# Exploiting Cultural Biases via Homoglyphs in Text-to-Image Synthesis

arXiv: https://arxiv.org/abs/2209.08891 · HTML: https://mlsanity.com/paper/2209.08891/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 36 of its 98 references that are in the map.
- First posted on arXiv 19 Sep 2022.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 30 of 36)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [Membership Inference Attacks against Machine Learning Models](https://mlsanity.com/paper/1610.05820/) (2016)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [Scaling Autoregressive Models for Content-Rich Text-to-Image Generation](https://mlsanity.com/paper/2206.10789/) (2022)
- [Improved Techniques for Training Score-Based Generative Models](https://mlsanity.com/paper/2006.09011/) (2020)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors](https://mlsanity.com/paper/2203.13131/) (2022)
- [Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers](https://mlsanity.com/paper/1801.04354/) (2018)
- [Multimodal datasets: misogyny, pornography, and malignant stereotypes](https://mlsanity.com/paper/2110.01963/) (2021)
- [Data and its (dis)contents: A survey of dataset development and use in machine learning research](https://mlsanity.com/paper/2012.05345/) (2020)
- [FairGAN: Fairness-aware Generative Adversarial Networks](https://mlsanity.com/paper/1805.11202/) (2018)
- [Poisoning and Backdooring Contrastive Learning](https://mlsanity.com/paper/2106.09667/) (2021)
- [Are Gender-Neutral Queries Really Gender-Neutral? Mitigating Gender Bias in Image Search](https://mlsanity.com/paper/2109.05433/) (2021)
- [A very preliminary analysis of DALL-E 2](https://mlsanity.com/paper/2204.13807/) (2022)
- [Bad Characters: Imperceptible NLP Attacks](https://mlsanity.com/paper/2106.09898/) (2021)
- [Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attacks](https://mlsanity.com/paper/2201.12179/) (2022)
- [Testing Relational Understanding in Text-Guided Image Generation](https://mlsanity.com/paper/2208.00005/) (2022)
- [Adversarial Attacks on Image Generation With Made-Up Words](https://mlsanity.com/paper/2208.04135/) (2022)

## Built on by (most cited first, 29 of 29)

- [Survey of Bias In Text-to-Image Generation: Definition, Evaluation, and Mitigation](https://mlsanity.com/paper/2404.01030/) (2024)
- [Rickrolling the Artist: Injecting Backdoors into Text Encoders for Text-to-Image Synthesis](https://mlsanity.com/paper/2211.02408/) (2022)
- [T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation](https://mlsanity.com/paper/2306.00905/) (2023)
- [Adversarial Attacks and Defenses on Text-to-Image Diffusion Models: A Survey](https://mlsanity.com/paper/2407.15861/) (2024)
- [Would Deep Generative Models Amplify Bias in Future Models?](https://mlsanity.com/paper/2404.03242/) (2024)
- [Navigating Cultural Chasms: Exploring and Unlocking the Cultural POV of Text-To-Image Models](https://mlsanity.com/paper/2310.01929/) (2023)
- [CulturalFrames: Assessing Cultural Expectation Alignment in Text-to-Image Models and Evaluation Metrics](https://mlsanity.com/paper/2506.08835/) (2025)
- [Backdooring Bias (B2) into Stable Diffusion Models](https://mlsanity.com/paper/2406.15213/) (2024)
- [Replication in Visual Diffusion Models: A Survey and Outlook](https://mlsanity.com/paper/2408.00001/) (2024)
- [Trustworthy Text-to-Image Diffusion Models: A Timely and Focused Survey](https://mlsanity.com/paper/2409.18214/) (2024)
- [ACE: Concept Editing in Diffusion Models without Performance Degradation](https://mlsanity.com/paper/2503.08116/) (2025)
- [CRCE: Coreference-Retention Concept Erasure in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2503.14232/) (2025)
- [Navigating Text-to-Image Generative Bias across Indic Languages](https://mlsanity.com/paper/2408.00283/) (2024)
- [BlurGuard: A Simple Approach for Robustifying Image Protection Against AI-Powered Editing](https://mlsanity.com/paper/2511.00143/) (2025)
- [Can Prompt Modifiers Control Bias? A Comparative Analysis of Text-to-Image Generative Models](https://mlsanity.com/paper/2406.05602/) (2024)
- [CollaFuse: Collaborative Diffusion Models](https://mlsanity.com/paper/2406.14429/) (2024)
- [Implicit Bias Injection Attacks against Text-to-Image Diffusion Models](https://mlsanity.com/paper/2504.01819/) (2025)
- [Hypernymy Understanding Evaluation of Text-to-Image Models via WordNet Hierarchy](https://mlsanity.com/paper/2310.09247/) (2023)
- [Discovering Universal Semantic Triggers for Text-to-Image Synthesis](https://mlsanity.com/paper/2402.07562/) (2024)
- [From Fake Perfects to Conversational Imperfects: Exploring Image-Generative AI as a Boundary Object for Participatory Design of Public Spaces](https://mlsanity.com/paper/2411.00949/) (2024)
- [Prompt and Prejudice](https://mlsanity.com/paper/2408.04671/) (2024)
- [RusCode: Russian Cultural Code Benchmark for Text-to-Image Generation](https://mlsanity.com/paper/2502.07455/) (2025)
- [Synthetic History: Evaluating Visual Representations of the Past in Diffusion Models](https://mlsanity.com/paper/2505.17064/) (2025)
- [Order Is Not Layout: Order-to-Space Bias in Image Generation](https://mlsanity.com/paper/2603.03714/) (2026)
- [Training-Free Debiasing of Diffusion Models via CLIP-Guided Denoising Optimization](https://mlsanity.com/paper/2607.00817/) (2026)
- [On the Limitations of Cross-Lingual Consistency in Multilingual Text-to-image Generation](https://mlsanity.com/paper/2608.11002/) (2026)
- [Trustworthy Large Models in Vision: A Survey](https://mlsanity.com/paper/2311.09680/) (2023)
- [CRAFT: Cultural Russian-Oriented Dataset Adaptation for Focused Text-to-Image Generation](https://mlsanity.com/paper/2505.04851/) (2025)
- [No Safe Dose: How Training Data Drives Unsafe Image Generation](https://mlsanity.com/paper/2605.28137/) (2026)

## Abstract (arXiv)

Models for text-to-image synthesis, such as DALL-E~2 and Stable Diffusion, have recently drawn a lot of interest from academia and the general public. These models are capable of producing high-quality images that depict a variety of concepts and styles when conditioned on textual descriptions. However, these models adopt cultural characteristics associated with specific Unicode scripts from their vast amount of training data, which may not be immediately apparent. We show that by simply inserting single non-Latin characters in a textual description, common models reflect cultural stereotypes and biases in their generated images. We analyze this behavior both qualitatively and quantitatively, and identify a model's text encoder as the root cause of the phenomenon. Additionally, malicious users or service providers may try to intentionally bias the image generation to create racist stereotypes by replacing Latin characters with similarly-looking characters from non-Latin scripts, so-called homoglyphs. To mitigate such unnoticed script attacks, we propose a novel homoglyph unlearning method to fine-tune a text encoder, making it robust against homoglyph manipulations.