mlsanityData to 7 Oct 2026

Spread Spurious Attribute: Improving Worst-group Accuracy with Spurious Attribute Estimation

arXiv 2204.02070 ↗First posted 5 Apr 2022cs.LG cs.CVcited by 88

Not on the map. Why?

Abstract (arXiv)

The paradigm of worst-group loss minimization has shown its promise in avoiding to learn spurious correlations, but requires costly additional supervision on spurious attributes. To resolve this, recent works focus on developing weaker forms of supervision -- e.g., hyperparameters discovered with a small number of validation samples with spurious attribute annotation -- but none of the methods retain comparable performance to methods using full supervision on the spurious attribute. In this paper, instead of searching for weaker supervisions, we ask: Given access to a fixed number of samples with spurious attribute annotations, what is the best achievable worst-group loss if we "fully exploit" them? To this end, we propose a pseudo-attribute-based algorithm, coined Spread Spurious Attribute (SSA), for improving the worst-group accuracy. In particular, we leverage samples both with and without spurious attribute annotations to train a model to predict the spurious attribute, then use the pseudo-attribute predicted by the trained model as supervision on the spurious attribute to train a new robust model having minimal worst-group loss. Our experiments on various benchmark datasets show that our algorithm consistently outperforms the baseline methods using the same number of validation samples with spurious attribute annotations. We also demonstrate that the proposed SSA can achieve comparable performances to methods using full (100%) spurious attribute supervision, by using a much smaller number of annotated samples -- from 0.6% and up to 1.5%, depending on the dataset.

Key facts. Cited by 88 papers in the map (data through 7 Oct 2026). Builds on 21 of its 34 references that are in the map. First posted on arXiv 5 Apr 2022.

Builds on

21 of 34 references.

References · 21

  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. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  4. 2020Supervised Contrastive Learningcited by 2,803
  5. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,665
  6. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  7. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  8. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
  9. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,232
  10. 2018Learning to Reweight Examples for Robust Deep Learningcited by 670
  11. 2018Recognition in Terra Incognitacited by 655
  12. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 522
9 more
  1. 2019Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classificationcited by 396
  2. 2020Environment Inference for Invariant Learningcited by 306
  3. 2020No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problemscited by 208
  4. 2020Large-Scale Methods for Distributionally Robust Optimizationcited by 179
  5. 2021CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learningcited by 174
  6. 2022Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlationscited by 171
  7. 2015Cost Sensitive Learning of Deep Feature Representations from Imbalanced Datacited by 167
  8. 2020Learning from Failure: Training Debiased Classifier from Biased Classifiercited by 133
  9. 2020Distribution Aligning Refinery of Pseudo-label for Imbalanced Semi-supervised Learningcited by 105

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 14; 2023: 21; 2024: 26; 2025: 21; 2026: 6.

88 papers.

Sort:

By area:

