# Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models

arXiv: https://arxiv.org/abs/2305.18189 · HTML: https://mlsanity.com/paper/2305.18189/

## Key facts

- Cited by 207 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 128 references that are in the map.
- First posted on arXiv 29 May 2023.

## Builds on (most cited first, 30 of 31)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (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)
- [On Measuring Social Biases in Sentence Encoders](https://mlsanity.com/paper/1903.10561/) (2019)
- [Social Bias Frames: Reasoning about Social and Power Implications of Language](https://mlsanity.com/paper/1911.03891/) (2019)
- [Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP](https://mlsanity.com/paper/2103.00453/) (2021)
- [CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilities](https://mlsanity.com/paper/2201.06796/) (2022)
- [Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases](https://mlsanity.com/paper/2006.03955/) (2020)
- ["I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset](https://mlsanity.com/paper/2205.09209/) (2022)
- [RedditBias: A Real-World Resource for Bias Evaluation and Debiasing of Conversational Language Models](https://mlsanity.com/paper/2106.03521/) (2021)
- [Challenges in Building Intelligent Open-domain Dialog Systems](https://mlsanity.com/paper/1905.05709/) (2019)
- [Bias Out-of-the-Box: An Empirical Analysis of Intersectional Occupational Biases in Popular Generative Language Models](https://mlsanity.com/paper/2102.04130/) (2021)
- [Assessing Social and Intersectional Biases in Contextualized Word Representations](https://mlsanity.com/paper/1911.01485/) (2019)
- [Long Time No See! Open-Domain Conversation with Long-Term Persona Memory](https://mlsanity.com/paper/2203.05797/) (2022)
- [Multi-Dimensional Gender Bias Classification](https://mlsanity.com/paper/2005.00614/) (2020)
- [American == White in Multimodal Language-and-Image AI](https://mlsanity.com/paper/2207.00691/) (2022)
- [Theory-Grounded Measurement of U.S. Social Stereotypes in English Language Models](https://mlsanity.com/paper/2206.11684/) (2022)
- [Evidence for Hypodescent in Visual Semantic AI](https://mlsanity.com/paper/2205.10764/) (2022)
- [Markedness in Visual Semantic AI](https://mlsanity.com/paper/2205.11378/) (2022)
- [SODAPOP: Open-Ended Discovery of Social Biases in Social Commonsense Reasoning Models](https://mlsanity.com/paper/2210.07269/) (2022)
- [Generalized Word Shift Graphs: A Method for Visualizing and Explaining Pairwise Comparisons Between Texts](https://mlsanity.com/paper/2008.02250/) (2020)
- [Unequal Representations: Analyzing Intersectional Biases in Word Embeddings Using Representational Similarity Analysis](https://mlsanity.com/paper/2011.12086/) (2020)

