Social Bias in Multilingual Language Models: A Survey
Not on the map. Why?
Abstract (arXiv)
Pretrained multilingual models exhibit the same social bias as models processing English texts. This systematic review analyzes emerging research that extends bias evaluation and mitigation approaches into multilingual and non-English contexts. We examine these studies with respect to linguistic diversity, cultural awareness, and their choice of evaluation metrics and mitigation techniques. Our survey illuminates gaps in the field's dominant methodological design choices (e.g., preference for certain languages, scarcity of multilingual mitigation experiments) while cataloging common issues encountered and solutions implemented in adapting bias benchmarks across languages and cultures. Drawing from the implications of our findings, we chart directions for future research that can reinforce the multilingual bias literature's inclusivity, cross-cultural appropriateness, and alignment with state-of-the-art NLP advancements.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 59 of its 110 references that are in the map. First posted on arXiv 27 Aug 2025.
Builds on
59 of 110 references.
References · 59
- 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,121
- 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 956
- 2016Semantics derived automatically from language corpora contain human-like biasescited by 901
- 2020The State and Fate of Linguistic Diversity and Inclusion in the NLP Worldcited by 798
- 2020CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Modelscited by 710
- 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 701
- 2024Aya Model: An Instruction Finetuned Open-Access Multilingual Language Modelcited by 295
- 2021Quality at a Glance: An Audit of Web-Crawled Multilingual Datasetscited by 243
- 2022Challenges and Strategies in Cross-Cultural NLPcited by 218
- 2023Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Promptingcited by 165
- 2022"I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Datasetcited by 162
- 2024LLaMA Beyond English: An Empirical Study on Language Capability Transfercited by 95
38 more
- 2021Mitigating Language-Dependent Ethnic Bias in BERTcited by 87
- 2024A Survey on Multilingual Large Language Models: Corpora, Alignment, and Biascited by 84
- 2021Quantifying Social Biases in NLP: A Generalization and Empirical Comparison of Extrinsic Fairness Metricscited by 83
- 2022COLD: A Benchmark for Chinese Offensive Language Detectioncited by 76
- 2023KoBBQ: Korean Bias Benchmark for Question Answeringcited by 71
- 2024The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harmcited by 70
- 2022Gender Bias in Masked Language Models for Multiple Languagescited by 60
- 2022Re-contextualizing Fairness in NLP: The Case of Indiacited by 55
- 2023Nationality Bias in Text Generationcited by 50
- 2023Click: Controllable Text Generation with Sequence Likelihood Contrastive Learningcited by 41
- 2023Flames: Benchmarking Value Alignment of LLMs in Chinesecited by 37
- 2022Towards Identifying Social Bias in Dialog Systems: Frame, Datasets, and Benchmarkscited by 36
- 2023Comparing Biases and the Impact of Multilingual Training across Multiple Languagescited by 36
- 2023COPAL-ID: Indonesian Language Reasoning with Local Culture and Nuancescited by 35
- 2024IndiBias: A Benchmark Dataset to Measure Social Biases in Language Models for Indian Contextcited by 32
- 2023On Evaluating and Mitigating Gender Biases in Multilingual Settingscited by 31
- 2024Preference Tuning For Toxicity Mitigation Generalizes Across Languagescited by 31
- 2021Socially Aware Bias Measurements for Hindi Language Representationscited by 29
- 2023On the Independence of Association Bias and Empirical Fairness in Language Modelscited by 29
- 2023CHBias: Bias Evaluation and Mitigation of Chinese Conversational Language Modelscited by 28
- 2023Sociodemographic Bias in Language Models: A Survey and Forward Pathcited by 28
- 2023Multilingual Holistic Bias: Extending Descriptors and Patterns to Unveil Demographic Biases in Languages at Scalecited by 24
- 2023KoSBi: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Applicationcited by 24
- 2024SeeGULL Multilingual: a Dataset of Geo-Culturally Situated Stereotypescited by 24
- 2023Global Voices, Local Biases: Socio-Cultural Prejudices across Languagescited by 23
- 2022Deep Learning on a Healthy Data Diet: Finding Important Examples for Fairnesscited by 21
- 2020AraWEAT: Multidimensional Analysis of Biases in Arabic Word Embeddingscited by 20
- 2024From One to Many: Expanding the Scope of Toxicity Mitigation in Language Modelscited by 18
- 2024Akal Badi ya Bias: An Exploratory Study of Gender Bias in Hindi Language Technologycited by 17
- 2020Evaluating Bias In Dutch Word Embeddingscited by 15
- 2023Unraveling Downstream Gender Bias from Large Language Models: A Study on AI Educational Writing Assistancecited by 15
- 2022Mapping the Multilingual Margins: Intersectional Biases of Sentiment Analysis Systems in English, Spanish, and Arabiccited by 13
- 2024Quite Good, but Not Enough: Nationality Bias in Large Language Models -- A Case Study of ChatGPTcited by 13
- 2022JASMINE: Arabic GPT Models for Few-Shot Learningcited by 11
- 2023Measuring Gender Bias in West Slavic Language Modelscited by 10
- 2023In-Depth Look at Word Filling Societal Bias Measurescited by 8
- 2022Measuring Harmful Representations in Scandinavian Language Modelscited by 7
- 2024Crossing Linguistic Horizons: Finetuning and Comprehensive Evaluation of Vietnamese Large Language Modelscited by 6
50 most cited of 59 All 59 (JSON)
Built on by
Citing papers per year: 2026: 2.
2 papers.