mlsanityData to 7 Oct 2026

An Exploration of Default Images in Text-to-Image Generation

arXiv 2505.09166 ↗First posted 14 May 2025cs.HC cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

In the creative practice of text-to-image (TTI) generation, images are synthesized from textual prompts. By design, TTI models always yield an output, even if the prompt contains unknown terms. In this case, the model may generate default images: images that closely resemble each other across many unrelated prompts. Studying default images is valuable for designing better solutions for prompt engineering and TTI generation. We present the first investigation into default images on Midjourney. We describe an initial study in which we manually created input prompts triggering default images, and several ablation studies. Building on these, we conduct a computational analysis of over 750,000 images, revealing consistent default images across unrelated prompts. We also conduct an online user study investigating how default images may affect user satisfaction. Our work lays the foundation for understanding default images in TTI generation, highlighting their practical relevance as well as challenges and future research directions.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 40 of its 106 references that are in the map. First posted on arXiv 14 May 2025.

Builds on

40 of 106 references.

References · 40

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  6. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  7. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  8. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,388
  9. 2023ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generationcited by 1,010
  10. 2023Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generationcited by 950
  11. 2025Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scalingcited by 787
  12. 2022Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priorscited by 487
28 more
  1. 2024Evaluating Text-to-Visual Generation with Image-to-Text Generationcited by 465
  2. 2021Design Guidelines for Prompt Engineering Text-to-Image Generative Modelscited by 216
  3. 2022The Role of ImageNet Classes in Fr'echet Inception Distancecited by 211
  4. 2019Seeing What a GAN Cannot Generatecited by 148
  5. 2023ChainForge: A Visual Toolkit for Prompt Engineering and LLM Hypothesis Testingcited by 88
  6. 2023Generating images of rare concepts using pre-trained diffusion modelscited by 78
  7. 2022The Creativity of Text-to-Image Generationcited by 73
  8. 2021EncoderMI: Membership Inference against Pre-trained Encoders in Contrastive Learningcited by 68
  9. 2024PromptCharm: Text-to-Image Generation through Multi-modal Prompting and Refinementcited by 61
  10. 2022Large-scale Text-to-Image Generation Models for Visual Artists' Creative Workscited by 60
  11. 2022Discovering the Hidden Vocabulary of DALLE-2cited by 42
  12. 2023Prompting AI Art: An Investigation into the Creative Skill of Prompt Engineeringcited by 40
  13. 2024ViSAGe: A Global-Scale Analysis of Visual Stereotypes in Text-to-Image Generationcited by 27
  14. 2024Adversarial Nibbler: An Open Red-Teaming Method for Identifying Diverse Harms in Text-to-Image Generationcited by 20
  15. 2020Soliciting Human-in-the-Loop User Feedback for Interactive Machine Learning Reduces User Trust and Impressions of Model Accuracycited by 19
  16. 2023PromptAid: Prompt Exploration, Perturbation, Testing and Iteration using Visual Analytics for Large Language Modelscited by 19
  17. 2023GenAssist: Making Image Generation Accessiblecited by 18
  18. 2023Quantifying Bias in Text-to-Image Generative Modelscited by 16
  19. 2024Prototyping with Prompts: Emerging Approaches and Challenges in Generative AI Design for Collaborative Software Teamscited by 16
  20. 2024Model Collapse in the Self-Consuming Chain of Diffusion Finetuning: A Novel Perspective from Quantitative Trait Modelingcited by 11
  21. 2022Schr"odinger's Bat: Diffusion Models Sometimes Generate Polysemous Words in Superpositioncited by 8
  22. 2023Exploring Social Bias in Downstream Applications of Text-to-Image Foundation Modelscited by 7
  23. 2021Black-Box Diagnosis and Calibration on GAN Intra-Mode Collapse: A Pilot Studycited by 3
  24. 2023The Cultivated Practices of Text-to-Image Generationcited by 3
  25. 2025MIRAGE: Multi-model Interface for Reviewing and Auditing Generative Text-to-Image AIcited by 3
  26. 2025Imagining the Far East: Exploring Perceived Biases in AI-Generated Images of East Asian Womencited by 3
  27. 2024DrawL: Understanding the Effects of Non-Mainstream Dialects in Prompted Image Generationcited by 2
  28. 2022Connective Reconstruction-based Novelty Detectioncited by 1

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025EMoE: Training-Free Expert Disagreement for Uncertainty-Aware Text-to-Image Diffusioncited by 0