## Built on by (most cited first, 30 of 207)

- [Towards Measuring the Representation of Subjective Global Opinions in Language Models](https://mlsanity.com/paper/2306.16388/) (2023)
- [SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents](https://mlsanity.com/paper/2310.11667/) (2023)
- [From Persona to Personalization: A Survey on Role-Playing Language Agents](https://mlsanity.com/paper/2404.18231/) (2024)
- [Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs](https://mlsanity.com/paper/2311.04892/) (2023)
- [Simulating Opinion Dynamics with Networks of LLM-based Agents](https://mlsanity.com/paper/2311.09618/) (2023)
- [CoMPosT: Characterizing and Evaluating Caricature in LLM Simulations](https://mlsanity.com/paper/2310.11501/) (2023)
- [A Survey on Fairness in Large Language Models](https://mlsanity.com/paper/2308.10149/) (2023)
- [From Individual to Society: A Survey on Social Simulation Driven by Large Language Model-based Agents](https://mlsanity.com/paper/2412.03563/) (2024)
- [Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Prompting](https://mlsanity.com/paper/2309.07034/) (2023)
- [Language Model Fine-Tuning on Scaled Survey Data for Predicting Distributions of Public Opinions](https://mlsanity.com/paper/2502.16761/) (2025)
- [Introducing v0.5 of the AI Safety Benchmark from MLCommons](https://mlsanity.com/paper/2404.12241/) (2024)
- [Fairness in Large Language Models: A Taxonomic Survey](https://mlsanity.com/paper/2404.01349/) (2024)
- [Rehearsal: Simulating Conflict to Teach Conflict Resolution](https://mlsanity.com/paper/2309.12309/) (2023)
- [Examining Gender and Racial Bias in Large Vision-Language Models Using a Novel Dataset of Parallel Images](https://mlsanity.com/paper/2402.05779/) (2024)
- [CULTURE-GEN: Revealing Global Cultural Perception in Language Models through Natural Language Prompting](https://mlsanity.com/paper/2404.10199/) (2024)
- [Emotionally Numb or Empathetic? Evaluating How LLMs Feel Using EmotionBench](https://mlsanity.com/paper/2308.03656/) (2023)
- [PAL: Pluralistic Alignment Framework for Learning from Heterogeneous Preferences](https://mlsanity.com/paper/2406.08469/) (2024)
- [Measuring Implicit Bias in Explicitly Unbiased Large Language Models](https://mlsanity.com/paper/2402.04105/) (2024)
- [Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Information](https://mlsanity.com/paper/2402.18144/) (2024)
- [Can Unconfident LLM Annotations Be Used for Confident Conclusions?](https://mlsanity.com/paper/2408.15204/) (2024)
- [You don't need a personality test to know these models are unreliable: Assessing the Reliability of Large Language Models on Psychometric Instruments](https://mlsanity.com/paper/2311.09718/) (2023)
- [Personas as a Way to Model Truthfulness in Language Models](https://mlsanity.com/paper/2310.18168/) (2023)
- [ChiMed-GPT: A Chinese Medical Large Language Model with Full Training Regime and Better Alignment to Human Preferences](https://mlsanity.com/paper/2311.06025/) (2023)
- [Opportunities and Risks of LLMs for Scalable Deliberation with Polis](https://mlsanity.com/paper/2306.11932/) (2023)
- [The Prompt Makes the Person(a): A Systematic Evaluation of Sociodemographic Persona Prompting for Large Language Models](https://mlsanity.com/paper/2507.16076/) (2025)
- [Angry Men, Sad Women: Large Language Models Reflect Gendered Stereotypes in Emotion Attribution](https://mlsanity.com/paper/2403.03121/) (2024)
- [Who's asking? User personas and the mechanics of latent misalignment](https://mlsanity.com/paper/2406.12094/) (2024)
- [MBBQ: A Dataset for Cross-Lingual Comparison of Stereotypes in Generative LLMs](https://mlsanity.com/paper/2406.07243/) (2024)
- [The Homogenizing Effect of Large Language Models on Human Expression and Thought](https://mlsanity.com/paper/2508.01491/) (2025)
- [Aligning Language Models with Demonstrated Feedback](https://mlsanity.com/paper/2406.00888/) (2024)

## Abstract (arXiv)

To recognize and mitigate harms from large language models (LLMs), we need to understand the prevalence and nuances of stereotypes in LLM outputs. Toward this end, we present Marked Personas, a prompt-based method to measure stereotypes in LLMs for intersectional demographic groups without any lexicon or data labeling. Grounded in the sociolinguistic concept of markedness (which characterizes explicitly linguistically marked categories versus unmarked defaults), our proposed method is twofold: 1) prompting an LLM to generate personas, i.e., natural language descriptions, of the target demographic group alongside personas of unmarked, default groups; 2) identifying the words that significantly distinguish personas of the target group from corresponding unmarked ones. We find that the portrayals generated by GPT-3.5 and GPT-4 contain higher rates of racial stereotypes than human-written portrayals using the same prompts. The words distinguishing personas of marked (non-white, non-male) groups reflect patterns of othering and exoticizing these demographics. An intersectional lens further reveals tropes that dominate portrayals of marginalized groups, such as tropicalism and the hypersexualization of minoritized women. These representational harms have concerning implications for downstream applications like story generation.