Pith. sign in

REVIEW 5 major objections 6 minor 1 cited by

Federated Out-of-Distribution Generalization: A Causal Augmentation View

T0 review · 5 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read The paper claims that transplanting detected objects onto random backgrounds, entirely inside each client, breaks background-label shortcuts and achieves state-of-the-art federated out-of-distribution accuracy.

desk verdict A plausible plug-and-play augmentation for federated OOD that deserves a major-revision round, but the reported numbers and unfinished definitions need fixing before I'd trust any of the claimed gains. read the letter →

arxiv 2504.19882 v1 pith:5KQZWVRB submitted 2025-04-28 cs.CV

classification cs.CV
keywords federatedlearningout-of-distributiongeneralizationcausalaugmentationcounterfactualsamplesspuriouscorrelationsalientobjectdetectionnon-IIDdataimageclassification
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

Federated image models trained across clients often learn shortcuts: they predict labels from background scenes, so accuracy collapses when deployment backgrounds differ from training backgrounds. The paper tries to establish that a purely local, causality-inspired augmentation removes this failure mode. Each client sharpens its images, locates the class-relevant object region with a saliency detector, and pastes that object onto randomly selected backgrounds from its own data; training on the transplanted images makes background information no longer predictive of the label. If the claim holds, federated models can gain out-of-distribution robustness without any client-to-client data sharing, and the augmentation module can be added to existing federated methods as a plug-in.

What carries the argument

The load-bearing machinery is the pair of causal modules. Causal Region Localization (CRL) sharpens edges with Canny detection and extracts a binary object mask from a pretrained saliency network, so the object is $I_O = I_{\mathrm{sharpened}} \odot I_{CR}$; Causal Augmentation (CA) then transplants the object onto a random background with $I_{CA} = \alpha I_O \oplus (1-\alpha) I_B^{\mathrm{random}}$ and trains the local encoder and classifier on both original and augmented features. Because the random background is drawn from the client's own data, each background becomes associated with many labels, and the model learns that background does not determine the class. The paper also tests an alignment term that pulls augmented and original features together.

What would settle it

Run FedCAug on a variant of NICO in which the objects are small, partially occluded, or blended into the background, and compare it against a version trained with ground-truth object masks; if saliency-guided augmentation fails to match the masked version, the gain is coming from the quality of region localization rather than from the causal augmentation principle.

Watch

Extended reading notes

Core claim

FedCAug's central claim is that spurious background-label correlations in federated image classification can be broken by counterfactual augmentation computed entirely inside each client. The causal region localization module sharpens object edges with Canny detection and uses a pretrained saliency network to obtain a binary object mask; the causal augmentation module fuses the masked object with a random background and adds a classification loss on the resulting feature. The authors report consistent top-1 accuracy gains over existing methods on NICO-Animal, NICO-Vehicle, and ColorMNIST, and show that when the subject is masked out and only the background is shown, FedCAug-trained models predict labels with far lower confidence than baseline models. They also argue the module is orthogonal to knowledge-distillation methods, improving the plain averaging baseline and two distillation baselines when integrated.

Load-bearing premise

The load-bearing premise is that the saliency mask on the sharpened image marks the true class-defining object and that the rest of the image is spurious background; if the detector instead locks onto background texture, color, or dataset bias, the counterfactual samples reinforce the spurious correlation rather than breaking it.

Editorial extensions

If this is right

  • Trained models should keep their accuracy on unseen backgrounds, because no single background is predictive of a class after training on transplanted images.
  • The method can be dropped into existing federated algorithms without changing their structure, and each integrated version outperforms its original baseline.
  • With the object masked out, FedCAug models are much less confident about the background-only input, showing the shortcut is measurably weaker.
  • The whole pipeline runs locally on each client, so the robustness gain does not come with added client-to-client data exposure.
  • Causally augmented samples preserve the subject better than diffusion-generated samples, avoiding the quality gap that limits generative augmentation methods.

