mlsanityData to 7 Oct 2026

A Friendly Face: Do Text-to-Image Systems Rely on Stereotypes when the Input is Under-Specified?

arXiv 2302.07159 ↗First posted 14 Feb 2023cs.CY cs.CLcited by 25

Not on the map. Why?

Abstract (arXiv)

As text-to-image systems continue to grow in popularity with the general public, questions have arisen about bias and diversity in the generated images. Here, we investigate properties of images generated in response to prompts which are visually under-specified, but contain salient social attributes (e.g., 'a portrait of a threatening person' versus 'a portrait of a friendly person'). Grounding our work in social cognition theory, we find that in many cases, images contain similar demographic biases to those reported in the stereotype literature. However, trends are inconsistent across different models and further investigation is warranted.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 20 of its 34 references that are in the map. First posted on arXiv 14 Feb 2023.

Builds on

20 of 34 references.

References · 20

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  5. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  6. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  7. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  8. 2018Women also Snowboard: Overcoming Bias in Captioning Modelscited by 236
  9. 2021Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implicationscited by 128
  10. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  11. 2022How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?cited by 92
  12. 2020Saving Face: Investigating the Ethical Concerns of Facial Recognition Auditingcited by 89
8 more
  1. 2015Seeing through the Human Reporting Bias: Visual Classifiers from Noisy Human-Centric Labelscited by 64
  2. 2022American == White in Multimodal Language-and-Image AIcited by 39
  3. 2022Theory-Grounded Measurement of U.S. Social Stereotypes in English Language Modelscited by 35
  4. 2022Underspecification in Scene Description-to-Depiction Taskscited by 30
  5. 2020Diversity and Inclusion Metrics in Subset Selectioncited by 27
  6. 2016Stereotyping and Bias in the Flickr30K Datasetcited by 25
  7. 2021Computer Vision and Conflicting Values: Describing People with Automated Alt Textcited by 14
  8. 2022Assessing Dataset Bias in Computer Visioncited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 9; 2024: 10; 2025: 4; 2026: 2.

25 papers.

Sort:

By area:

  1. 2023Unified Concept Editing in Diffusion Modelscited by 377
  2. 2023Editing Implicit Assumptions in Text-to-Image Diffusion Modelscited by 128
  3. 2023The Bias Amplification Paradox in Text-to-Image Generationcited by 69
  4. 2024Survey of Bias In Text-to-Image Generation: Definition, Evaluation, and Mitigationcited by 66
  5. 2024UnsafeBench: Benchmarking Image Safety Classifiers on Real-World and AI-Generated Imagescited by 52
  6. 2023Inspecting the Geographical Representativeness of Images from Text-to-Image Modelscited by 44
  7. 2023'Person' == Light-skinned, Western Man, and Sexualization of Women of Color: Stereotypes in Stable Diffusioncited by 33
  8. 2023ReFACT: Updating Text-to-Image Models by Editing the Text Encodercited by 32
  9. 2023Fair Text-to-Image Diffusion via Fair Mappingcited by 31
  10. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  11. 2024Mapping the Ethics of Generative AI: A Comprehensive Scoping Reviewcited by 19
  12. 2024BiasDora: Exploring Hidden Biased Associations in Vision-Language Modelscited by 12
13 more
  1. 2024A Unified Framework and Dataset for Assessing Societal Bias in Vision-Language Modelscited by 11
  2. 2024Social Perception of Faces in a Vision-Language Modelcited by 10
  3. 2023Analysing Gender Bias in Text-to-Image Models using Object Detectioncited by 7
  4. 2024Image-Perfect Imperfections: Safety, Bias, and Authenticity in the Shadow of Text-To-Image Model Evolutioncited by 7
  5. 2024The Male CEO and the Female Assistant: Evaluation and Mitigation of Gender Biases in Text-To-Image Generation of Dual Subjectscited by 6
  6. 2025Bias Amplification in Stable Diffusion's Representation of Stigma Through Skin Tones and Their Homogeneitycited by 5
  7. 2025Do Existing Testing Tools Really Uncover Gender Bias in Text-to-Image Models?cited by 4
  8. 2025Multi-Group Proportional Representation for Text-to-Image Modelscited by 3
  9. 2024Explainable Search and Discovery of Visual Cultural Heritage Collections with Multimodal Large Language Modelscited by 1
  10. 2024Implicit Priors Editing in Stable Diffusion via Targeted Token Adjustmentcited by 1
  11. 2025The Digital Mirror: Gender Bias and Occupational Stereotypes in AI-Generated Imagescited by 1
  12. 2026Gender Artifacts from Art History to Text-to-Image Generationcited by 1
  13. 2026IMPLICIT-Bench: Measuring Implicit Bias in Text-to-Image Models under Neutral Promptscited by 0