Pith. sign in

REVIEW 4 major objections 6 minor 35 references

Spatially-Delineated Domain-Adapted AI Classification: An Application for Oncology Data

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that two self-supervised spatial tasks let a tumor-region classifier adapt to a different unlabeled region, beating published domain-adaptation baselines by up to 24 percentage points on oncology data.

desk verdict A plausible but under-verified SSL-for-point-cloud UDA paper with a real oncology application; the core idea is a fresh combination, but the evaluation needs error bars and explicit split details before the accuracy margins can be trusted. read the letter →

arxiv 2501.11695 v2 pith:3K6BTKTD submitted 2025-01-20 cs.LG cs.AI

Majid Farhadloo , Arun Sharma , Alexey Leontovich , Svetomir N. Markovic , Shashi Shekhar This is my paper · ORCID
classification cs.LGcs.AI
keywords spatially-delineateddomain-adaptedAIclassificationunsuperviseddomainadaptationself-supervisedlearningspatialarrangementsmulti-typepointmapsmultiplexedimmunofluorescenceoncologydatavariability
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a classifier trained on labeled spatial point maps from one tumor region can be adapted to classify unlabeled maps from another tumor region by adding two self-supervised tasks that explicitly model spatial arrangements. The setting is unsupervised domain adaptation for multi-type point maps, where each map records the positions and types of cells and the class is a clinical outcome such as responder versus non-responder. The authors argue that prior domain-adaptation methods align features or geometry but overlook the spatial arrangements among points, so they propose spatial mix-up masking and spatial contrastive predictive coding inside a multi-task network. On multiplexed immunofluorescence oncology data, the proposed framework reports accuracy gains over the best published baseline as large as 24 percentage points. If the gains hold, the approach would let clinical classifiers transfer between tissue regions without new annotations.

What carries the argument

The load-bearing components are two self-supervised tasks defined on spatial arrangements. Spatial mix-up masking (SMUM) creates synthetic point maps by mixing the spatial coordinates of paired source and target instances and by replacing points inside a randomly placed geometric mask in a source map with points from a target map; three auxiliary classifiers then learn to distinguish real maps from mixed ones, to classify mask-mixed maps, and to regress the mixing proportions as soft labels. Spatial contrastive predictive coding (SCPC) treats latent representations of maps from the same place-type as positive pairs and maps from different place-types as negative pairs, using an autoregressive context model to predict another instance's latent embedding, thereby pushing the encoder to represent arrangements that are consistent within a place-type and distinct across place-types. Together these tasks expose the encoder to cross-domain spatial variation without requiring target labels.

What would settle it

Re-run the four transfer tasks with subsets of increasing size up to the full point maps and with a patient-level split that keeps all subsets from one slide in the same partition; if accuracy drops toward the no-adaptation baseline in either case, the 1,024-point subsetting rather than the spatial self-supervision is carrying the result.

Watch

Extended reading notes

Core claim

The central claim is that explicitly targeting spatial arrangements during self-supervision yields a domain-adapted classifier that outperforms existing deep neural network techniques on multi-type point maps. In the proposed multi-task framework, source-labeled and target-unlabeled point maps are passed through an encoder such as SAMCNet or DGCNN, while auxiliary tasks label synthetic samples created by mixing and masking sub-regions across place-types and contrast same-place-type latent representations against different-place-type ones. The paper reports that with SAMCNet as encoder, the method improves classification accuracy over the best competitor by 4, 24, 18, and 14 percentage points on the four tasks transferring between tumor-core, interface, and normal regions, and that with DGCNN it reports gains on three of the four tasks. The sensitivity analysis attributes the improvement to the combination of both spatial self-supervised modules rather than either alone.

Load-bearing premise

The load-bearing premise is that cutting every cell map into uniform subsets of 1,024 points preserves the spatial arrangement that distinguishes the two clinical classes; if the decisive arrangement spans more than 1,024 cells, or if subsets from the same person appear in both training and test sets, the reported accuracy gains may be inflated or may not transfer to full tissue maps.

