mlsanityData to 13 Oct 2026

Mitigating Dataset Harms Requires Stewardship: Lessons from 1000 Papers

arXiv 2108.02922 ↗First posted 6 Aug 2021cs.LG cs.CYcited by 49

Not on the map. Why?

Abstract (arXiv)

Machine learning datasets have elicited concerns about privacy, bias, and unethical applications, leading to the retraction of prominent datasets such as DukeMTMC, MS-Celeb-1M, and Tiny Images. In response, the machine learning community has called for higher ethical standards in dataset creation. To help inform these efforts, we studied three influential but ethically problematic face and person recognition datasets -- Labeled Faces in the Wild (LFW), MS-Celeb-1M, and DukeMTM -- by analyzing nearly 1000 papers that cite them. We found that the creation of derivative datasets and models, broader technological and social change, the lack of clarity of licenses, and dataset management practices can introduce a wide range of ethical concerns. We conclude by suggesting a distributed approach to harm mitigation that considers the entire life cycle of a dataset.

Key facts. Cited by 49 papers in the map (data through 13 Oct 2026). Builds on 39 of its 104 references that are in the map. First posted on arXiv 6 Aug 2021.

Builds on

39 of 104 references.

References · 39

  1. 2014Deep Learning Face Attributes in the Wildcited by 3,686
  2. 2018Datasheets for Datasetscited by 1,377
  3. 2018Model Cards for Model Reportingcited by 1,076
  4. 2017VGGFace2: A dataset for recognising faces across pose and agecited by 913
  5. 2016Semantics derived automatically from language corpora contain human-like biasescited by 895
  6. 2016Performance Measures and a Data Set for Multi-Target, Multi-Camera Trackingcited by 687
  7. 2016MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognitioncited by 624
  8. 2018AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Biascited by 319
  9. 2017Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitrocited by 280
  10. 2018Deep Face Recognition: A Surveycited by 275
  11. 2018Deep Facial Expression Recognition: A Surveycited by 270
  12. 2020Data and its (dis)contents: A survey of dataset development and use in machine learning researchcited by 259
27 more
  1. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  2. 2015End-to-end people detection in crowded scenescited by 140
  3. 2020REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasetscited by 136
  4. 2015A Light CNN for Deep Face Representation with Noisy Labelscited by 126
  5. 2019Towards a Critical Race Methodology in Algorithmic Fairnesscited by 125
  6. 2019Lessons from Archives: Strategies for Collecting Sociocultural Data in Machine Learningcited by 116
  7. 2020Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biasescited by 107
  8. 2018The Dataset Nutrition Label: A Framework To Drive Higher Data Quality Standardscited by 97
  9. 2017Improving Person Re-identification by Attribute and Identity Learningcited by 95
  10. 2020Saving Face: Investigating the Ethical Concerns of Facial Recognition Auditingcited by 89
  11. 2021A Study of Face Obfuscation in ImageNetcited by 83
  12. 2021Towards Measuring Fairness in AI: the Casual Conversations Datasetcited by 80
  13. 2019Diversity in Facescited by 77
  14. 2017Racial Disparity in Natural Language Processing: A Case Study of Social Media African-American Englishcited by 70
  15. 2020Masked Face Recognition Dataset and Applicationcited by 66
  16. 2019Unsupervised Tracklet Person Re-Identificationcited by 55
  17. 2018Racial categories in machine learningcited by 48
  18. 2017Heterogeneous Face Attribute Estimation: A Deep Multi-Task Learning Approachcited by 43
  19. 2020Bringing the People Back In: Contesting Benchmark Machine Learning Datasetscited by 43
  20. 2014Effective Face Frontalization in Unconstrained Imagescited by 34
  21. 2021One Label, One Billion Faces: Usage and Consistency of Racial Categories in Computer Visioncited by 33
  22. 2021About Face: A Survey of Facial Recognition Evaluationcited by 27
  23. 2019Towards Standardization of Data Licenses: The Montreal Data Licensecited by 19
  24. 2020An Ethical Highlighter for People-Centric Dataset Creationcited by 12
  25. 2019Algorithmic Injustices: Towards a Relational Ethicscited by 11
  26. 2020Like a Researcher Stating Broader Impact For the Very First Timecited by 5
  27. 2018Privacy, ethics, and data access: A case study of the Fragile Families Challengecited by 2

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 6; 2022: 9; 2023: 12; 2024: 17; 2025: 4; 2026: 1.

49 papers.