Reading between the lines

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

  • The paper does not test it, but the two-step recipe of saliency extraction plus background transplantation should also improve non-federated image classifiers whenever background is a nuisance variable.
  • A testable consequence of the causal claim: error analysis shows small or multiple subjects confuse attention, so gains should shrink as object size decreases or scene clutter increases.
  • Because no data leaves the client, the augmentation could be combined with knowledge distillation to fight both covariate shift and label distribution skew at once.
  • If the mechanism is really causal-region quality, swapping the saliency detector for a stronger segmentation model should further improve FedCAug by roughly the amount that false-positive background regions are removed.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 6 minor

Summary. The paper proposes FedCAug, a federated learning method for out-of-distribution (OOD) generalization. It uses a Causal Region Localization (CRL) module that sharpens images with Canny edge detection and applies a pre-trained PoolNet saliency detector to identify a binary 'causal region' mask; a Causal Augmentation (CA) module then cuts the detected object and pastes it onto random backgrounds within the client to generate counterfactual samples. Training minimizes cross-entropy on the original images and an additional cross-entropy loss on the augmented images. Experiments on NICO-Animal, NICO-Vehicle, and ColorMNIST compare against seven baselines, with ablations, orthogonality tests with existing federated methods, and qualitative visualizations. The paper claims that FedCAug reduces reliance on background-label correlations and outperforms state-of-the-art methods.

Significance. If the reported results hold, FedCAug offers a simple, privacy-preserving augmentation layer that can be plugged into existing federated algorithms without any client data sharing. The use of standard external benchmarks (PoolNet pretrained on saliency data, NICO, ColorMNIST) is a strength, as is the inclusion of ablation studies and error analysis. However, the central 'causal' premise is not independently verified, and the manuscript contains internal numerical inconsistencies as well as missing definitions of key components. The paper does not provide code, and the exact method as described is not fully reproducible. These issues need to be resolved before the claimed contributions can be assessed reliably, but the core idea is potentially useful and within the scope of a major revision.

major comments (5)
  1. [Tables II, III, and IV (Sections IV-B, IV-C, IV-D)] The paper reports inconsistent accuracy values for the same FedCAugFedAvg configuration on NICO-Vehicle. Table II lists A7: 69.91±0.5 and B7: 62.46±0.3, while Table IV lists A7: 66.91±0.5 and B7: 61.46±0.3, and Table III lists 66.91/61.46 for the same configuration. This discrepancy is load-bearing because the claimed superiority over baselines on NICO-Vehicle depends directly on which numbers are correct. The authors must reconcile the tables and re-check the corresponding conclusions, including the claims in Section IV-B about consistent improvements.
  2. [Section III-C and Section IV-A3 (Eq. (5), Table III)] The full method is incompletely specified. Table III reports results for '+ Align' (CRL + CA(CE + Align)) and Section IV-C describes an alignment mechanism, but the alignment loss is never defined in the method section or elsewhere. The objective L_total in Eq. (9) only includes L_CE and L_CA, with no Align term. Additionally, α in Eq. (5) is called a hyperparameter, but its value or search range is not reported in Section IV-A3, which lists only λ_weighted. Without defining Align and specifying α, the exact FedCAug algorithm cannot be reproduced and the ablation comparisons cannot be interpreted.
  3. [Section III-B, Eqs. (3)-(4)] There is a formal inconsistency in the definition of the causal region. Eq. (3) writes ICR as a 2×2 coordinate matrix with top-left and bottom-right corners, while Eq. (4) applies ICR in a Hadamard product with the image, which requires a full-resolution binary mask, and the text calls ICR a binary matrix of 0s and 1s. Please clarify whether ICR is a bounding box or a per-pixel binary mask, and update the notation and module description accordingly.
  4. [Section III-B with Section IV-F, Fig. 6] The central causal premise is not validated. PoolNet is a generic salient-object detector trained without access to class labels, but the paper treats its output as the 'causal representation region' and assumes the background is spurious. No quantitative evidence is provided that ICR aligns with the actual class-causal object regions on NICO-Animal or NICO-Vehicle, and the paper's own error analysis (Section IV-F, Fig. 6(c)-(d)) acknowledges failures on small targets and multi-subject images. A concrete test, such as measuring Intersection-over-Union between ICR and ground-truth object masks, or comparing against class-conditioned segmentation, is needed to support the claim that the augmentation breaks background-label correlations rather than acting as a generic cut-paste regularizer.
  5. [Table II and Section IV-B] The statistical support for the main claim is missing. With only three trials, many reported gains are within one standard deviation of the baseline; for example, NICO-Animal B7: FedCAugFedAvg 55.49±0.2 vs FPL 55.39±0.2, and NICO-Vehicle B7: 62.46±0.3 vs FPL 61.76±0.6 in Table II. The manuscript should report significance tests (e.g., paired t-tests or confidence intervals), or explicitly state that the differences are not statistically evaluated. Without this, the claim of 'superior performance compared to state-of-the-art methods' is not supported.
minor comments (6)
  1. [Section III-B, Eq. (2); Section III-C, Eq. (5)] The operator ⊕ is used for image fusion without being defined. Please specify the operation (e.g., alpha blending or weighted sum) for Eqs. (2) and (5).
  2. [Table I caption] The caption contains an ungrammatical phrase 'A7 REPRESENTS TO THE DATA'; please rephrase, and clarify how the A7/B7 splits are constructed.
  3. [Index Terms] The index term 'Casual augmentation' should be 'Causal augmentation'.
  4. [Section IV-E1, Fig. 3] The procedure for masking subjects to obtain background-only images is not described. Please state how these background inputs were generated so that the experiment is reproducible.
  5. [References] Several references use 'et al.' without full author lists, e.g., [35] 'Bao and et al.', which is inconsistent with standard journal style; please complete the author information.
  6. [Figure 4 caption] The numeric values in the prediction lists are not explained; please clarify whether they are logits, softmax probabilities, or another quantity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FedCAug's claims are evaluated against external benchmarks, and the 'causal' terminology is an assumption rather than a fitted or self-referential output.

