RusCode: Russian Cultural Code Benchmark for Text-to-Image Generation
Not on the map. Why?
Abstract (arXiv)
Text-to-image generation models have gained popularity among users around the world. However, many of these models exhibit a strong bias toward English-speaking cultures, ignoring or misrepresenting the unique characteristics of other language groups, countries, and nationalities. The lack of cultural awareness can reduce the generation quality and lead to undesirable consequences such as unintentional insult, and the spread of prejudice. In contrast to the field of natural language processing, cultural awareness in computer vision has not been explored as extensively. In this paper, we strive to reduce this gap. We propose a RusCode benchmark for evaluating the quality of text-to-image generation containing elements of the Russian cultural code. To do this, we form a list of 19 categories that best represent the features of Russian visual culture. Our final dataset consists of 1250 text prompts in Russian and their translations into English. The prompts cover a wide range of topics, including complex concepts from art, popular culture, folk traditions, famous people's names, natural objects, scientific achievements, etc. We present the results of a human evaluation of the side-by-side comparison of Russian visual concepts representations using popular generative models.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 30 of its 52 references that are in the map. First posted on arXiv 11 Feb 2025.
Builds on
30 of 52 references.
References · 30
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
- 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
- 2022DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Modelscited by 230
- 2023Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Biascited by 230
- 2023Stable Bias: Analyzing Societal Representations in Diffusion Modelscited by 225
- 2022Challenges and Strategies in Cross-Cultural NLPcited by 216
- 2023Sociotechnical Safety Evaluation of Generative AI Systemscited by 154
- 2023Social Biases through the Text-to-Image Generation Lenscited by 126
- 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
- 2024Benchmarking Vision Language Models for Cultural Understandingcited by 100
- 2024CLIcK: A Benchmark Dataset of Cultural and Linguistic Intelligence in Koreancited by 65
18 more
- 2021IndoNLG: Benchmark and Resources for Evaluating Indonesian Natural Language Generationcited by 57
- 2023Typology of Risks of Generative Text-to-Image Modelscited by 55
- 2022IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languagescited by 54
- 2023AI's Regimes of Representation: A Community-centered Study of Text-to-Image Models in South Asiacited by 54
- 2023COPAL-ID: Indonesian Language Reasoning with Local Culture and Nuancescited by 35
- 2022Exploiting Cultural Biases via Homoglyphs in Text-to-Image Synthesiscited by 29
- 2024Do Generative AI Models Output Harm while Representing Non-Western Cultures: Evidence from A Community-Centered Approachcited by 25
- 2024Kandinsky 3: Text-to-Image Synthesis for Multifunctional Generative Frameworkcited by 25
- 2024The Dark Side of Dataset Scaling: Evaluating Racial Classification in Multimodal Modelscited by 23
- 2024RuBLiMP: Russian Benchmark of Linguistic Minimal Pairscited by 23
- 2024MERA: A Comprehensive LLM Evaluation in Russiancited by 22
- 2023Cultural Compass: Predicting Transfer Learning Success in Offensive Language Detection with Cultural Featurescited by 18
- 2024Bridging Cultural Nuances in Dialogue Agents through Cultural Value Surveyscited by 18
- 2022MDIA: A Benchmark for Multilingual Dialogue Generation in 46 Languagescited by 13
- 2024Heron-Bench: A Benchmark for Evaluating Vision Language Models in Japanesecited by 12
- 2024Multilingual Text Style Transfer: Datasets & Models for Indian Languagescited by 9
- 2023FusionFrames: Efficient Architectural Aspects for Text-to-Video Generation Pipelinecited by 7
- 2024YaART: Yet Another ART Rendering Technologycited by 5
Built on by
Citing papers per year: 2025: 1.
1 paper.