# An Analysis of Social Biases Present in BERT Variants Across Multiple Languages

arXiv: https://arxiv.org/abs/2211.14402 · HTML: https://mlsanity.com/paper/2211.14402/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 24 references that are in the map.
- First posted on arXiv 25 Nov 2022.

## Builds on (most cited first, 13 of 13)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering](https://mlsanity.com/paper/2012.00955/) (2020)
- [Persistent Anti-Muslim Bias in Large Language Models](https://mlsanity.com/paper/2101.05783/) (2021)
- [Towards Understanding and Mitigating Social Biases in Language Models](https://mlsanity.com/paper/2106.13219/) (2021)
- [Mitigating Gender Bias in Natural Language Processing: Literature Review](https://mlsanity.com/paper/1906.08976/) (2019)
- [Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly](https://mlsanity.com/paper/1911.03343/) (2019)
- [ParsBERT: Transformer-based Model for Persian Language Understanding](https://mlsanity.com/paper/2005.12515/) (2020)
- [Mitigating Language-Dependent Ethnic Bias in BERT](https://mlsanity.com/paper/2109.05704/) (2021)
- [Understanding by Understanding Not: Modeling Negation in Language Models](https://mlsanity.com/paper/2105.03519/) (2021)
- [Gender Bias in Multilingual Embeddings and Cross-Lingual Transfer](https://mlsanity.com/paper/2005.00699/) (2020)
- [GREEK-BERT: The Greeks visiting Sesame Street](https://mlsanity.com/paper/2008.12014/) (2020)

## Built on by (most cited first, 6 of 6)

- [Having Beer after Prayer? Measuring Cultural Bias in Large Language Models](https://mlsanity.com/paper/2305.14456/) (2023)
- [Fairness in Language Models Beyond English: Gaps and Challenges](https://mlsanity.com/paper/2302.12578/) (2023)
- [Social Bias in Large Language Models For Bangla: An Empirical Study on Gender and Religious Bias](https://mlsanity.com/paper/2407.03536/) (2024)
- [Addressing Bias in Generative AI: Challenges and Research Opportunities in Information Management](https://mlsanity.com/paper/2502.10407/) (2025)
- [NER-to-MRC: Named-Entity Recognition Completely Solving as Machine Reading Comprehension](https://mlsanity.com/paper/2305.03970/) (2023)
- [StyleBART: Decorate Pretrained Model with Style Adapters for Unsupervised Stylistic Headline Generation](https://mlsanity.com/paper/2310.17743/) (2023)

## Abstract (arXiv)

Although large pre-trained language models have achieved great success in many NLP tasks, it has been shown that they reflect human biases from their pre-training corpora. This bias may lead to undesirable outcomes when these models are applied in real-world settings. In this paper, we investigate the bias present in monolingual BERT models across a diverse set of languages (English, Greek, and Persian). While recent research has mostly focused on gender-related biases, we analyze religious and ethnic biases as well and propose a template-based method to measure any kind of bias, based on sentence pseudo-likelihood, that can handle morphologically complex languages with gender-based adjective declensions. We analyze each monolingual model via this method and visualize cultural similarities and differences across different dimensions of bias. Ultimately, we conclude that current methods of probing for bias are highly language-dependent, necessitating cultural insights regarding the unique ways bias is expressed in each language and culture (e.g. through coded language, synecdoche, and other similar linguistic concepts). We also hypothesize that higher measured social biases in the non-English BERT models correlate with user-generated content in their training.