & Related papers  (2023-01-03T18:23:42Z)
• An Analysis of Social Biases Present in BERT Variants Across Multiple Languages [0.0]
  We investigate the bias present in monolingual BERT models across a diverse set of languages. We propose a template-based method to measure any kind of bias, based on sentence pseudo-likelihood. We conclude that current methods of probing for bias are highly language-dependent.
  arXiv  Detail & Related papers  (2022-11-25T23:38:08Z)
• No Language Left Behind: Scaling Human-Centered Machine Translation [69.28110770760506]
  We create datasets and models aimed at narrowing the performance gap between low and high-resource languages. We propose multiple architectural and training improvements to counteract overfitting while training on thousands of tasks. Our model achieves an improvement of 44% BLEU relative to the previous state-of-the-art.
  arXiv  Detail & Related papers  (2022-07-11T07:33:36Z)
• Comparison of Interactive Knowledge Base Spelling Correction Models for Low-Resource Languages [81.90356787324481]
  Spelling normalization for low resource languages is a challenging task because the patterns are hard to predict. This work shows a comparison of a neural model and character language models with varying amounts on target language data. Our usage scenario is interactive correction with nearly zero amounts of training examples, improving models as more data is collected.
  arXiv  Detail & Related papers  (2020-10-20T17:31:07Z)

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.