mlsanityData to 14 Oct 2026

No Classification without Representation: Assessing Geodiversity Issues in Open Data Sets for the Developing World

arXiv 1711.08536 ↗First posted 22 Nov 2017stat.MLcited by 153

Not on the map. Why?

Abstract (arXiv)

Modern machine learning systems such as image classifiers rely heavily on large scale data sets for training. Such data sets are costly to create, thus in practice a small number of freely available, open source data sets are widely used. We suggest that examining the geo-diversity of open data sets is critical before adopting a data set for use cases in the developing world. We analyze two large, publicly available image data sets to assess geo-diversity and find that these data sets appear to exhibit an observable amerocentric and eurocentric representation bias. Further, we analyze classifiers trained on these data sets to assess the impact of these training distributions and find strong differences in the relative performance on images from different locales. These results emphasize the need to ensure geo-representation when constructing data sets for use in the developing world.

Key facts. Cited by 153 papers in the map (data through 14 Oct 2026). First posted on arXiv 22 Nov 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 27; 2022: 29; 2023: 31; 2024: 29; 2025: 25; 2026: 12.

153 papers.

Sort:

By area:

  1. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,295
  2. 2021AI and the Everything in the Whole Wide World Benchmarkcited by 349
  3. 2021Visually Grounded Reasoning across Languages and Culturescited by 189
  4. 2022Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localizationcited by 150
  5. 2021SituatedQA: Incorporating Extra-Linguistic Contexts into QAcited by 145
  6. 2021Re-imagining Algorithmic Fairness in India and Beyondcited by 104
  7. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  8. 2021Reduced, Reused and Recycled: The Life of a Dataset in Machine Learning Researchcited by 102
  9. 2023Art and the science of generative AI: A deeper divecited by 95
  10. 2023Evaluating the Social Impact of Generative AI Systems in Systems and Societycited by 92
  11. 2021Do Datasets Have Politics? Disciplinary Values in Computer Vision Dataset Developmentcited by 85
  12. 2021A Survey on Bias in Visual Datasetscited by 84
38 more
  1. 2021A Study of Face Obfuscation in ImageNetcited by 83
  2. 2023The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AIcited by 73
  3. 2025Meta CLIP 2: A Worldwide Scaling Recipecited by 64
  4. 2021Broaden the Vision: Geo-Diverse Visual Commonsense Reasoningcited by 59
  5. 2023FACET: Fairness in Computer Vision Evaluation Benchmarkcited by 58
  6. 2021PASS: An ImageNet replacement for self-supervised pretraining without humanscited by 51
  7. 2023GeoDE: a Geographically Diverse Evaluation Dataset for Object Recognitioncited by 50
  8. 2024A Decade's Battle on Dataset Bias: Are We There Yet?cited by 47
  9. 2021A Step Toward More Inclusive People Annotations for Fairnesscited by 46
  10. 2024Consent in Crisis: The Rapid Decline of the AI Data Commonscited by 46
  11. 2023Inspecting the Geographical Representativeness of Images from Text-to-Image Modelscited by 44
  12. 2022Gender and Racial Bias in Visual Question Answering Datasetscited by 39
  13. 2023T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generationcited by 39
  14. 2022Gender Artifacts in Visual Datasetscited by 38
  15. 2023Overwriting Pretrained Bias with Finetuning Datacited by 32
  16. 2021Representation Matters: Assessing the Importance of Subgroup Allocations in Training Datacited by 31
  17. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  18. 2024Position: Measure Dataset Diversity, Don't Just Claim Itcited by 26
  19. 2022Fairness Indicators for Systematic Assessments of Visual Feature Extractorscited by 24
  20. 2022Bugs in the Data: How ImageNet Misrepresents Biodiversitycited by 24
  21. 2021UDIS: Unsupervised Discovery of Bias in Deep Visual Recognition Modelscited by 23
  22. 2021Responsible AI: Gender bias assessment in emotion recognitioncited by 22
  23. 2024Listen and Speak Fairly: A Study on Semantic Gender Bias in Speech Integrated Large Language Modelscited by 21
  24. 2023Better, Not Just More: Data-Centric Machine Learning for Earth Observationcited by 19
  25. 2021A Framework for Deprecating Datasets: Standardizing Documentation, Identification, and Communicationcited by 18
  26. 2023Ethical Considerations for Responsible Data Curationcited by 17
  27. 2023GeoNet: Benchmarking Unsupervised Adaptation across Geographiescited by 17
  28. 2024See It from My Perspective: How Language Affects Cultural Bias in Image Understandingcited by 17
  29. 2024Towards Geographic Inclusion in the Evaluation of Text-to-Image Modelscited by 16
  30. 2024No Filter: Cultural and Socioeconomic Diversity in Contrastive Vision-Language Modelscited by 16
  31. 2022Data Representativity for Machine Learning and AI Systemscited by 15
  32. 2023Out of Context: Investigating the Bias and Fairness Concerns of "Artificial Intelligence as a Service"cited by 15
  33. 2024Understanding Bias in Large-Scale Visual Datasetscited by 15
  34. 2024ONEBench to Test Them All: Sample-Level Benchmarking Over Open-Ended Capabilitiescited by 15
  35. 2025Palm: A Culturally Inclusive and Linguistically Diverse Dataset for Arabic LLMscited by 15
  36. 2021Metadata Normalizationcited by 14
  37. 2024Multilingual Diversity Improves Vision-Language Representationscited by 14
  38. 2024From Local Concepts to Universals: Evaluating the Multicultural Understanding of Vision-Language Modelscited by 14

50 most cited of 153 All 153 (JSON)