full rationale

The paper's derivation chain is self-contained: FedCAug's components are a Canny sharpening step (Eqs. 1-2), a pretrained external saliency detector PoolNet (Eq. 3), a Hadamard extraction of the detected foreground (Eq. 4), a random-background composition (Eq. 5), and cross-entropy losses on original and augmented images (Eqs. 7-9). Nothing in this chain is fitted to the evaluation targets; the accuracy gains on NICO-Animal, NICO-Vehicle, and ColorMNIST are genuine empirical outcomes against external benchmarks, and the background-reliance claim is tested by a background-only prediction experiment and Grad-CAM visualizations. The word 'causal' is an interpretive label applied to PoolNet detections, and the paper's own error analysis (Sec. IV-F) concedes that small or multiple subjects can defeat the localization premise; that is an assumption-quality or correctness risk, not a circular reduction. The only self-citations are background citations in the introduction (refs. 1-3 and related) and they are not load-bearing for the proposed method or its evaluation. No equation is defined in terms of the quantity it purports to predict, and no fitted parameter is renamed as a prediction. Hence no circularity.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central method depends on two hyperparameters (alpha, lambda_weighted) and on the untested assumption that saliency maps equal causal regions. No new particles, forces, or entities are introduced. The claim that background replacement removes spurious correlation is an assumption about the data generation process, not a derived consequence.

free parameters (2)
  • alpha
    Eq. 5 blends object and random background with alpha; no value or tuning range is reported, yet it controls the strength of augmentation and the reported results.
  • lambda_weighted = selected from {0.1, 0.3, 0.5}
    Eq. 2 uses lambda_weighted to fuse edges; the paper states the set but not which value per dataset was used.
assumptions (4)
  • domain assumption PoolNet saliency maps identify causal (label-relevant) image regions.
    Eq. 3-4 use PoolNet output as ICR to separate object IO from background IB; no evidence is given that saliency equals causality.
  • domain assumption Inserting an object into a random background within each client removes the background-label spurious correlation.
    Eq. 5 and LCA assume random backgrounds are independent of labels and that the generated sample's label is the object's label.
  • domain assumption Canny edge sharpening improves causal region segmentation and does not corrupt labels.
    Eq. 1-2 apply sharpening before PoolNet; no validation is provided.
  • domain assumption Image labels remain valid after background replacement.
    Augmented images share the label of the original image; for NICO, backgrounds are part of the class context and pasting one class object into another background may create ambiguous samples.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Federated Out-of-Distribution Generalization: A Causal Augmentation View." pith.science (2026). https://pith.science/paper/5KQZWVRB

