mlsanityData to 7 Oct 2026

Stable Diffusion Exposed: Gender Bias from Prompt to Image

arXiv 2312.03027 ↗First posted 5 Dec 2023cs.CVcited by 25

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

Several studies have raised awareness about social biases in image generative models, demonstrating their predisposition towards stereotypes and imbalances. This paper contributes to this growing body of research by introducing an evaluation protocol that analyzes the impact of gender indicators at every step of the generation process on Stable Diffusion images. Leveraging insights from prior work, we explore how gender indicators not only affect gender presentation but also the representation of objects and layouts within the generated images. Our findings include the existence of differences in the depiction of objects, such as instruments tailored for specific genders, and shifts in overall layouts. We also reveal that neutral prompts tend to produce images more aligned with masculine prompts than their feminine counterparts. We further explore where bias originates through representational disparities and how it manifests in the images via prompt-image dependencies, and provide recommendations for developers and users to mitigate potential bias in image generation.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 70 of its 91 references that are in the map. First posted on arXiv 5 Dec 2023.

Builds on

70 of 91 references.

Closest ancestors · 11

Same direction, newest first

  1. 2023GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment17 Oct
  2. 2023TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition24 Jul
  3. 2023Inspecting the Geographical Representativeness of Images from Text-to-Image Models18 May
  4. 2023Multimodal Composite Association Score: Measuring Gender Bias in Generative Multimodal Models26 Apr
  5. 2023HRS-Bench: Holistic, Reliable and Scalable Benchmark for Text-to-Image Models11 Apr
  6. 2022Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models7 Dec
  7. 2022What the DAAM: Interpreting Stable Diffusion Using Cross Attention10 Oct
  8. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation25 Aug
  9. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding23 May
  10. 2022Hierarchical Text-Conditional Image Generation with CLIP Latents13 Apr
  11. 2022DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models8 Feb

Other references · 59

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  6. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  7. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  8. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  9. 2023Segment Anythingcited by 8,158
  10. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  11. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  12. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
38 more
  1. 2022Generative Adversarial Networkscited by 5,603
  2. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  3. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,274
  4. 2016Improved Techniques for Training GANscited by 2,938
  5. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  6. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,388
  7. 2014CIDEr: Consensus-based Image Description Evaluationcited by 2,027
  8. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,287
  9. 2016Semantics derived automatically from language corpora contain human-like biasescited by 894
  10. 2016Generative Adversarial Text to Image Synthesiscited by 735
  11. 2023Extracting Training Data from Diffusion Modelscited by 723
  12. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 691
  13. 2021ClipCap: CLIP Prefix for Image Captioningcited by 557
  14. 2023Erasing Concepts from Diffusion Modelscited by 530
  15. 2017Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraintscited by 420
  16. 2020TextCaps: a Dataset for Image Captioning with Reading Comprehensioncited by 398
  17. 2023Unified Concept Editing in Diffusion Modelscited by 377
  18. 2021Multimodal datasets: misogyny, pornography, and malignant stereotypescited by 326
  19. 2022CogView2: Faster and Better Text-to-Image Generation via Hierarchical Transformerscited by 305
  20. 2023Diffusion Self-Guidance for Controllable Image Generationcited by 305
  21. 2023Recognize Anything: A Strong Image Tagging Modelcited by 305
  22. 2022Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scalecited by 262
  23. 2023Stable Bias: Analyzing Societal Representations in Diffusion Modelscited by 225
  24. 2023DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Modelscited by 162
  25. 2020DF-GAN: A Simple and Effective Baseline for Text-to-Image Synthesiscited by 161
  26. 2023Holistic Evaluation of Text-To-Image Modelscited by 160
  27. 2023Social Biases through the Text-to-Image Generation Lenscited by 126
  28. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  29. 2023The Bias Amplification Paradox in Text-to-Image Generationcited by 69
  30. 2023Toward Verifiable and Reproducible Human Evaluation for Text-to-Image Generationcited by 66
  31. 2023ITI-GEN: Inclusive Text-to-Image Generationcited by 64
  32. 2023Into the LAIONs Den: Investigating Hate in Multimodal Datasetscited by 54
  33. 2023Uncurated Image-Text Datasets: Shedding Light on Demographic Biascited by 51
  34. 2023TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Modelscited by 51
  35. 2023Evaluating Data Attribution for Text-to-Image Modelscited by 45
  36. 2022Contrastive Language-Vision AI Models Pretrained on Web-Scraped Multimodal Data Exhibit Sexual Objectification Biascited by 43
  37. 2022Gender and Racial Bias in Visual Question Answering Datasetscited by 39
  38. 2022American == White in Multimodal Language-and-Image AIcited by 39

