mlsanityData to 7 Oct 2026

MATE: Masked Autoencoders are Online 3D Test-Time Learners

arXiv 2211.11432 ↗First posted 21 Nov 2022cs.CVcited by 23

Not on the map. Why?

Abstract (arXiv)

Our MATE is the first Test-Time-Training (TTT) method designed for 3D data, which makes deep networks trained for point cloud classification robust to distribution shifts occurring in test data. Like existing TTT methods from the 2D image domain, MATE also leverages test data for adaptation. Its test-time objective is that of a Masked Autoencoder: a large portion of each test point cloud is removed before it is fed to the network, tasked with reconstructing the full point cloud. Once the network is updated, it is used to classify the point cloud. We test MATE on several 3D object classification datasets and show that it significantly improves robustness of deep networks to several types of corruptions commonly occurring in 3D point clouds. We show that MATE is very efficient in terms of the fraction of points it needs for the adaptation. It can effectively adapt given as few as 5% of tokens of each test sample, making it extremely lightweight. Our experiments show that MATE also achieves competitive performance by adapting sparsely on the test data, which further reduces its computational overhead, making it ideal for real-time applications.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 32 of its 41 references that are in the map. First posted on arXiv 21 Nov 2022.

Builds on

32 of 41 references.

References · 32

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  4. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  5. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  6. 2020Exploring Simple Siamese Representation Learningcited by 2,853
  7. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,818
  8. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  9. 2016Deep CORAL: Correlation Alignment for Deep Domain Adaptationcited by 1,194
  10. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 844
  11. 2018Open3D: A Modern Library for 3D Data Processingcited by 717
  12. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
20 more
  1. 2022Continual Test-Time Domain Adaptationcited by 520
  2. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 456
  3. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 376
  4. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  5. 2022Test-Time Training with Masked Autoencoderscited by 208
  6. 2017Central Moment Discrepancy (CMD) for Domain-Invariant Representation Learningcited by 192
  7. 2022Parameter-free Online Test-time Adaptationcited by 184
  8. 2021Spatio-temporal Self-Supervised Representation Learning for 3D Point Cloudscited by 180
  9. 2020Train in Germany, Test in The USA: Making 3D Object Detectors Generalizecited by 150
  10. 2021Revisiting Point Cloud Shape Classification with a Simple and Effective Baselinecited by 145
  11. 2021The Norm Must Go On: Dynamic Unsupervised Domain Adaptation by Normalizationcited by 131
  12. 2020Complete & Label: A Domain Adaptation Approach to Semantic Segmentation of LiDAR Point Cloudscited by 116
  13. 2019PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representationcited by 114
  14. 2022PillarNet: Real-Time and High-Performance Pillar-based 3D Object Detectioncited by 99
  15. 2022Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptionscited by 78
  16. 2022MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentationcited by 76
  17. 2022Benchmarking and Analyzing Point Cloud Classification under Corruptionscited by 70
  18. 2021Unsupervised Domain Adaptive 3D Detection with Multi-Level Consistencycited by 63
  19. 20213D-VField: Adversarial Augmentation of Point Clouds for Domain Generalization in 3D Object Detectioncited by 43
  20. 2021Domain Adaptation on Point Clouds via Geometry-Aware Implicitscited by 26

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 8; 2025: 7; 2026: 5.

23 papers.

Sort:

By area:

  1. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  2. 2024Masked Image Modeling: A Surveycited by 28
  3. 2024T4P: Test-Time Training of Trajectory Prediction via Masked Autoencoder and Actor-specific Token Memorycited by 20
  4. 2024Backpropagation-free Network for 3D Test-time Adaptationcited by 14
  5. 2024TALoS: Enhancing Semantic Scene Completion via Test-time Adaptation on the Line of Sightcited by 12
  6. 2024CloudFixer: Test-Time Adaptation for 3D Point Clouds via Diffusion-Guided Geometric Transformationcited by 11
  7. 2024Test-Time Adaptation in Point Clouds: Leveraging Sampling Variation with Weight Averagingcited by 7
  8. 2023Domain Generalization of 3D Object Detection by Density-Resamplingcited by 6
  9. 2024AdapTable: Test-Time Adaptation for Tabular Data via Shift-Aware Uncertainty Calibrator and Label Distribution Handlercited by 6
  10. 2024TTT-KD: Test-Time Training for 3D Semantic Segmentation through Knowledge Distillation from Foundation Modelscited by 5
  11. 2025Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysiscited by 4
  12. 2025SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Cloudscited by 4
11 more
  1. 2025Purge-Gate: Backpropagation-Free Test-Time Adaptation for Point Clouds Classification via Token Purgingcited by 3
  2. 2025APCoTTA: Continual Test-Time Adaptation for Semantic Segmentation of Airborne LiDAR Point Cloudscited by 2
  3. 2026Turning Adaptation into Assets: Cross-Domain Bridging for Online Vision-Language Navigationcited by 2
  4. 2023Calibrating Panoramic Depth Estimation for Practical Localization and Mappingcited by 1
  5. 20253D Test-time Adaptation via Graph Spectral Driven Point Shiftcited by 1
  6. 2025Adapt-As-You-Walk Through the Clouds: Training-Free Online Test-Time Adaptation of 3D Vision-Language Foundation Modelscited by 1
  7. 2025T3SVFND: Towards an Evolving Fake News Detector for Emergencies with Test-time Training on Short Video Platformscited by 0
  8. 2026MetaDAT: Generalizable Trajectory Prediction via Meta Pre-training and Data-Adaptive Test-Time Updatingcited by 0
  9. 2026Adapting Point Cloud Analysis via Multimodal Bayesian Distribution Learningcited by 0
  10. 2026MAMVI: 3D Test-Time Adaptation via Masked Multi-View Point Cloudscited by 0
  11. 2026The JEPA Predictor: A Transferable Operator for Occluded Feature Completioncited by 0