  1. 2024Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Modelscited by 412
  2. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 379
  3. 2022On Feature Learning in the Presence of Spurious Correlationscited by 145
  4. 2023Change is Hard: A Closer Look at Subpopulation Shiftcited by 119
  5. 2023Simple and Fast Group Robustness by Automatic Feature Reweightingcited by 68
  6. 2022Contrastive Adapters for Foundation Model Group Robustnesscited by 62
  7. 2023Towards Last-layer Retraining for Group Robustness with Fewer Annotationscited by 54
  8. 2022Learning Debiased Classifier with Biased Committeecited by 51
  9. 2023Mitigating Spurious Correlations in Multi-modal Models during Fine-tuningcited by 47
  10. 2023Identifying Spurious Biases Early in Training through the Lens of Simplicity Biascited by 44
  11. 2024The Clever Hans Mirage: A Comprehensive Survey on Spurious Correlations in Machine Learningcited by 39
  12. 2025Steering Out-of-Distribution Generalization with Concept Ablation Fine-Tuningcited by 34
38 more
  1. 2024MM-SpuBench: Towards Better Understanding of Spurious Biases in Multimodal LLMscited by 33
  2. 2025Steering CLIP's vision transformer with sparse autoencoderscited by 33
  3. 2024Calibrating Multi-modal Representations: A Pursuit of Group Robustness without Annotationscited by 30
  4. 2022Metadata Archaeology: Unearthing Data Subsets by Leveraging Training Dynamicscited by 26
  5. 2023FACTS: First Amplify Correlations and Then Slice to Discover Biascited by 26
  6. 2024Navigating Shortcuts, Spurious Correlations, and Confounders: From Origins via Detection to Mitigationcited by 22
  7. 2023FLAC: Fairness-Aware Representation Learning by Suppressing Attribute-Class Associationscited by 18
  8. 2022You Only Need a Good Embeddings Extractor to Fix Spurious Correlationscited by 17
  9. 2023FRAPPE: A Group Fairness Framework for Post-Processing Everythingcited by 16
  10. 2024A Survey of Deep Graph Learning under Distribution Shifts: from Graph Out-of-Distribution Generalization to Adaptationcited by 16
  11. 2024Improving Group Robustness on Spurious Correlation Requires Preciser Group Inferencecited by 15
  12. 2023Distributionally Robust Optimization with Probabilistic Groupcited by 14
  13. 2022Spuriosity Rankings: Sorting Data to Measure and Mitigate Biasescited by 13
  14. 2023SYNAuG: Exploiting Synthetic Data for Data Imbalance Problemscited by 12
  15. 2024Unsupervised Concept Discovery Mitigates Spurious Correlationscited by 11
  16. 2022AGRO: Adversarial Discovery of Error-prone groups for Robust Optimizationcited by 10
  17. 2024Learning Robust Classifiers with Self-Guided Spurious Correlation Mitigationcited by 10
  18. 2024Navigating the Shortcut Maze: A Comprehensive Analysis of Shortcut Learning in Text Classification by Language Modelscited by 10
  19. 2023Preemptively Pruning Clever-Hans Strategies in Deep Neural Networkscited by 9
  20. 2025Project-Probe-Aggregate: Efficient Fine-Tuning for Group Robustnesscited by 9
  21. 2024ExMap: Leveraging Explainability Heatmaps for Unsupervised Group Robustness to Spurious Correlationscited by 8
  22. 2024Efficient Bias Mitigation Without Privileged Informationcited by 8
  23. 2022Beyond Invariance: Test-Time Label-Shift Adaptation for Distributions with "Spurious" Correlationscited by 7
  24. 2024Spuriousness-Aware Meta-Learning for Learning Robust Classifierscited by 7
  25. 2025Mitigating Spurious Correlation via Distributionally Robust Learning with Hierarchical Ambiguity Setscited by 7
  26. 2023Bias Amplification Enhances Minority Group Performancecited by 6
  27. 2024Improving Robustness to Multiple Spurious Correlations by Multi-Objective Optimizationcited by 6
  28. 2025Group-robust Sample Reweighting for Subpopulation Shifts via Influence Functionscited by 6
  29. 2022Learning with Noisy Labels over Imbalanced Subpopulationscited by 5
  30. 2022Breaking the Spurious Causality of Conditional Generation via Fairness Intervention with Corrective Samplingcited by 5
  31. 2024SLIM: Spuriousness Mitigation with Minimal Human Annotationscited by 5
  32. 2024The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlationscited by 5
  33. 2024Scalable Out-of-distribution Robustness in the Presence of Unobserved Confounderscited by 5
  34. 2022Improving group robustness under noisy labels using predictive uncertaintycited by 4
  35. 2023Challenges and Opportunities in Improving Worst-Group Generalization in Presence of Spurious Featurescited by 4
  36. 2025Error Slice Discovery via Manifold Compactnesscited by 4
  37. 2025Gradient Extrapolation for Debiased Representation Learningcited by 4
  38. 2025ShortcutProbe: Probing Prediction Shortcuts for Learning Robust Modelscited by 4

50 most cited of 88 All 88 (JSON)