Editorial extensions

If this is right

  • Clinical classifiers for immunotherapy response can be trained on one tumor region and applied to another without re-labeling, as long as the arrangement signal is preserved in the sampled subsets.
  • The two spatial self-supervised tasks can be added to any point-cloud encoder, converting a purely supervised classifier into a domain-adapting one with a small set of auxiliary heads.
  • Transfers involving the interface region, which sits between periphery and core, should show the largest accuracy gains because its spatial arrangements lie between the other two place-types.
  • The spatial associations surfaced by the model, such as macrophage proximity to tumor-cell and CD8-cell complexes rather than the tumor-CD8 pair alone, become testable biological hypotheses about modulator cells in the tumor microenvironment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Going beyond the paper: because the same-place-type contrast objective learns to distinguish place-types in latent space, the SCPC loss could also serve as a detector for out-of-distribution tissue regions or as a graded measure of how far a region's arrangement is from the training distribution.
  • A testable extension is to vary the farthest-point-sampling subset size, for example using 512, 1,024, and 2,048 points; if accuracy improves monotonically with subset size, the arrangement signal spans large neighborhoods and the 1,024-point cap may be underestimating the true signal.
  • The paper treats each multi-type point map as a permutation-invariant set, so patient-level context is left outside the model; grouping subsets by patient or by field of view during train-test splits could change the magnitude of the reported gains.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper studies unsupervised domain adaptation for multi-type point maps, where a classifier is trained on labeled source place-type instances and applied to an unlabeled target place-type, using oncology MxIF data. It introduces a multi-task framework with two spatial self-supervised modules: spatial mix-up masking (SMUM), which creates mixed or masked samples and predicts their source/target composition, and spatial contrastive predictive coding (SCPC), which contrasts same-place-type versus different-place-type latent representations. Experiments compare the framework with DANN, DefRec+PCM, and GAST over four transfer tasks using DGCNN and SAMCNet encoders, and report accuracy improvements for the proposed method. A case study interprets top spatial relationships in tumor regions.

Significance. If the evaluation is sound, the paper is a useful extension of point-cloud domain adaptation to spatially arranged multi-type point maps, with a clear application to MxIF oncology data and a clinically motivated case study. The proposed auxiliary SSL losses do not appear circular: they are defined on source/target composition and place-type identity rather than on target class labels. The main weaknesses are evidential, not conceptual: the data-partitioning description is ambiguous, the reported margins lack statistical support, and one set of claimed margins is inconsistent with the table. The paper's strengths include comparison with three published baselines and the authors' statement that code will be available, which should enable verification once the split and statistics are clarified.

major comments (4)
  1. [Section 5.1 (Dataset Preparation)] The text states that 'we divided the data into 80% training and 20% testing' and then, 'due to the limited number of learning samples, we used the farthest point sampling strategy to break each instance into uniform subsets of 1,024 points'. This leaves unclear whether the train/test split was performed on original point maps before subsetting and whether all subsets generated from one instance were kept in the same partition. If subsets from a single slide can appear in both training and testing, the model can memorize slide-level artifacts and the reported gains in Tables 1 and 2 would be inflated. Please state explicitly that the split is at the instance level, that all FPS subsets of an instance stay in the same partition, and report the numbers of instances and subsets in each split.
  2. [Tables 1–3 and Section 5.2] All reported accuracies are point estimates without error bars, confidence intervals, multiple training seeds, or significance tests. Because the target test sets consist of roughly 17–29 original point maps per task, a difference of 4–18 percentage points can easily arise from sampling noise; the 3–10 point differences in Table 1 are especially fragile. Please report bootstrap or other confidence intervals, run each configuration with several seeds, and evaluate at the instance level rather than the FPS-subset level, with a paired significance test where appropriate.
  3. [Section 5.2 and Table 1] The claim that the DGCNN-based proposed method improved accuracy over the best competitor by 6.0%, 7.0%, and 8.0% in PT1ToPT2, PT2ToPT1, and PT3ToPT2 is not consistent with Table 1: the corresponding margins are 0.54−0.51 = 0.03, 0.67−0.57 = 0.10, and 0.61−0.53 = 0.08. Please correct the text or the table and verify all computed margins.
  4. [Section 4.2 (Spatial Contrastive Predictive Coding)] Equation (4.6) is not fully specified: the paper does not define how the context vector c_t is computed, how the predicted latent z_hat_j is generated, or how positive and negative pairs are formed beyond 'same place-type'. Since SCPC is a central proposed component and Table 3 attributes gains to it, please provide the exact pairing and autoregressive procedure, including handling of varying instance sizes and batch construction.
minor comments (6)
  1. [Section 5.1] The sentence 'experiment code is available here' does not include a URL; please provide the repository link.
  2. [Table 3] The SMUM row reports identical values for PT1ToPT2 and PT2ToPT1; please verify whether this is a copy-paste error.
  3. [Sections 4.1–4.2 and 5.1] The values of hyperparameters (alpha, tau, masking geometry and proportion, loss weights, and subset size) are not reported; adding a table with the final settings would improve reproducibility.
  4. [Section 5.1] The phrase 'weighted average of accuracy, precision, recall, and F1-score' should define how accuracy is aggregated, since accuracy is normally not a per-class score.
  5. [Section 6] The case-study claims would be stronger with quantitative validation of the top spatial relationships, such as stability across runs or comparison to a random baseline with significance testing.
  6. [Tables 1–3 and Equation (4.5)] Minor typographical issues: 'Accuray' in the table headers should be 'Accuracy', and the text contains 'P_TmaskM isk' instead of 'P_TmaskMix'.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the SSL auxiliary losses are defined on place-type identity and mix proportions, not on target labels, and the reported gains are empirical comparisons against baselines sharing the same encoder.

