Social Bias in Language Models

Overview

This topic studies how language models reflect or produce social disparities tied to traits such as gender, race, disability, religion, class, and language variety, in both text generation and consequential tasks. Researchers use counterfactual prompts, stereotype and decision-making benchmarks, human and community judgments, and audits across languages, dialects, modalities, and multi-turn interactions. Mitigation approaches include data augmentation and fine-tuning, prompting and retrieval, decoding-time controls, and interventions on model weights or internal representations. A central challenge is that measured bias changes with the task and context, while interventions can reduce visible disparities without removing underlying associations or can shift harms and degrade useful capabilities.

Latest papers

CardsList

217 papers20 papers in the last four weeks, level with the four weeks before. 0.2% of all new papers.

  1. What Does a Bias Head Measure? Non-Identifiability in Head-Level Bias Attribution

    Nov 17, 2026S. M Seefat AlamAttention Head AnalysisLLM Auditing

  2. Beyond Attribute-Centric Fairness: Modelling and Mitigating Unified Bias Representation in Omni-modal Large Language Models

    Nov 17, 2026Runqi Lin, Suqin Yuan, Zihao Fu +1Social Bias in Language ModelsLanguage Model Debiasing

  3. A Crowd-Audited Atlas of Implicit Bias in LLM-Powered Agents

    Nov 16, 2026Jason Alexander, Yuxuan Li, Michael Licamele +2Social Bias in Language ModelsDemographic Bias in Language Models

  4. Revealing the Biases of Large Language Models By Inferring Prior Distributions

    Nov 13, 2026Joshua Ryo Cheng, Thomas L. Griffiths, Jake C. SnellSocial Bias in Language ModelsLanguage Model Bias Evaluation

  5. Fairness in Explainable AI Narratives: Decomposing Social Disparities in LLM-Generated Explanations

    Nov 9, 2026Wannes Van den Bulck, Sofie Goethals, David MartensSocial Bias in Language ModelsFaithfulness of Language Model Explanations

  6. BDBBQ: A Human-AI Collaborative Benchmark for Bengali Social Bias in LLMs

    Nov 9, 2026Provat Saha, Ashika Habib Khan, Tahseen Tanha +2Social Bias in Language ModelsLanguage Model Bias Evaluation

  7. WarBias: A Bilingual LLM Benchmark of Stereotypes About Internally Displaced People and Veterans

    Nov 8, 2026Viktoriia Makovska, Tetiana Zakharchenko, Oleksandra OstafiichukSocial Bias in Language ModelsLanguage Model Bias Evaluation

  8. CompAIssionate Other: Identity Assumptions in LLM Responses to Distress

    Nov 7, 2026Drishti Goel, Ethan M. Nguyen, Violeta J. Rodriguez +1Social Bias in Language ModelsMental Health Conversational Agents

  9. Safer for Whom? A Systematic Exploration of AI Safety in Eating Disorder Interactions

    Nov 7, 2026Jinyuan Wang, Minh Duc Chu, Frances Ross +2LLM SafetyLanguage Model Safety Evaluation

  10. Policy Over Purpose: Auditing LLM Content Moderation of Reclaimed Slur Usage

    Nov 6, 2026Christina Chance, Kai-Wei ChangContent ModerationLLM Auditing

  11. The Male Abstraction Bias in Internet Culture and Large Language Models

    Nov 3, 2026Douglas Guilbeault, Bhargav Srinivasa DesikanGender Bias in Language ModelsSocial Bias in Language Models

  12. User Attributes and Political Ideology Shape LLM Receptiveness in Online Political Debates

    Nov 2, 2026Akram Elbouanani, Yousra Taleb, Gianluca Demartini +1Political Bias in Language ModelsSocial Bias in Language Models

  13. Architecting Fairness-Aware LLM Interactions

    Oct 31, 2026Alessandra Parziale, Valeria Pontillo, Fabio Palomba +1Algorithmic FairnessSocial Bias in Language Models

  14. Multi-Test Nation Bias Benchmarking of LLMs with Explicit Unbiased Baselines

    Oct 30, 2026Jonghyeon Choi, Yeonjun Choi, Beakcheol JangPolitical Bias in Language ModelsLanguage Model Bias Evaluation

  15. Language Models Show a "Women-Are-Wonderful" Effect in Blame Attribution That is Not Just Leniency

    Oct 30, 2026Anika Sharma, Snehalkumar 'Neil' S. Gaikwad, Snigdha Chaturvedi +1Gender Bias in Language ModelsSocial Bias in Language Models

  16. Qistas: A Pre-Registered, Portable Protocol for Fairness Evaluation of Large Language Models

    Oct 29, 2026Nawar S. Alseelawi, Mustafa S. AljumailyLanguage Model Bias EvaluationCross-Cultural Language Model Evaluation

  17. RAGing Against Misogyny: Improving Implicit Misogyny Explanations

    Oct 26, 2026Ivana Crescenzi, Arianna Muti, Maarten Sap +2Implicit BiasSocial Bias in Language Models

  18. How Does Post-Training Shape LLM Value Preferences?

    Oct 23, 2026Yang Liu, Shaoyang Xu, Masahiro Kaneko +3LLM Post-TrainingSocial Bias in Language Models

  19. Investigating the Impact of Learner Errors on Social Bias in Large Language Models

    Oct 21, 2026Federico Marcuzzi, Martin Christoph, Iryna Gurevych +1Social Bias in Language ModelsLanguage Model Bias Evaluation

  20. Effects of Model Scale and Quantization on Social Bias and Abstention in Question Answering

    Oct 21, 2026Kuniko Paxton, Ifeoluwa Wuraola, Dhavalkumar ThakkerSocial Bias in Language ModelsLanguage Model Bias Evaluation

  21. Beyond Model Weights: Sampling-Dependent Fairness in Diffusion Language Models

    Oct 20, 2026Koshiro Saito, Daisuke Oba, Danushka Bollegala +1Social Bias in Language ModelsDiffusion Language Models

  22. Do LLMs Really Lack Difference Awareness? Uncovering Heuristic Substitution in Group-Differentiating Fairness

    Oct 18, 2026Jingshen Zhang, Bo Wang, Yanlin Fu +4Algorithmic FairnessSocial Bias in Language Models

  23. Cultural (In)Competence: Investigating Culture-Specific Stereotype Detection Capabilities of LLMs

    Oct 15, 2026Maximilian Spliethöver, Jannes Lötfering, Dominik Bachmann +4Social Bias in Language ModelsCultural Bias in Language Models

  24. Name-Bias in Large Language Models

    Oct 15, 2026Kalyan Bhetwal, Ceren Mert, Ian WhitehouseSocial Bias in Language ModelsDemographic Bias in Language Models

  25. Resisting Harmful Instructions through Training-Free Weight Editing

    Oct 15, 2026Ali Ebrahimpour-Boroojeny, Xiaocong Yang, Samuel Schapiro +4LLM SafetyLLM Safety Alignment

  26. Social Priors: Descriptions of AI Users Shape How Language Models Trust and Cooperate

    Oct 12, 2026Harsh Kumar, George Eilender, Ashton AndersonSocial Bias in Language ModelsBelief Updating in LLMs

  27. AI Representations of Mental Health Stigma: Quantifying the Gap Between AI Estimations and Human Perceptions Across Gender Groups Using GPT-Based LLMs

    Oct 12, 2026Soojong Kim, Hye Min Kim, Kwanho Kim +1Social Bias in Language ModelsGender Bias in Language Models