# Surfacing Subtle Stereotypes: A Multilingual, Debate-Oriented Evaluation of Modern LLMs

arXiv: https://arxiv.org/abs/2511.01187 · HTML: https://mlsanity.com/paper/2511.01187/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 46 references that are in the map.
- First posted on arXiv 3 Nov 2025.

## Builds on (most cited first, 30 of 30)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [Persistent Anti-Muslim Bias in Large Language Models](https://mlsanity.com/paper/2101.05783/) (2021)
- [AraBERT: Transformer-based Model for Arabic Language Understanding](https://mlsanity.com/paper/2003.00104/) (2020)
- [Societal Biases in Language Generation: Progress and Challenges](https://mlsanity.com/paper/2105.04054/) (2021)
- [ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic](https://mlsanity.com/paper/2101.01785/) (2020)
- [The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts](https://mlsanity.com/paper/2401.13136/) (2024)
- [Does Refusal Training in LLMs Generalize to the Past Tense?](https://mlsanity.com/paper/2407.11969/) (2024)
- [Large Language Model Safety: A Holistic Survey](https://mlsanity.com/paper/2412.17686/) (2024)
- [AI's Regimes of Representation: A Community-centered Study of Text-to-Image Models in South Asia](https://mlsanity.com/paper/2305.11844/) (2023)
- [IndiBias: A Benchmark Dataset to Measure Social Biases in Language Models for Indian Context](https://mlsanity.com/paper/2403.20147/) (2024)
- [Who is Responsible? The Data, Models, Users or Regulations? A Comprehensive Survey on Responsible Generative AI for a Sustainable Future](https://mlsanity.com/paper/2502.08650/) (2025)
- [The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understanding](https://mlsanity.com/paper/2502.08946/) (2025)
- [Multi-OphthaLingua: A Multilingual Benchmark for Assessing and Debiasing LLM Ophthalmological QA in LMICs](https://mlsanity.com/paper/2412.14304/) (2024)
- [OASIS Uncovers: High-Quality T2I Models, Same Old Stereotypes](https://mlsanity.com/paper/2501.00962/) (2025)
- [Implicit Discourse Relation Classification For Nigerian Pidgin](https://mlsanity.com/paper/2406.18776/) (2024)
- [Exploring Bias and Prediction Metrics to Characterise the Fairness of Machine Learning for Equity-Centered Public Health Decision-Making: A Narrative Review](https://mlsanity.com/paper/2408.13295/) (2024)
- [Low-Resource Cross-Lingual Adaptive Training for Nigerian Pidgin](https://mlsanity.com/paper/2307.00382/) (2023)
- [Modeling Orthographic Variation Improves NLP Performance for Nigerian Pidgin](https://mlsanity.com/paper/2404.18264/) (2024)
- [Desert Camels and Oil Sheikhs: Arab-Centric Red Teaming of Frontier LLMs](https://mlsanity.com/paper/2410.24049/) (2024)

## Built on by (most cited first, 3 of 3)

- [LLMpedia: A Transparent Framework to Materialize an LLM's Encyclopedic Knowledge at Scale](https://mlsanity.com/paper/2603.24080/) (2026)
- [LLM Safety Alignment in Low-Resource Languages: A Systematic Literature Review](https://mlsanity.com/paper/2608.14626/) (2026)
- [Noise Out, Bias In: Targeted Bias Injection in Diffusion Language Models via Closed-Loop Activation Steering](https://mlsanity.com/paper/2610.05894/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are widely deployed for open-ended communication, yet most bias evaluations still rely on English, classification-style tasks. We introduce corpusname, a new multilingual, debate-style benchmark designed to reveal how narrative bias appears in realistic generative settings. Our dataset includes 8,400 structured debate prompts spanning four sensitive domains -- Women's Rights, Backwardness, Terrorism, and Religion -- across seven languages ranging from high-resource (English, Chinese) to low-resource (Swahili, Nigerian Pidgin). Using four flagship models (GPT-4o, Claude~3.5~Haiku, DeepSeek-Chat, and LLaMA-3-70B), we generate over 100,000 debate responses and automatically classify which demographic groups are assigned stereotyped versus modern roles. Results show that all models reproduce entrenched stereotypes despite safety alignment: Arabs are overwhelmingly linked to Terrorism and Religion (≥89%), Africans to socioeconomic ``backwardness'' (up to 77%), and Western groups are consistently framed as modern or progressive. Biases grow sharply in lower-resource languages, revealing that alignment trained primarily in English does not generalize globally. Our findings highlight a persistent divide in multilingual fairness: current alignment methods reduce explicit toxicity but fail to prevent biased outputs in open-ended contexts. We release our corpusname benchmark and analysis framework to support the next generation of multilingual bias evaluation and safer, culturally inclusive model alignment.