full rationale

The paper's central claim is an empirical accuracy comparison (Section 5.2) between a multi-task framework with spatial mix-up masking and spatial contrastive predictive coding and published baselines (DANN, DefRec+PCM, GAST) using the same backbone encoders (DGCNN and SAMCNet). The SSL objectives in Eqs. (4.3)-(4.6) are constructed from source/target place-type membership and from mask/mix proportions, not from the target classification labels, so the target accuracy is not a refit of the target labels. The proposed losses do not by construction determine the reported margins; the margins are measured on held-out target data. Self-citations to SAMCNet [25] and related prior work are used as architectural building blocks and as related work, not as an external 'uniqueness theorem' or as a derivation that forces the results. The internal inconsistency between the 6.0% margin claimed in the text and the 3-point margin shown in Table 1 for PT1ToPT2, and the possible train/test overlap of farthest-point-sampling subsets, are experimental-validity concerns rather than circular derivation. No step in the paper's claimed derivation chain reduces by definition to its inputs.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The method introduces several hyperparameters and relies on assumptions about the preservation of spatial signal and the informativeness of place-type identity. No new physical entities are introduced.

free parameters (5)
  • alpha (beta distribution shape for spatial mix-up)
    Controls the mix ratio in Eq. 4.1; tuned on validation set, exact value not reported.
  • tau (temperature in SCPC contrastive loss)
    Scales similarity in Eq. 4.6; tuned on validation set, exact value not reported.
  • Masking geometry and proportion
    Randomized geometry, center, and proportion in Section 4.1.2; ranges not specified.
  • Multi-task loss weights
    Weights combining L_clsmix, L_clsmaskMix, L_clssoftMix, and L_SCPC are not reported.
  • Number of points per subset (1,024) and farthest point sampling parameters
    Chosen due to limited sample sizes; no sensitivity analysis is provided.
assumptions (5)
  • domain assumption Label spaces of source and target place-types are identical (Y_S = Y_T)
    Section 3.2 explicitly assumes this for the problem formulation.
  • domain assumption Spatial arrangements are discriminative for class labels
    Core premise of the problem definition, stated in the introduction and used throughout.
  • domain assumption Place-type identity is a useful self-supervision signal (same place-type = positive pair)
    Eq. 4.6 defines positives this way; no evidence that this aligns with class boundaries.
  • ad hoc to paper Subsampling a point map to 1,024 points preserves class-relevant spatial patterns
    Section 5.1 applies farthest point sampling; the paper provides no verification that outcomes survive this reduction.
  • standard math Standard optimization assumptions (e.g., SGD convergence, representational capacity of DGCNN/SAMCNet)
    Implicit in all deep learning experiments.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Spatially-Delineated Domain-Adapted AI Classification: An Application for Oncology Data." pith.science (2026). https://pith.science/paper/3K6BTKTD

