mlsanityData to 7 Oct 2026

Robust Perception through Equivariance

arXiv 2212.06079 ↗First posted 12 Dec 2022cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

Deep networks for computer vision are not reliable when they encounter adversarial examples. In this paper, we introduce a framework that uses the dense intrinsic constraints in natural images to robustify inference. By introducing constraints at inference time, we can shift the burden of robustness from training to the inference algorithm, thereby allowing the model to adjust dynamically to each individual image's unique and potentially novel characteristics at inference time. Among different constraints, we find that equivariance-based constraints are most effective, because they allow dense constraints in the feature space without overly constraining the representation at a fine-grained level. Our theoretical results validate the importance of having such dense constraints at inference time. Our empirical experiments show that restoring feature equivariance at inference time defends against worst-case adversarial perturbations. The method obtains improved adversarial robustness on four datasets (ImageNet, Cityscapes, PASCAL VOC, and MS-COCO) on image recognition, semantic segmentation, and instance segmentation tasks. Project page is available at equi4robust.cs.columbia.edu.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 52 of its 66 references that are in the map. First posted on arXiv 12 Dec 2022.

Builds on

52 of 66 references.

References · 52

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2017Mask R-CNNcited by 7,259
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  8. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  9. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  10. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  11. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  12. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,154
38 more
  1. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  2. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  3. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  4. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  5. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  6. 2019MixMatch: A Holistic Approach to Semi-Supervised Learningcited by 1,390
  7. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  8. 2016Group Equivariant Convolutional Networkscited by 1,154
  9. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  10. 2019Using Self-Supervised Learning Can Improve Model Robustness and Uncertaintycited by 511
  11. 2020Overfitting in adversarially robust deep learningcited by 504
  12. 2019General E(2)-Equivariant Steerable CNNscited by 441
  13. 2020CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instancescited by 441
  14. 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
  15. 2019Making Convolutional Networks Shift-Invariant Againcited by 390
  16. 2016Steerable CNNscited by 298
  17. 2017Dilated Residual Networkscited by 293
  18. 2017Parseval Networks: Improving Robustness to Adversarial Examplescited by 288
  19. 2020Domain Generalization using Causal Matchingcited by 224
  20. 2017Learning Steerable Filters for Rotation Equivariant CNNscited by 193
  21. 2017On the Robustness of Semantic Segmentation Models to Adversarial Attackscited by 126
  22. 2019Scale-Equivariant Steerable Networkscited by 110
  23. 2020Robust Learning Through Cross-Task Consistencycited by 105
  24. 2016Exploiting Cyclic Symmetry in Convolutional Neural Networkscited by 91
  25. 2019Metric Learning for Adversarial Robustnesscited by 77
  26. 2016TI-POOLING: transformation-invariant pooling for feature learning in Convolutional Neural Networkscited by 66
  27. 2022Evaluating the Adversarial Robustness of Adaptive Test-time Defensescited by 64
  28. 2021A Closer Look at Self-training for Zero-Label Semantic Segmentationcited by 57
  29. 2020Truly shift-invariant convolutional neural networkscited by 53
  30. 2017Houdini: Fooling Deep Structured Prediction Modelscited by 50
  31. 2021Improving Transferability of Representations via Augmentation-Aware Self-Supervisioncited by 48
  32. 2021Adversarial Attacks are Reversible with Natural Supervisioncited by 47
  33. 2021Online Adversarial Purification based on Self-Supervisioncited by 45
  34. 2020Certifying Confidence via Randomized Smoothingcited by 36
  35. 2016Learning rotation invariant convolutional filters for texture classificationcited by 24
  36. 2020Rotation Equivariant Siamese Networks for Trackingcited by 15
  37. 2019Bidirectional Inference Networks: A Class of Deep Bayesian Networks for Health Profilingcited by 11
  38. 2021Truly shift-equivariant convolutional neural networks with adaptive polyphase upsamplingcited by 10

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 2; 2024: 2; 2025: 1; 2026: 2.

9 papers.

  1. 2022Understanding Zero-Shot Adversarial Robustness for Large-Scale Modelscited by 117
  2. 2023Convolutional Visual Prompt for Robust Visual Perceptioncited by 17
  3. 2025Structure-aware Semantic Discrepancy and Consistency for 3D Medical Image Self-supervised Learningcited by 4
  4. 2022Adversarially Robust Video Perception by Seeing Motioncited by 2
  5. 2023Robustifying Language Models with Test-Time Adaptationcited by 2
  6. 2024G-RepsNet: A Fast and General Construction of Equivariant Networks for Arbitrary Matrix Groupscited by 1
  7. 2024Towards Adversarial Robustness And Backdoor Mitigation in SSLcited by 0
  8. 2026Beyond Instance-Level Self-Supervision in 3D Multi-Modal Medical Imagingcited by 0
  9. 2026Cross-View Correspondence Is a Measurement Intervention: Two-Sided Validation for Agent Evaluation and Credit Assignmentcited by 0