# The Bias Amplification Paradox in Text-to-Image Generation

arXiv: https://arxiv.org/abs/2308.00755 · HTML: https://mlsanity.com/paper/2308.00755/

## Key facts

- Cited by 70 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 56 references that are in the map.
- First posted on arXiv 1 Aug 2023.

## Builds on (most cited first, 30 of 39)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus](https://mlsanity.com/paper/2104.08758/) (2021)
- [Large Language Models Struggle to Learn Long-Tail Knowledge](https://mlsanity.com/paper/2211.08411/) (2022)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints](https://mlsanity.com/paper/1707.09457/) (2017)
- [Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting](https://mlsanity.com/paper/1901.09451/) (2019)
- [Multimodal datasets: misogyny, pornography, and malignant stereotypes](https://mlsanity.com/paper/2110.01963/) (2021)
- [Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale](https://mlsanity.com/paper/2211.03759/) (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)
- [A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity](https://mlsanity.com/paper/2305.13169/) (2023)
- [Adversarial Removal of Demographic Attributes from Text Data](https://mlsanity.com/paper/1808.06640/) (2018)
- [Fair Diffusion: Instructing Text-to-Image Generation Models on Fairness](https://mlsanity.com/paper/2302.10893/) (2023)
- [Bias Out-of-the-Box: An Empirical Analysis of Intersectional Occupational Biases in Popular Generative Language Models](https://mlsanity.com/paper/2102.04130/) (2021)
- [Social Biases through the Text-to-Image Generation Lens](https://mlsanity.com/paper/2304.06034/) (2023)
- [How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?](https://mlsanity.com/paper/2210.15230/) (2022)
- [A Systematic Study of Bias Amplification](https://mlsanity.com/paper/2201.11706/) (2022)
- [Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions](https://mlsanity.com/paper/2207.14251/) (2022)
- [Directional Bias Amplification](https://mlsanity.com/paper/2102.12594/) (2021)
- [Uncurated Image-Text Datasets: Shedding Light on Demographic Bias](https://mlsanity.com/paper/2304.02828/) (2023)
- [Inspecting the Geographical Representativeness of Images from Text-to-Image Models](https://mlsanity.com/paper/2305.11080/) (2023)
- [Quantifying Social Biases Using Templates is Unreliable](https://mlsanity.com/paper/2210.04337/) (2022)

## Built on by (most cited first, 30 of 70)

- [Unified Concept Editing in Diffusion Models](https://mlsanity.com/paper/2308.14761/) (2023)
- [What's In My Big Data?](https://mlsanity.com/paper/2310.20707/) (2023)
- [Dataset Diffusion: Diffusion-based Synthetic Dataset Generation for Pixel-Level Semantic Segmentation](https://mlsanity.com/paper/2309.14303/) (2023)
- [Finetuning Text-to-Image Diffusion Models for Fairness](https://mlsanity.com/paper/2311.07604/) (2023)
- [No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance](https://mlsanity.com/paper/2404.04125/) (2024)
- [TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models](https://mlsanity.com/paper/2312.01261/) (2023)
- [T2ISafety: Benchmark for Assessing Fairness, Toxicity, and Privacy in Image Generation](https://mlsanity.com/paper/2501.12612/) (2025)
- [Securing Large Language Models: Addressing Bias, Misinformation, and Prompt Attacks](https://mlsanity.com/paper/2409.08087/) (2024)
- [Stable Diffusion Exposed: Gender Bias from Prompt to Image](https://mlsanity.com/paper/2312.03027/) (2023)
- [Would Deep Generative Models Amplify Bias in Future Models?](https://mlsanity.com/paper/2404.03242/) (2024)
- [On Linear Representations and Pretraining Data Frequency in Language Models](https://mlsanity.com/paper/2504.12459/) (2025)
- [Rethinking Training for De-biasing Text-to-Image Generation: Unlocking the Potential of Stable Diffusion](https://mlsanity.com/paper/2408.12692/) (2024)
- [Large Language Model Reasoning Failures](https://mlsanity.com/paper/2602.06176/) (2026)
- [Eureka: Evaluating and Understanding Large Foundation Models](https://mlsanity.com/paper/2409.10566/) (2024)
- [Quantifying Bias in Text-to-Image Generative Models](https://mlsanity.com/paper/2312.13053/) (2023)
- [Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)](https://mlsanity.com/paper/2410.19314/) (2024)
- [A Large Scale Analysis of Gender Biases in Text-to-Image Generative Models](https://mlsanity.com/paper/2503.23398/) (2025)
- [Where Culture Fades: Revealing the Cultural Gap in Text-to-Image Generation](https://mlsanity.com/paper/2511.17282/) (2025)
- [Exploring Bias in over 100 Text-to-Image Generative Models](https://mlsanity.com/paper/2503.08012/) (2025)
- [FairGen: Enhancing Fairness in Text-to-Image Diffusion Models via Self-Discovering Latent Directions](https://mlsanity.com/paper/2412.18810/) (2024)
- [OASIS Uncovers: High-Quality T2I Models, Same Old Stereotypes](https://mlsanity.com/paper/2501.00962/) (2025)
- [CuRe: Cultural Gaps in the Long Tail of Text-to-Image Systems](https://mlsanity.com/paper/2506.08071/) (2025)
- [Image-Perfect Imperfections: Safety, Bias, and Authenticity in the Shadow of Text-To-Image Model Evolution](https://mlsanity.com/paper/2408.17285/) (2024)
- [GRADE: Quantifying Sample Diversity in Text-to-Image Models](https://mlsanity.com/paper/2410.22592/) (2024)
- [How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions](https://mlsanity.com/paper/2506.16679/) (2025)
- [Bias in Gender Bias Benchmarks: How Spurious Features Distort Evaluation](https://mlsanity.com/paper/2509.07596/) (2025)
- [Ethical-Lens: Curbing Malicious Usages of Open-Source Text-to-Image Models](https://mlsanity.com/paper/2404.12104/) (2024)
- [Can Prompt Modifiers Control Bias? A Comparative Analysis of Text-to-Image Generative Models](https://mlsanity.com/paper/2406.05602/) (2024)
- [Concept-as-Tree: A Controllable Synthetic Data Framework Makes Stronger Personalized VLMs](https://mlsanity.com/paper/2503.12999/) (2025)
- [Deeper Diffusion Models Amplify Bias](https://mlsanity.com/paper/2505.17560/) (2025)

## Abstract (arXiv)

Bias amplification is a phenomenon in which models exacerbate biases or stereotypes present in the training data. In this paper, we study bias amplification in the text-to-image domain using Stable Diffusion by comparing gender ratios in training vs. generated images. We find that the model appears to amplify gender-occupation biases found in the training data (LAION) considerably. However, we discover that amplification can be largely attributed to discrepancies between training captions and model prompts. For example, an inherent difference is that captions from the training data often contain explicit gender information while our prompts do not, which leads to a distribution shift and consequently inflates bias measures. Once we account for distributional differences between texts used for training and generation when evaluating amplification, we observe that amplification decreases drastically. Our findings illustrate the challenges of comparing biases in models and their training data, and highlight confounding factors that impact analyses.