@misc{pith2026250111695,
  author       = {Pith},
  title        = {Pith review of: Spatially-Delineated Domain-Adapted AI Classification: An Application for Oncology Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3K6BTKTD}},
  note         = {Machine review of arXiv:2501.11695}
}
read the original abstract

Given multi-type point maps from different place-types (e.g., tumor regions), our objective is to develop a classifier trained on the source place-type to accurately distinguish between two classes of the target place-type based on their point arrangements. This problem is societally important for many applications, such as generating clinical hypotheses for designing new immunotherapies for cancer treatment. The challenge lies in the spatial variability, the inherent heterogeneity and variation observed in spatial properties or arrangements across different locations (i.e., place-types). Previous techniques focus on self-supervised tasks to learn domain-invariant features and mitigate domain differences; however, they often neglect the underlying spatial arrangements among data points, leading to significant discrepancies across different place-types. We explore a novel multi-task self-learning framework that targets spatial arrangements, such as spatial mix-up masking and spatial contrastive predictive coding, for spatially-delineated domain-adapted AI classification. Experimental results on real-world datasets (e.g., oncology data) show that the proposed framework provides higher prediction accuracy than baseline methods.

Figures

Figures reproduced from arXiv: 2501.11695 by the authors.

Figure 1
Figure 1. Spatial differences in arrangements between [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Spatial variability in a tissue slide, empha [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Three multi-type point maps are categorized [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: The overall framework of the proposed work. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 28 canonical work pages

  1. [1]

    Causes and consequences of spatial heterogeneity in ecosystem function

    Monica G Turner and F Stuart Chapin III. Causes and consequences of spatial heterogeneity in ecosystem function. In Ecosystem function in heterogeneous landscapes, pages 9–30. Springer, 2005

    work page 2005

  2. [2]

    Fastmapping: Software to create field maps and identify management zones in precision agriculture

    PPaccioretti, MarianoCórdoba, andMónicaBalzarini. Fastmapping: Software to create field maps and identify management zones in precision agriculture. Computers and Electronics in Agriculture, 175:105556, 2020

    work page 2020

  3. [3]

    Management of rhesus alloim- munization in pregnancy

    Kenneth J Moise Jr. Management of rhesus alloim- munization in pregnancy. Obstetrics & Gynecology, 112(1):164–176, 2008

    work page 2008

  4. [4]

    Preparing medical imaging data for machine learning.Radiology, 295(1):4–15, 2020

    Martin J Willemink, Wojciech A Koszek, Cailin Hard- ell, Jie Wu, Dominik Fleischmann, Hugh Harvey, Les R Folio, Ronald M Summers, Daniel L Rubin, and Matthew P Lungren. Preparing medical imaging data for machine learning.Radiology, 295(1):4–15, 2020

    work page 2020

  5. [5]

    Domain- adversarial training of neural networks.Journal of ma- chine learning research, 17(59):1–35, 2016

    Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, François Lavio- lette, Mario March, and Victor Lempitsky. Domain- adversarial training of neural networks.Journal of ma- chine learning research, 17(59):1–35, 2016

    work page 2016

  6. [6]

    Jay Kuo, and Yun Fu

    Can Qin, Haoxuan You, Lichen Wang, C.-C. Jay Kuo, and Yun Fu. Pointdan: A multi-scale 3d domain adap- tion network for point cloud representation. In H. Wal- lach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox, and R. Garnett, editors,Advances in Neural Information Processing Systems 32, pages 7190–7201. Curran Associates, Inc., 2019

    work page 2019

  7. [7]

    Adversarial training on point clouds for sim-to-real 3d object detection.IEEE Robotics and Automation Letters, 6(4):6662–6669, 2021

    Robert DeBortoli, Li Fuxin, Ashish Kapoor, and Geof- frey A Hollinger. Adversarial training on point clouds for sim-to-real 3d object detection.IEEE Robotics and Automation Letters, 6(4):6662–6669, 2021

    work page 2021

  8. [8]

