# Hidden Bias in the Machine: Stereotypes in Text-to-Image Models

arXiv: https://arxiv.org/abs/2506.13780 · HTML: https://mlsanity.com/paper/2506.13780/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 26 references that are in the map.
- First posted on arXiv 9 Jun 2025.

## Builds on (most cited first, 16 of 16)

- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis](https://mlsanity.com/paper/2310.00426/) (2023)
- [Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation](https://mlsanity.com/paper/2403.12015/) (2024)
- [Imagen Editor and EditBench: Advancing and Evaluating Text-Guided Image Inpainting](https://mlsanity.com/paper/2212.06909/) (2022)
- [DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models](https://mlsanity.com/paper/2202.04053/) (2022)
- [Stable Bias: Analyzing Societal Representations in Diffusion Models](https://mlsanity.com/paper/2303.11408/) (2023)
- [SINE: SINgle Image Editing with Text-to-Image Diffusion Models](https://mlsanity.com/paper/2212.04489/) (2022)
- [Social Biases through the Text-to-Image Generation Lens](https://mlsanity.com/paper/2304.06034/) (2023)
- [Survey of Bias In Text-to-Image Generation: Definition, Evaluation, and Mitigation](https://mlsanity.com/paper/2404.01030/) (2024)
- [Look, Imagine and Match: Improving Textual-Visual Cross-Modal Retrieval with Generative Models](https://mlsanity.com/paper/1711.06420/) (2017)
- [TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models](https://mlsanity.com/paper/2312.01261/) (2023)
- [Multilingual Diversity Improves Vision-Language Representations](https://mlsanity.com/paper/2405.16915/) (2024)
- [Analyzing Quality, Bias, and Performance in Text-to-Image Generative Models](https://mlsanity.com/paper/2407.00138/) (2024)
- [Text-to-Image Synthesis: A Decade Survey](https://mlsanity.com/paper/2411.16164/) (2024)
- [Agonistic Image Generation: Unsettling the Hegemony of Intention](https://mlsanity.com/paper/2502.15242/) (2025)

## Built on by (most cited first, 2 of 2)

- [Toward Trustworthy Portrait Editing: Evaluation of Demographic Misrepresentation in I2I Models](https://mlsanity.com/paper/2602.16149/) (2026)
- [Beyond wheelchairs and blindfolds: Investigating disability stereotypes in T2I models with INCLUDE-BENCH](https://mlsanity.com/paper/2607.08515/) (2026)

## Abstract (arXiv)

Text-to-Image (T2I) models have transformed visual content creation, producing highly realistic images from natural language prompts. However, concerns persist around their potential to replicate and magnify existing societal biases. To investigate these issues, we curated a diverse set of prompts spanning thematic categories such as occupations, traits, actions, ideologies, emotions, family roles, place descriptions, spirituality, and life events. For each of the 160 unique topics, we crafted multiple prompt variations to reflect a wide range of meanings and perspectives. Using Stable Diffusion 1.5 (UNet-based) and Flux-1 (DiT-based) models with original checkpoints, we generated over 16,000 images under consistent settings. Additionally, we collected 8,000 comparison images from Google Image Search. All outputs were filtered to exclude abstract, distorted, or nonsensical results. Our analysis reveals significant disparities in the representation of gender, race, age, somatotype, and other human-centric factors across generated images. These disparities often mirror and reinforce harmful stereotypes embedded in societal narratives. We discuss the implications of these findings and emphasize the need for more inclusive datasets and development practices to foster fairness in generative visual systems.