Sort:

By area:

  1. 2022When and why vision-language models behave like bags-of-words, and what to do about it?cited by 606
  2. 2024Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuningcited by 174
  3. 2024Proving membership in LLM pretraining data via data watermarkscited by 40
  4. 2022Gender Artifacts in Visual Datasetscited by 38
  5. 2023Interpretable Diffusion via Information Decompositioncited by 37
  6. 2022Data Governance in the Age of Large-Scale Data-Driven Language Technologycited by 33
  7. 2022Rethinking Bias Mitigation: Fairer Architectures Make for Fairer Face Recognitioncited by 26
  8. 2024Position: Measure Dataset Diversity, Don't Just Claim Itcited by 26
  9. 2024Real Risks of Fake Data: Synthetic Data, Diversity-Washing and Consent Circumventioncited by 21
  10. 2022Labeling instructions matter in biomedical image analysiscited by 20
  11. 2021A Framework for Deprecating Datasets: Standardizing Documentation, Identification, and Communicationcited by 18
  12. 2021SynthBio: A Case Study in Human-AI Collaborative Curation of Text Datasetscited by 18
37 more
  1. 2024Wikibench: Community-Driven Data Curation for AI Evaluation on Wikipediacited by 18
  2. 2023Ethical Considerations for Responsible Data Curationcited by 17
  3. 2024Who's in and who's out? A case study of multimodal CLIP-filtering in DataCompcited by 15
  4. 2023REFORMS: Reporting Standards for Machine Learning Based Sciencecited by 13
  5. 2021Feature and Label Embedding Spaces Matter in Addressing Image Classifier Biascited by 11
  6. 2021SoK: Machine Learning Governancecited by 10
  7. 2024Benchmark Data Repositories for Better Benchmarkingcited by 10
  8. 2023Intrinsic Self-Supervision for Data Quality Auditscited by 8
  9. 2023Dissecting Multimodality in VideoQA Transformer Models by Impairing Modality Fusioncited by 8
  10. 2025In the Picture: Medical Imaging Datasets, Artifacts, and their Living Reviewcited by 8
  11. 2025Beyond Real Faces: Synthetic Datasets Can Achieve Reliable Recognition Performance without Privacy Compromisecited by 8
  12. 2021Can I use this publicly available dataset to build commercial AI software? -- A Case Study on Publicly Available Image Datasetscited by 6
  13. 2022Regulating Facial Processing Technologies: Tensions Between Legal and Technical Considerations in the Application of Illinois BIPAcited by 5
  14. 2024Copycats: the many lives of a publicly available medical imaging datasetcited by 5
  15. 2024GPT Deciphering Fedspeak: Quantifying Dissent Among Hawks and Dovescited by 5
  16. 2023Fairness and Bias in Truth Discovery Algorithms: An Experimental Analysiscited by 4
  17. 2022Are Commercial Face Detection Models as Biased as Academic Models?cited by 3
  18. 2023RAI4IoE: Responsible AI for Enabling the Internet of Energycited by 3
  19. 2023The Surveillance AI Pipelinecited by 3
  20. 2024Promoting the Responsible Development of Speech Datasets for Mental Health and Neurological Disorders Researchcited by 3
  21. 2024Let Guidelines Guide You: A Prescriptive Guideline-Centered Data Annotation Methodologycited by 3
  22. 2025Completeness of Datasets Documentation on ML/AI repositories: an Empirical Investigationcited by 3
  23. 2022SMAuC -- The Scientific Multi-Authorship Corpuscited by 2
  24. 2023On Responsible Machine Learning Datasets with Fairness, Privacy, and Regulatory Normscited by 2
  25. 2024A Taxonomy of Challenges to Curating Fair Datasetscited by 2
  26. 2024CounterFace: A Synthetic Face Dataset for Fine-Grained Counterfactual Evaluation of Face Recognition Systemscited by 2
  27. 2024A Systematic Review of NeurIPS Dataset Management Practicescited by 2
  28. 2024Beyond the Numbers: Transparency in Relation Extraction Benchmark Creation and Leaderboardscited by 2
  29. 2024LicenseGPT: A Fine-tuned Foundation Model for Publicly Available Dataset License Compliancecited by 2
  30. 2021STEREO: Scientific Text Reuse in Open Access Publicationscited by 1
  31. 2023On Comparing Fair Classifiers under Data Biascited by 1
  32. 2023An investigation of licensing of datasets for machine learning based on the GQM modelcited by 1
  33. 2023Considerations for Ethical Speech Recognition Datasetscited by 1
  34. 2025Medical Imaging AI Competitions Lack Fairnesscited by 1
  35. 2026From Reflection to Repair: A Scoping Review of Dataset Documentation Toolscited by 1
  36. 2022Evaluation for Changecited by 0
  37. 2024Mitigating Downstream Model Risks via Model Provenancecited by 0