# If your data distribution shifts, use self-learning

arXiv: https://arxiv.org/abs/2104.12928 · HTML: https://mlsanity.com/paper/2104.12928/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Builds on 62 of its 105 references that are in the map.
- First posted on arXiv 27 Apr 2021.

## Builds on (most cited first, 30 of 62)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Understanding deep learning requires rethinking generalization](https://mlsanity.com/paper/1611.03530/) (2016)
- [Unsupervised Representation Learning by Predicting Image Rotations](https://mlsanity.com/paper/1803.07728/) (2018)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [Big Self-Supervised Models are Strong Semi-Supervised Learners](https://mlsanity.com/paper/2006.10029/) (2020)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Adversarial Discriminative Domain Adaptation](https://mlsanity.com/paper/1702.05464/) (2017)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Dota 2 with Large Scale Deep Reinforcement Learning](https://mlsanity.com/paper/1912.06680/) (2019)
- [CyCADA: Cycle-Consistent Adversarial Domain Adaptation](https://mlsanity.com/paper/1711.03213/) (2017)

## Built on by (most cited first, 30 of 36)

- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning](https://mlsanity.com/paper/2505.15134/) (2025)
- [Test-Time Adaptation via Conjugate Pseudo-labels](https://mlsanity.com/paper/2207.09640/) (2022)
- [A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Others](https://mlsanity.com/paper/2212.04825/) (2022)
- [On Pitfalls of Test-Time Adaptation](https://mlsanity.com/paper/2306.03536/) (2023)
- [RDumb: A simple approach that questions our progress in continual test-time adaptation](https://mlsanity.com/paper/2306.05401/) (2023)
- [SuperAnimal pretrained pose estimation models for behavioral analysis](https://mlsanity.com/paper/2203.07436/) (2022)
- [The Entropy Enigma: Success and Failure of Entropy Minimization](https://mlsanity.com/paper/2405.05012/) (2024)
- [A Survey on the Robustness of Computer Vision Models against Common Corruptions](https://mlsanity.com/paper/2305.06024/) (2023)
- [Spuriosity Didn't Kill the Classifier: Using Invariant Predictions to Harness Spurious Features](https://mlsanity.com/paper/2307.09933/) (2023)
- [Rethinking Precision of Pseudo Label: Test-Time Adaptation via Complementary Learning](https://mlsanity.com/paper/2301.06013/) (2023)
- [Effective Robustness against Natural Distribution Shifts for Models with Different Training Data](https://mlsanity.com/paper/2302.01381/) (2023)
- [Uncovering Adversarial Risks of Test-Time Adaptation](https://mlsanity.com/paper/2301.12576/) (2023)
- [Test-Time Adaptation Induces Stronger Accuracy and Agreement-on-the-Line](https://mlsanity.com/paper/2310.04941/) (2023)
- [textttBATCLIP: Bimodal Online Test-Time Adaptation for CLIP](https://mlsanity.com/paper/2412.02837/) (2024)
- [MedBN: Robust Test-Time Adaptation against Malicious Test Samples](https://mlsanity.com/paper/2403.19326/) (2024)
- [Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation](https://mlsanity.com/paper/2210.10019/) (2022)
- [MANO: Exploiting Matrix Norm for Unsupervised Accuracy Estimation Under Distribution Shifts](https://mlsanity.com/paper/2405.18979/) (2024)
- [CLIPTTA: Robust Contrastive Vision-Language Test-Time Adaptation](https://mlsanity.com/paper/2507.14312/) (2025)
- [On the Adversarial Risk of Test Time Adaptation: An Investigation into Realistic Test-Time Data Poisoning](https://mlsanity.com/paper/2410.04682/) (2024)
- [Do Machine Learning Models Learn Statistical Rules Inferred from Data?](https://mlsanity.com/paper/2303.01433/) (2023)
- [Probabilistic Test-Time Generalization by Variational Neighbor-Labeling](https://mlsanity.com/paper/2307.04033/) (2023)
- [Improving self-training under distribution shifts via anchored confidence with theoretical guarantees](https://mlsanity.com/paper/2411.00586/) (2024)
- [textttAVROBUSTBENCH: Benchmarking the Robustness of Audio-Visual Recognition Models at Test-Time](https://mlsanity.com/paper/2506.00358/) (2025)
- [Maintain Plasticity in Long-timescale Continual Test-time Adaptation](https://mlsanity.com/paper/2412.20034/) (2024)
- [Reliability-Gated Source Anchoring for Continual Test-Time Adaptation](https://mlsanity.com/paper/2605.14063/) (2026)
- [Continual Test-Time Adaptation in Computer Vision: Methods, Benchmarks, and Future Directions](https://mlsanity.com/paper/2607.08164/) (2026)
- [ESP-Zero: Unsupervised enhancement of zero-shot classification for Extremely Sparse Point cloud](https://mlsanity.com/paper/2404.19639/) (2024)
- [Realistic Evaluation of Test-Time Adaptation Algorithms: Unsupervised Hyperparameter Selection](https://mlsanity.com/paper/2407.14231/) (2024)
- [Control+Shift: Generating Controllable Distribution Shifts](https://mlsanity.com/paper/2409.07940/) (2024)

## Abstract (arXiv)

We demonstrate that self-learning techniques like entropy minimization and pseudo-labeling are simple and effective at improving performance of a deployed computer vision model under systematic domain shifts. We conduct a wide range of large-scale experiments and show consistent improvements irrespective of the model architecture, the pre-training technique or the type of distribution shift. At the same time, self-learning is simple to use in practice because it does not require knowledge or access to the original training data or scheme, is robust to hyperparameter choices, is straight-forward to implement and requires only a few adaptation epochs. This makes self-learning techniques highly attractive for any practitioner who applies machine learning algorithms in the real world. We present state-of-the-art adaptation results on CIFAR10-C (8.5% error), ImageNet-C (22.0% mCE), ImageNet-R (17.4% error) and ImageNet-A (14.8% error), theoretically study the dynamics of self-supervised adaptation methods and propose a new classification dataset (ImageNet-D) which is challenging even with adaptation.