    Maximum classifier discrepancy for unsupervised domain adaptation

    Kuniaki Saito, Kohei Watanabe, Yoshitaka Ushiku, and Tatsuya Harada. Maximum classifier discrepancy for unsupervised domain adaptation. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 3723–3732, 2018

    2018

Show all 35 references
  1. [9]

    Ad- versarial domain adaptation with domain mixup

    Minghao Xu, Jian Zhang, Bingbing Ni, Teng Li, Chengjie Wang, Qi Tian, and Wenjun Zhang. Ad- versarial domain adaptation with domain mixup. In Proceedings of the AAAI conference on artificial intel- ligence, volume 34, pages 6502–6509, 2020

    2020

  2. [10]

    Self- supervised learning for domain adaptation on point clouds

    Idan Achituve, Haggai Maron, and Gal Chechik. Self- supervised learning for domain adaptation on point clouds. In Proceedings of the IEEE/CVF winter con- ference on applications of computer vision, pages 123– 133, 2021

    2021

  3. [11]

    Domain adaptation on point clouds via geometry-aware implicits

    Yuefan Shen, Yanchao Yang, Mi Yan, He Wang, Youyi Zheng, and Leonidas J Guibas. Domain adaptation on point clouds via geometry-aware implicits. InProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7223–7232, 2022

    2022

  4. [12]

    Geometry-aware self-training for unsupervised domain adaptation on object point clouds

    Longkun Zou, Hui Tang, Ke Chen, and Kui Jia. Geometry-aware self-training for unsupervised domain adaptation on object point clouds. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6403–6412, 2021

    2021

  5. [13]

    A learnable self-supervised task for unsupervised domain adaptation on point cloud classification and segmentation.Frontiers of Computer Science, 17(6):176708, 2023

    Shaolei Liu, Xiaoyuan Luo, Kexue Fu, Manning Wang, and Zhijian Song. A learnable self-supervised task for unsupervised domain adaptation on point cloud classification and segmentation.Frontiers of Computer Science, 17(6):176708, 2023

    2023

  6. [14]

    Graph convolutional networks: a com- prehensive review

    Si Zhang, Hanghang Tong, Jiejun Xu, and Ross Ma- ciejewski. Graph convolutional networks: a com- prehensive review. Computational Social Networks, 6(1):1–23, 2019

    2019

  7. [15]

    Multiplexed imaging in oncology

    Chrysafis Andreou, Ralph Weissleder, and Moritz F Kircher. Multiplexed imaging in oncology. Nature Biomedical Engineering, 6(5):527–540, 2022

    2022

  8. [16]

    Cscd: towards spatially resolving the heteroge- neous landscape of mxif oncology data

    Yan Li, Majid Farhadloo, Santhoshi Krishnan, Yiqun Xie, Timothy L Frankel, Shashi Shekhar, and Arvind Rao. Cscd: towards spatially resolving the heteroge- neous landscape of mxif oncology data. InProceedings of the 10th ACM SIGSPATIAL International Work- shop on Analytics for ...

    2022

  9. [17]

    Towards spatially-lucid ai classification in non-euclidean space: An application for mxif oncol- ogy data

    Majid Farhadloo, Arun Sharma, Jayant Gupta, Alexey Leontovich, Svetomir N Markovic, and Shashi Shekhar. Towards spatially-lucid ai classification in non-euclidean space: An application for mxif oncol- ogy data. In Proceedings of the 2024 SIAM Interna- tional Conference on Data...

    2024

  10. [18]

    Spatial computing opportunities in biomedical decision support: The atlas-ehr vision

    Majid Farhadloo, Arun Sharma, Shashi Shekhar, and Svetomir Markovic. Spatial computing opportunities in biomedical decision support: The atlas-ehr vision. ACM Transactions on Spatial Algorithms and Systems, 10(3):1–36, 2024

    2024

  11. [19]

    Computational pathology: Exploring the spatial dimension of tumor ecology

    S Nawaz and Y Yuan. Computational pathology: Exploring the spatial dimension of tumor ecology. Cancer letters, 380(1):296–303, 2016

    2016

  12. [20]

    Discovering spatial co-location patterns: A summary of results

    Shashi Shekhar and Yan Huang. Discovering spatial co-location patterns: A summary of results. InInter- national symposium on spatial and temporal databases, pages 236–256. Springer, 2001

