mlsanityData to 7 Oct 2026

Discovering environments with XRM

arXiv 2309.16748 ↗First posted 28 Sep 2023cs.LG cs.AI stat.MLcited by 26

Not on the map. Why?

Abstract (arXiv)

Environment annotations are essential for the success of many out-of-distribution (OOD) generalization methods. Unfortunately, these are costly to obtain and often limited by human annotators' biases. To achieve robust generalization, it is essential to develop algorithms for automatic environment discovery within datasets. Current proposals, which divide examples based on their training error, suffer from one fundamental problem. These methods introduce hyper-parameters and early-stopping criteria, which require a validation set with human-annotated environments, the very information subject to discovery. In this paper, we propose Cross-Risk-Minimization (XRM) to address this issue. XRM trains twin networks, each learning from one random half of the training data, while imitating confident held-out mistakes made by its sibling. XRM provides a recipe for hyper-parameter tuning, does not require early-stopping, and can discover environments for all training and validation data. Algorithms built on top of XRM environments achieve oracle worst-group-accuracy, addressing a long-standing challenge in OOD generalization. Code available at urlhttps://github.com/facebookresearch/XRM.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Builds on 56 of its 88 references that are in the map. First posted on arXiv 28 Sep 2023.

Builds on

56 of 88 references.

References · 56

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  6. 2017On Calibration of Modern Neural Networkscited by 4,860
  7. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  8. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  9. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,665
  10. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  11. 2019Invariant Risk Minimizationcited by 1,832
  12. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
38 more
  1. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,366
  2. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,232
  3. 2016Deep CORAL: Correlation Alignment for Deep Domain Adaptationcited by 1,194
  4. 2020In Search of Lost Domain Generalizationcited by 1,042
  5. 2016End to End Learning for Self-Driving Carscited by 1,000
  6. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 762
  7. 2021Domain Generalization: A Surveycited by 760
  8. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 737
  9. 2013Domain Generalization via Invariant Feature Representationcited by 622
  10. 2015Causal inference using invariant prediction: identification and confidence intervalscited by 533
  11. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 522
  12. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 492
  13. 2019Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classificationcited by 396
  14. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 379
  15. 2020The Pitfalls of Simplicity Bias in Neural Networkscited by 349
  16. 2020Noise or Signal: The Role of Image Backgrounds in Object Recognitioncited by 342
  17. 2020Environment Inference for Invariant Learningcited by 306
  18. 2020Gradient Starvation: A Learning Proclivity in Neural Networkscited by 262
  19. 2020Fairness without Demographics through Adversarially Reweighted Learningcited by 251
  20. 2020No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problemscited by 208
  21. 2022Improving Out-of-Distribution Robustness via Selective Augmentationcited by 202
  22. 2020Adaptive Risk Minimization: Learning to Adapt to Domain Shiftcited by 183
  23. 2021Simple data balancing achieves competitive worst-group-accuracycited by 177
  24. 2022Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlationscited by 171
  25. 2022Diverse Weight Averaging for Out-of-Distribution Generalizationcited by 148
  26. 2022On Feature Learning in the Presence of Spurious Correlationscited by 145
  27. 2021On Calibration and Out-of-domain Generalizationcited by 134
  28. 2020Learning from Failure: Training Debiased Classifier from Biased Classifiercited by 133
  29. 2023Change is Hard: A Closer Look at Subpopulation Shiftcited by 119
  30. 2022MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflictscited by 81
  31. 2022Agree to Disagree: Diversity through Disagreement for Better Transferabilitycited by 69
  32. 2023Simple and Fast Group Robustness by Automatic Feature Reweightingcited by 68
  33. 2022ZIN: When and How to Learn Invariance Without Environment Partition?cited by 58
  34. 2022Diversify and Disambiguate: Learning From Underspecified Datacited by 55
  35. 2023Towards Last-layer Retraining for Group Robustness with Fewer Annotationscited by 54
  36. 2022Pareto Invariant Risk Minimization: Towards Mitigating the Optimization Dilemma in Out-of-Distribution Generalizationcited by 38
  37. 2022Rich Feature Construction for the Optimization-Generalization Dilemmacited by 34
  38. 2023Spuriosity Didn't Kill the Classifier: Using Invariant Predictions to Harness Spurious Featurescited by 21

50 most cited of 56 All 56 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 10; 2025: 8; 2026: 7.

26 papers.

Sort:

By area:

  1. 2023Does Invariant Graph Learning via Environment Augmentation Learn Invariance?cited by 42
  2. 2024Unifying Causal Representation Learning with the Invariance Principlecited by 30
  3. 2024The Pitfalls of Memorization: When Memorization Hurts Generalizationcited by 16
  4. 2024Compositional Risk Minimizationcited by 13
  5. 2024Unsupervised Concept Discovery Mitigates Spurious Correlationscited by 11
  6. 2024DRoP: Distributionally Robust Data Pruningcited by 7
  7. 2025Severing Spurious Correlations with Data Pruningcited by 7
  8. 2024Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selectioncited by 5
  9. 2025Trustworthy AI: Safety, Bias, and Privacy -- A Surveycited by 5
  10. 2025AIM-Fair: Advancing Algorithmic Fairness via Selectively Fine-Tuning Biased Models with Contextual Synthetic Datacited by 5
  11. 2024Optimizing importance weighting in the presence of sub-population shiftscited by 3
  12. 2025Measure-Theoretic Anti-Causal Representation Learningcited by 3
14 more
  1. 2024FairLoRA: Unpacking Bias Mitigation in Vision Models with Fairness-Driven Low-Rank Adaptationcited by 2
  2. 2024Curriculum-enhanced GroupDRO: Challenging the Norm of Avoiding Curriculum Learning in Subpopulation Shift Setupscited by 1
  3. 2025Class-Conditional Distribution Balancing for Group Robust Classificationcited by 1
  4. 2025Fine-Grained Class-Conditional Distribution Balancing for Debiased Learningcited by 1
  5. 2025Subgroups Matter for Robust Bias Mitigationcited by 1
  6. 2026Dual-Branch Cross-Projection Debiasing through Diffusion-based Disentanglementcited by 1
  7. 2024Mind the GAP: Improving Robustness to Subpopulation Shifts with Group-Aware Priorscited by 0
  8. 2025Revisiting Mixout: An Overlooked Path to Robust Finetuningcited by 0
  9. 2026Environment-Conditioned Tail Reweighting for Invariant Learning under Mixed Shiftscited by 0
  10. 2026Bias Leaves a Gradient Trail: Label-Free Bias Identification via Gradient Probes on Concept Decompositionscited by 0
  11. 2026Discovering Latent Groups for Robust Classificationcited by 0
  12. 2026ProME: Prototype-Margin Environments with Repair-Aware Selection for Group-Robust Learningcited by 0
  13. 2026Learning the Robustness Mechanism with Bilevel Optimizationcited by 0
  14. 2026Structure-agnostic Causal Representation Learningcited by 0