# Towards Real-World Test-Time Adaptation: Tri-Net Self-Training with Balanced Normalization

arXiv: https://arxiv.org/abs/2309.14949 · HTML: https://mlsanity.com/paper/2309.14949/

## Key facts

- Cited by 34 papers in the map (data through 14 Oct 2026).
- Builds on 42 of its 52 references that are in the map.
- First posted on arXiv 26 Sep 2023.

## Builds on (most cited first, 30 of 42)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [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)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [Learning Transferable Features with Deep Adaptation Networks](https://mlsanity.com/paper/1502.02791/) (2015)
- [Class-Balanced Loss Based on Effective Number of Samples](https://mlsanity.com/paper/1901.05555/) (2019)
- [Deep CORAL: Correlation Alignment for Deep Domain Adaptation](https://mlsanity.com/paper/1607.01719/) (2016)
- [CyCADA: Cycle-Consistent Adversarial Domain Adaptation](https://mlsanity.com/paper/1711.03213/) (2017)
- [Generalizing to Unseen Domains: A Survey on Domain Generalization](https://mlsanity.com/paper/2103.03097/) (2021)
- [RobustBench: a standardized adversarial robustness benchmark](https://mlsanity.com/paper/2010.09670/) (2020)
- [Deep Visual Domain Adaptation: A Survey](https://mlsanity.com/paper/1802.03601/) (2018)
- [Deep Domain Confusion: Maximizing for Domain Invariance](https://mlsanity.com/paper/1412.3474/) (2014)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning](https://mlsanity.com/paper/1908.02983/) (2019)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [Semantic Foggy Scene Understanding with Synthetic Data](https://mlsanity.com/paper/1708.07819/) (2017)
- [Contrastive Test-Time Adaptation](https://mlsanity.com/paper/2204.10377/) (2022)
- [Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shift](https://mlsanity.com/paper/2006.10963/) (2020)
- [Test-Time Training with Masked Autoencoders](https://mlsanity.com/paper/2209.07522/) (2022)
- [Central Moment Discrepancy (CMD) for Domain-Invariant Representation Learning](https://mlsanity.com/paper/1702.08811/) (2017)
- [Parameter-free Online Test-time Adaptation](https://mlsanity.com/paper/2201.05718/) (2022)
- [Robust Test-Time Adaptation in Dynamic Scenarios](https://mlsanity.com/paper/2303.13899/) (2023)
- [Understanding Self-Training for Gradual Domain Adaptation](https://mlsanity.com/paper/2002.11361/) (2020)