@misc{pith2026250419882,
  author       = {Pith},
  title        = {Pith review of: Federated Out-of-Distribution Generalization: A Causal Augmentation View},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5KQZWVRB}},
  note         = {Machine review of arXiv:2504.19882}
}
read the original abstract

Federated learning aims to collaboratively model by integrating multi-source information to obtain a model that can generalize across all client data. Existing methods often leverage knowledge distillation or data augmentation to mitigate the negative impact of data bias across clients. However, the limited performance of teacher models on out-of-distribution samples and the inherent quality gap between augmented and original data hinder their effectiveness and they typically fail to leverage the advantages of incorporating rich contextual information. To address these limitations, this paper proposes a Federated Causal Augmentation method, termed FedCAug, which employs causality-inspired data augmentation to break the spurious correlation between attributes and categories. Specifically, it designs a causal region localization module to accurately identify and decouple the background and objects in the image, providing rich contextual information for causal data augmentation. Additionally, it designs a causality-inspired data augmentation module that integrates causal features and within-client context to generate counterfactual samples. This significantly enhances data diversity, and the entire process does not require any information sharing between clients, thereby contributing to the protection of data privacy. Extensive experiments conducted on three datasets reveal that FedCAug markedly reduces the model's reliance on background to predict sample labels, achieving superior performance compared to state-of-the-art methods.

Figures

Figures reproduced from arXiv: 2504.19882 by the authors.

Figure 1
Figure 1. FedCAug can learn better causal representations by reducing back￾ground noise interference through causal decoupling and causal augmentation, thereby providing instructive knowledge for the image classification task. and data augmentation-based methods. Knowledge distillation￾based methods focus on guiding models to learn domain￾invariant features, reducing the influence of irrelevant at￾tributes in image recognitio… view at source ↗
Figure 2
Figure 2. Illustration of the framework of FedCAug, it sharpens images and locates class-relevant regions by using the Causal Region Localization (CRL) module; then it fuses the images with common sense background through the Causal Augmentation (CA) module, providing guiding information for the model to learn causal features. IV. EXPERIMENTS A. Experiment Settings 1) Datasets: Following the existing work, experiments were co… view at source ↗
Figure 3
Figure 3. The prediction confidence obtained from background images after [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Visualization of the visual Attention. (a) The [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: A comparative demonstration of causal augmented samples and [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Error analysis of FedCAug. (a) FedCAug enhances attention to causal regions and increases confidence in predictions. (b) FedCAug can leverage causal region localization and causal augmentation to correct prediction errors. (c) Smaller visual targets may prevent the mod…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Federated Deconfounding and Debiasing Learning for Out-of-Distribution Generalization

    cs.CV 2025-05 conditional novelty 5.0 of 10

    FedDDL improves federated out-of-distribution generalization by generating background-mixed counterfactual samples and aligning clients with causal prototypes, yielding an average Top-1 gain of about 4.5 percent over ...