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 7; 2025: 12; 2026: 6.

25 papers.

Sort:

By area:

  1. 2024Would Deep Generative Models Amplify Bias in Future Models?cited by 24
  2. 2024Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)cited by 15
  3. 2025Generating an Image From 1,000 Words: Enhancing Text-to-Image With Structured Captionscited by 13
  4. 2025A Large Scale Analysis of Gender Biases in Text-to-Image Generative Modelscited by 12
  5. 2024Replication in Visual Diffusion Models: A Survey and Outlookcited by 11
  6. 2024Securing the Future of GenAI: Policy and Technologycited by 4
  7. 2025Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Modelscited by 4
  8. 2025EMMA: Concept Erasure Benchmark with Comprehensive Semantic Metrics and Diverse Categoriescited by 4
  9. 2025Model-Agnostic Gender Bias Control for Text-to-Image Generation via Sparse Autoencodercited by 3
  10. 2025When Algorithms Play Favorites: Lookism in the Generation and Perception of Facescited by 2
  11. 2025The Iconicity of the Generated Imagecited by 2
  12. 2026Aesthetics as Structural Harm: Algorithmic Lookism Across Text-to-Image Generation and Classificationcited by 2
13 more
  1. 2026Stay Fair! Ensuring Group Fairness in Diffusion Models Across Guidance Scalescited by 2
  2. 2024"My Kind of Woman": Analysing Gender Stereotypes in AI through The Averageness Theory and EU Lawcited by 1
  3. 2024SiCo: An Interactive Size-Controllable Virtual Try-On Approach for Informed Decision-Makingcited by 1
  4. 2025Using LLMs as prompt modifier to avoid biases in AI image generatorscited by 1
  5. 2025Synthetic History: Evaluating Visual Representations of the Past in Diffusion Modelscited by 1
  6. 2025How Bias Binds: Measuring Hidden Associations for Bias Control in Text-to-Image Compositionscited by 1
  7. 2026Who Defines Fairness? Target-Based Prompting for Demographic Representation in Generative Modelscited by 1
  8. 2026Gender Artifacts from Art History to Text-to-Image Generationcited by 1
  9. 2026Training-Free Debiasing of Diffusion Models via CLIP-Guided Denoising Optimizationcited by 1
  10. 2024Attention Shift: Steering AI Away from Unsafe Contentcited by 0
  11. 2025FairT2I: Mitigating Social Bias in Text-to-Image Generation via Large Language Model-Assisted Detection and Attribute Rebalancingcited by 0
  12. 2025When Cars Have Stereotypes: Auditing Demographic Bias in Objects from Text-to-Image Modelscited by 0
  13. 2026Efficient Fairness Auditing Across Guidance Scales in Text-to-Image Diffusion Models via Causal Abstractioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Generated Bias: Auditing Internal Bias Dynamics of Text-To-Image Generative Models
  2. 2024AI-generated faces influence gender stereotypes and racial homogenization
  3. 2025Hidden Bias in the Machine: Stereotypes in Text-to-Image Models
  4. 2026Newer Is Not Fairer: Gender Stereotyping in Text-to-Image AI Across Model Generations
  5. 2023Exploring Social Bias in Downstream Applications of Text-to-Image Foundation Models
  6. 2025Prompting Away Stereotypes? Evaluating Bias in Text-to-Image Models for Occupations
  7. 2025SCALEX: Scalable Concept and Latent Exploration for Diffusion Models
  8. 2025Automated Evaluation of Gender Bias Across 13 Large Multimodal Models