Simulating Identity, Propagating Bias: Abstraction and Stereotypes in LLM-Generated Text
Not on the map. Why?
Abstract (arXiv)
Persona-prompting is a growing strategy to steer LLMs toward simulating particular perspectives or linguistic styles through the lens of a specified identity. While this method is often used to personalize outputs, its impact on how LLMs represent social groups remains underexplored. In this paper, we investigate whether persona-prompting leads to different levels of linguistic abstraction - an established marker of stereotyping - when generating short texts linking socio-demographic categories with stereotypical or non-stereotypical attributes. Drawing on the Linguistic Expectancy Bias framework, we analyze outputs from six open-weight LLMs under three prompting conditions, comparing 11 persona-driven responses to those of a generic AI assistant. To support this analysis, we introduce Self-Stereo, a new dataset of self-reported stereotypes from Reddit. We measure abstraction through three metrics: concreteness, specificity, and negation. Our results highlight the limits of persona-prompting in modulating abstraction in language, confirming criticisms about the ecology of personas as representative of socio-demographic groups and raising concerns about the risk of propagating stereotypes even when seemingly evoking the voice of a marginalized group.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 21 of its 42 references that are in the map. First posted on arXiv 10 Sep 2025.
Builds on
21 of 42 references.
References · 21
- 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 911
- 2020CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Modelscited by 703
- 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 683
- 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
- 2024Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inferencecited by 527
- 2018Gender Bias in Coreference Resolutioncited by 438
- 2023Toxicity in ChatGPT: Analyzing Persona-assigned Language Modelscited by 356
- 2023Character-LLM: A Trainable Agent for Role-Playingcited by 335
- 2023Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Modelscited by 204
- 2024Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalizationcited by 203
- 2023Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMscited by 191
- 2024Quantifying the Persona Effect in LLM Simulationscited by 164
9 more
- 2023Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Promptingcited by 80
- 2024Evaluating Large Language Model Biases in Persona-Steered Generationcited by 78
- 2023SeeGULL: A Stereotype Benchmark with Broad Geo-Cultural Coverage Leveraging Generative Modelscited by 66
- 2023On the steerability of large language models toward data-driven personascited by 53
- 2023The Ecological Fallacy in Annotation: Modelling Human Label Variation goes beyond Sociodemographicscited by 45
- 2024Bias in Language Models: Beyond Trick Tests and Toward RUTEd Evaluationcited by 41
- 2024Angry Men, Sad Women: Large Language Models Reflect Gendered Stereotypes in Emotion Attributioncited by 39
- 2024Persona Setting Pitfall: Persistent Outgroup Biases in Large Language Models Arising from Social Identity Adoptioncited by 9
- 2025Prompt Engineering: How Prompt Vocabulary affects Domain Knowledgecited by 2
Built on by
Citing papers per year: 2026: 2.
2 papers.