mlsanityData to 7 Oct 2026

Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selection

arXiv 2406.16846 ↗First posted 24 Jun 2024cs.LG cs.CY stat.MLcited by 5

Not on the map. Why?

Abstract (arXiv)

Machine learning models can fail on subgroups that are underrepresented during training. While techniques such as dataset balancing can improve performance on underperforming groups, they require access to training group annotations and can end up removing large portions of the dataset. In this paper, we introduce Data Debiasing with Datamodels (D3M), a debiasing approach which isolates and removes specific training examples that drive the model's failures on minority groups. Our approach enables us to efficiently train debiased classifiers while removing only a small number of examples, and does not require training group annotations or additional hyperparameter tuning.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 42 of its 51 references that are in the map. First posted on arXiv 24 Jun 2024.

Builds on

42 of 51 references.

References · 42

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  3. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,665
  4. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,346
  5. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
  6. 2019Invariant Risk Minimizationcited by 1,832
  7. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  8. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,232
  9. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 911
  10. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  11. 2020Estimating Training Data Influence by Tracing Gradient Descentcited by 565
  12. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 522
30 more
  1. 2024LESS: Selecting Influential Data for Targeted Instruction Tuningcited by 514
  2. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 379
  3. 2020The Pitfalls of Simplicity Bias in Neural Networkscited by 349
  4. 2021Multimodal datasets: misogyny, pornography, and malignant stereotypescited by 326
  5. 2023Data Selection for Language Models via Importance Resamplingcited by 322
  6. 2020An Investigation of Why Overparameterization Exacerbates Spurious Correlationscited by 310
  7. 2023TRAK: Attributing Model Behavior at Scalecited by 273
  8. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  9. 2019Hidden Stratification Causes Clinically Meaningful Failures in Machine Learning for Medical Imagingcited by 215
  10. 2022Datamodels: Predicting Predictions from Training Datacited by 186
  11. 2021Simple data balancing achieves competitive worst-group-accuracycited by 177
  12. 2022Domino: Discovering Systematic Errors with Cross-Modal Embeddingscited by 155
  13. 2022If Influence Functions are the Answer, Then What is the Question?cited by 142
  14. 2021Scaling Up Influence Functionscited by 130
  15. 2024DsDm: Model-Aware Dataset Selection with Datamodelscited by 103
  16. 2022The Fallacy of AI Functionalitycited by 88
  17. 2023Fairness And Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, And Mitigation Strategiescited by 83
  18. 2022Distilling Model Failures as Directions in Latent Spacecited by 82
  19. 2023Simple and Fast Group Robustness by Automatic Feature Reweightingcited by 68
  20. 2021Editing a classifier by rewriting its prediction rulescited by 64
  21. 2023In-context Example Selection with Influencescited by 60
  22. 2022Measuring the Effect of Training Data on Deep Learning Predictions via Randomized Experimentscited by 50
  23. 2022ModelDiff: A Framework for Comparing Learning Algorithmscited by 30
  24. 2023Don't blame Dataset Shift! Shortcut Learning due to Gradients and Cross Entropycited by 30
  25. 2023Discovering environments with XRMcited by 26
  26. 2022On the Limitations of Dataset Balancing: The Lost Battle Against Spurious Correlationscited by 19
  27. 2023When More is Less: Incorporating Additional Datasets Can Hurt Performance By Introducing Spurious Correlationscited by 18
  28. 2022FairIF: Boosting Fairness in Deep Learning via Influence Functions with Validation Set Sensitive Attributescited by 12
  29. 2023Unifying Corroborative and Contributive Attributions in Large Language Modelscited by 10
  30. 2023Quantifying and mitigating the impact of label errors on model disparity metricscited by 7

Built on by

Cited by year

Citing papers per year: 2024: 2; 2025: 3.

5 papers.

  1. 2024ScaleBiO: Scalable Bilevel Optimization for LLM Data Reweightingcited by 35
  2. 2025Diversity-driven Data Selection for Language Model Tuning through Sparse Autoencodercited by 11
  3. 2024Towards User-Focused Research in Training Data Attribution for Human-Centered Explainable AIcited by 3
  4. 2025Efficiently Verifiable Proofs of Data Attributioncited by 1
  5. 2025From Ethical Declarations to Provable Independence: An Ontology-Driven Optimal-Transport Framework for Certifiably Fair AI Systemscited by 0