Mitigation of Gender and Ethnicity Bias in AI-Generated Stories through Model Explanations
Not on the map. Why?
Abstract (arXiv)
Language models have been shown to propagate social bias through their output, particularly in the representation of gender and ethnicity. This paper investigates gender and ethnicity biases in AI-generated occupational stories. Representation biases are measured before and after applying our proposed mitigation strategy, Bias Analysis and Mitigation through Explanation (BAME), revealing improvements in demographic representation ranging from 2% to 20%. BAME leverages model-generated explanations to inform targeted prompt engineering, effectively reducing biases without modifying model parameters. By analyzing stories generated across 25 occupational groups, three large language models (Claude 3.5 Sonnet, Llama 3.1 70B Instruct, and GPT-4 Turbo), and multiple demographic dimensions, we identify persistent patterns of overrepresentation and underrepresentation linked to training data stereotypes. Our findings demonstrate that guiding models with their own internal reasoning mechanisms can significantly enhance demographic parity, thereby contributing to the development of more transparent generative AI systems.
Key facts. Cited by 1 paper in the map (data through 13 Oct 2026). Builds on 34 of its 63 references that are in the map. First posted on arXiv 3 Sep 2025.
Builds on
34 of 63 references.
References · 34
- 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,367
- 2019A Survey on Bias and Fairness in Machine Learningcited by 1,915
- 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,309
- 2021Ethical and social risks of harm from Language Modelscited by 964
- 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
- 2023Bias and Fairness in Large Language Models: A Surveycited by 636
- 2021BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generationcited by 411
- 2019Mitigating Gender Bias in Natural Language Processing: Literature Reviewcited by 287
- 2022DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Modelscited by 230
- 2023"Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letterscited by 198
- 2019A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cyclecited by 193
- 2020Algorithmic Fairnesscited by 143
22 more
- 2021Bias Out-of-the-Box: An Empirical Analysis of Intersectional Occupational Biases in Popular Generative Language Modelscited by 141
- 2021Debiasing Pre-trained Contextualised Embeddingscited by 107
- 2024Escalation Risks from Language Models in Military and Diplomatic Decision-Makingcited by 62
- 2024Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypescited by 59
- 2015On the relation between accuracy and fairness in binary classificationcited by 47
- 2022Representation Bias in Data: A Survey on Identification and Resolution Techniquescited by 37
- 2024Large Language Models Portray Socially Subordinate Groups as More Homogeneous, Consistent with a Bias Observed in Humanscited by 31
- 2023Fairness and Bias in Algorithmic Hiring: a Multidisciplinary Surveycited by 30
- 2024LLMs are Biased Teachers: Evaluating LLM Bias in Personalized Educationcited by 28
- 2023Tailoring Language Generation Models under Total Variation Distancecited by 27
- 2019Fairness in representation: quantifying stereotyping as a representational harmcited by 26
- 2023Multimodal Composite Association Score: Measuring Gender Bias in Generative Multimodal Modelscited by 21
- 2023Perspectives on the Social Impacts of Reinforcement Learning with Human Feedbackcited by 15
- 2024The Pursuit of Fairness in Artificial Intelligence Models: A Surveycited by 13
- 2020Bias in Data-driven AI Systems -- An Introductory Surveycited by 10
- 2023BAD: BiAs Detection for Large Language Models in the context of candidate screeningcited by 8
- 2024AXOLOTL: Fairness through Assisted Self-Debiasing of Large Language Model Outputscited by 7
- 2025Addressing Bias in Generative AI: Challenges and Research Opportunities in Information Managementcited by 7
- 2023Gender Bias in Multimodal Models: A Transnational Feminist Approach Considering Geographical Region and Culturecited by 5
- 2024Unboxing Occupational Bias: Grounded Debiasing of LLMs with U.S. Labor Datacited by 5
- 2024When AI Eats Itself: On the Caveats of AI Autophagycited by 4
- 2024The Life Cycle of Large Language Models: A Review of Biases in Educationcited by 3
Built on by
Citing papers per year: 2026: 1.
1 paper.