    2001

  13. [21]

    An introduction to spatial data mining.UCGIS., 2020

    Jamal Golmohammadi, Yiqun Xie, Jayant Gupta, Ma- jid Farhadloo, Yan Li, Jiannan Cai, Samantha Detor, Abigail Roh, and Shashi Shekhar. An introduction to spatial data mining.UCGIS., 2020

    2020

  14. [22]

    mixup: Beyond empirical risk mini- mization

    Hongyi Zhang. mixup: Beyond empirical risk mini- mization. arXiv preprint arXiv:1710.09412, 2017

    2017 arXiv

  15. [23]

    Rep- resentation learning with contrastive predictive coding

    AaronvandenOord, YazheLi, andOriolVinyals. Rep- resentation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748, 2018

    2018 arXiv

  16. [24]

    Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E Sarma, MichaelMBronstein, andJustinMSolomon. Dynamic graph cnn for learning on point clouds.ACM Transac- Copyright © 2025 by SIAM Unauthorized reproduction of this article is prohibited tions on Graphics (tog), 38(5):1–12, 2019

    2025

  17. [25]

    Samcnet: towards a spatially explainable ai approach for classi- fying mxif oncology data

    Majid Farhadloo, Carl Molnar, Gaoxiang Luo, Yan Li, Shashi Shekhar, Rachel L Maus, Svetomir Markovic, Alexey Leontovich, and Raymond Moore. Samcnet: towards a spatially explainable ai approach for classi- fying mxif oncology data. In Proceedings of the 28th ACM SIGKDD conferen...

    2022

  18. [26]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D.Cournapeau, M.Brucher, M.Perrot, andE.Duches- nay. Scikit-learn: MachinelearninginPython. Journal of Machine Learning Resea...

    2011

  19. [27]

    Sr- net: A spatial-relationship aware point-set classifica- tion method for multiplexed pathology images

    Yan Li, Majid Farhadloo, Santhoshi Krishnan, Timo- thy L Frankel, Shashi Shekhar, and Arvind Rao. Sr- net: A spatial-relationship aware point-set classifica- tion method for multiplexed pathology images. In Proceedings of 2nd ACM SIGKDD Workshop on Deep Learning for Spatiotemp...

    2021

  20. [28]

    Generative adversarial networks

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks. Communications of the ACM, 63(11):139– 144, 2020

    2020

  21. [29]

    Conditional adversarial domain adaptation

    Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan. Conditional adversarial domain adaptation. Advances in neural information processing systems, 31, 2018

    2018

  22. [30]

    Generation for unsupervised domain adaptation: A gan-based approach for object classification with 3d point cloud data

    Junxuan Huang, Junsong Yuan, and Chunming Qiao. Generation for unsupervised domain adaptation: A gan-based approach for object classification with 3d point cloud data. In ICASSP 2022-2022 IEEE Inter- national Conference on Acoustics, Speech and Signal Processing (ICASSP), page...

    2022

  23. [31]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16000–16009, 2022

    2022

  24. [32]

    Domain generalization with mixstyle

    KaiyangZhou, YongxinYang, YuQiao, andTaoXiang. Domain generalization with mixstyle. arXiv preprint arXiv:2104.02008, 2021

    2021 arXiv

  25. [33]

    Point-bert: Pre-training 3d point cloud transformers with masked point model- ing

    Xumin Yu, Lulu Tang, Yongming Rao, Tiejun Huang, Jie Zhou, and Jiwen Lu. Point-bert: Pre-training 3d point cloud transformers with masked point model- ing. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 19313– 19322, 2022

    2022

  26. [34]

    Self-ensembling for visual domain adaptation

    Geoff French, Michal Mackiewicz, and Mark Fisher. Self-ensembling for visual domain adaptation. In In- ternational Conference on Learning Representations, 2018

    2018

  27. [35]

    Bidi- rectional learning for domain adaptation of semantic segmentation

    Yunsheng Li, Lu Yuan, and Nuno Vasconcelos. Bidi- rectional learning for domain adaptation of semantic segmentation. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition, pages 6936–6945, 2019. Copyright © 2025 by SIAM Unauthorized reproductio...

    2019

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.