## Built on by (most cited first, 30 of 34)

- [Improving the Generalization of Segmentation Foundation Model under Distribution Shift via Weakly Supervised Adaptation](https://mlsanity.com/paper/2312.03502/) (2023)
- [PointSAM: Pointly-Supervised Segment Anything Model for Remote Sensing Images](https://mlsanity.com/paper/2409.13401/) (2024)
- [Persistent Test-time Adaptation in Recurring Testing Scenarios](https://mlsanity.com/paper/2311.18193/) (2023)
- [SLOT: Sample-specific Language Model Optimization at Test-time](https://mlsanity.com/paper/2505.12392/) (2025)
- [Learning to Adapt Frozen CLIP for Few-Shot Test-Time Domain Adaptation](https://mlsanity.com/paper/2506.17307/) (2025)
- [Monitoring Risks in Test-Time Adaptation](https://mlsanity.com/paper/2507.08721/) (2025)
- [Decoupled Prototype Learning for Reliable Test-Time Adaptation](https://mlsanity.com/paper/2401.08703/) (2024)
- [Temporal Test-Time Adaptation with State-Space Models](https://mlsanity.com/paper/2407.12492/) (2024)
- [Parameter-Selective Continual Test-Time Adaptation](https://mlsanity.com/paper/2407.02253/) (2024)
- [UniTTA: Unified Benchmark and Versatile Framework Towards Realistic Test-Time Adaptation](https://mlsanity.com/paper/2407.20080/) (2024)
- [Low Saturation Confidence Distribution-based Test-Time Adaptation for Cross-Domain Remote Sensing Image Classification](https://mlsanity.com/paper/2408.16265/) (2024)
- [OCRT: Boosting Foundation Models in the Open World with Object-Concept-Relation Triad](https://mlsanity.com/paper/2503.18695/) (2025)
- [CLIP-Guided Source-Free Object Detection in Aerial Images](https://mlsanity.com/paper/2401.05168/) (2024)
- [Bootstrap Segmentation Foundation Model under Distribution Shift via Object-Centric Learning](https://mlsanity.com/paper/2408.16310/) (2024)
- [On the Adversarial Risk of Test Time Adaptation: An Investigation into Realistic Test-Time Data Poisoning](https://mlsanity.com/paper/2410.04682/) (2024)
- [BoTTA: Benchmarking on-device Test Time Adaptation](https://mlsanity.com/paper/2504.10149/) (2025)
- [Adapting In-Domain Few-Shot Segmentation to New Domains without Source Domain Retraining](https://mlsanity.com/paper/2504.21414/) (2025)
- [Test-time Correlation Alignment](https://mlsanity.com/paper/2505.00533/) (2025)
- [Exploring Human-in-the-Loop Test-Time Adaptation by Synergizing Active Learning and Model Selection](https://mlsanity.com/paper/2405.18911/) (2024)
- [Generalizing Segmentation Foundation Model Under Sim-to-real Domain-shift for Guidewire Segmentation in X-ray Fluoroscopy](https://mlsanity.com/paper/2410.07460/) (2024)
- [Augmented Contrastive Clustering with Uncertainty-Aware Prototyping for Time Series Test Time Adaptation](https://mlsanity.com/paper/2501.01472/) (2025)
- [FRET: Feature Redundancy Elimination for Test Time Adaptation](https://mlsanity.com/paper/2505.10641/) (2025)
- [Probabilistic Hyper-Graphs using Multiple Randomly Masked Autoencoders for Semi-supervised Multi-modal Multi-task Learning](https://mlsanity.com/paper/2510.10068/) (2025)
- [ReSAM: Refine, Requery, and Reinforce: Self-Prompting Point-Supervised Segmentation for Remote Sensing Images](https://mlsanity.com/paper/2511.21606/) (2025)
- [Improving Out-of-Distribution Data Handling and Corruption Resistance via Modern Hopfield Networks](https://mlsanity.com/paper/2408.11309/) (2024)
- [Improving Adversarial Robustness for 3D Point Cloud Recognition at Test-Time through Purified Self-Training](https://mlsanity.com/paper/2409.14940/) (2024)
- [R.I.P.: A Simple Black-box Attack on Continual Test-time Adaptation](https://mlsanity.com/paper/2412.01154/) (2024)
- [Learn Faster and Remember More: Balancing Exploration and Exploitation for Continual Test-time Adaptation](https://mlsanity.com/paper/2508.12643/) (2025)
- [One Pool Is Not Enough: Multi-Cluster Memory for Practical Test-Time Adaptation](https://mlsanity.com/paper/2603.21135/) (2026)
- [Entropy Minimization without Model Collapse: Mitigating Prediction Bias in Medical Imaging](https://mlsanity.com/paper/2606.02339/) (2026)

## Abstract (arXiv)

Test-Time Adaptation aims to adapt source domain model to testing data at inference stage with success demonstrated in adapting to unseen corruptions. However, these attempts may fail under more challenging real-world scenarios. Existing works mainly consider real-world test-time adaptation under non-i.i.d. data stream and continual domain shift. In this work, we first complement the existing real-world TTA protocol with a globally class imbalanced testing set. We demonstrate that combining all settings together poses new challenges to existing methods. We argue the failure of state-of-the-art methods is first caused by indiscriminately adapting normalization layers to imbalanced testing data. To remedy this shortcoming, we propose a balanced batchnorm layer to swap out the regular batchnorm at inference stage. The new batchnorm layer is capable of adapting without biasing towards majority classes. We are further inspired by the success of self-training (ST) in learning from unlabeled data and adapt ST for test-time adaptation. However, ST alone is prone to over adaption which is responsible for the poor performance under continual domain shift. Hence, we propose to improve self-training under continual domain shift by regularizing model updates with an anchored loss. The final TTA model, termed as TRIBE, is built upon a tri-net architecture with balanced batchnorm layers. We evaluate TRIBE on four datasets representing real-world TTA settings. TRIBE consistently achieves the state-of-the-art performance across multiple evaluation protocols. The code is available at https://github.com/Gorilla-Lab-SCUT/TRIBE.