Reference graph

Works this paper leans on

59 extracted references · 56 canonical work pages · cited by 1 Pith paper

  1. [1]

    Attentive modeling and distillation for out-of- distribution generalization of federated learning,

    Zhuang Qi and et al., “Attentive modeling and distillation for out-of- distribution generalization of federated learning,” in ICME. IEEE, 2024, pp. 1–6

  2. [2]

    Cross-training with multi-view knowledge fusion for heterogenous federated learning,

    Zhuang Qi and et al., “Cross-training with multi-view knowledge fusion for heterogenous federated learning,” arXiv preprint arXiv:2405.20046, pp. 1–12, 2024

  3. [3]

    Clustering-based curriculum construction for sample-balanced federated learning,

    Zhuang Qi and et al., “Clustering-based curriculum construction for sample-balanced federated learning,” in CICAI. Springer, 2022, pp. 155– 166

  4. [4]

    pfedes: Model heterogeneous personalized federated learning with feature extractor sharing,

    Liping Yi et al., “pfedes: Model heterogeneous personalized federated learning with feature extractor sharing,” in Proc. AAAI, 2025

  5. [5]

    Federated model heterogeneous matryoshka represen- tation learning,

    Liping Yi et al., “Federated model heterogeneous matryoshka represen- tation learning,” in Proc. NeurIPS, 2024

  6. [6]

    FedSSA: Semantic Similarity-based Aggregation for Efficient Model-Heterogeneous Personalized Federated Learning,

    Liping Yi, Han Yu, Zhuan Shi, Gang Wang, Xiaoguang Liu, Lizhen Cui, and Xiaoxiao Li, “FedSSA: Semantic Similarity-based Aggregation for Efficient Model-Heterogeneous Personalized Federated Learning,” in Proc. IJCAI, 2024

  7. [7]

    Fedgh: Heterogeneous federated learning with generalized global header,

    Liping Yi, Gang Wang, Xiaoguang Liu, Zhuan Shi, and Han Yu, “Fedgh: Heterogeneous federated learning with generalized global header,” in Proc. MM, Ottawa, ON, Canada . 2023, pp. 8686–8696, ACM

  8. [8]

    QSFL: A two-level uplink communication optimization framework for federated learning,

    Liping Yi, Gang Wang, and Xiaoguang Liu, “QSFL: A two-level uplink communication optimization framework for federated learning,” in Proc. ICML. 2022, vol. 162, pp. 25501–25513, PMLR

Show all 59 references
  1. [9]

    Dafkd: Domain-aware federated knowledge distillation,

    Haozhao Wang, Yichen Li, Wenchao Xu, Ruixuan Li, Yufeng Zhan, and Zhigang Zeng, “Dafkd: Domain-aware federated knowledge distillation,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17-24, 2023 . 2023, pp. 20412–20421, IEEE

  2. [10]

    Fedcda: Federated learning with cross-rounds divergence- aware aggregation,

    Haozhao Wang, Haoran Xu, Yichen Li, Yuan Xu, Ruixuan Li, and Tian- wei Zhang, “Fedcda: Federated learning with cross-rounds divergence- aware aggregation,” in The Twelfth International Conference on Learn- ing Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024

  3. [11]

    Feddse: Distribution-aware sub-model extraction for federated learning over resource-constrained devices,

    Haozhao Wang, Yabo Jia, Meng Zhang, Qinghao Hu, Hao Ren, Peng Sun, Yonggang Wen, and Tianwei Zhang, “Feddse: Distribution-aware sub-model extraction for federated learning over resource-constrained devices,” in Proceedings of the ACM on Web Conference 2024, WWW 2024, Singapore...

  4. [12]

    Lg-fgad: An effective federated graph anomaly detection framework,

    Jinyu Cai, Yunhe Zhang, Jicong Fan, and See-Kiong Ng, “Lg-fgad: An effective federated graph anomaly detection framework,” in Proceedings of the International Joint Conference on Artificial Intelligence , 2024

  5. [13]

    A survey on heterogeneous federated learning,

    Dashan Gao and et al., “A survey on heterogeneous federated learning,” arXiv preprint arXiv:2210.04505 , 2022

  6. [14]

    Cross-training with prototypical distillation for improving the generalization of federated learning,

    Tianhan Liu and et al., “Cross-training with prototypical distillation for improving the generalization of federated learning,” in ICME. IEEE, 2023, pp. 648–653

  7. [15]

    Cross-silo prototypical calibration for federated learning with non-iid data,

    Zhuang Qi and et al., “Cross-silo prototypical calibration for federated learning with non-iid data,” in MM, 2023, pp. 3099–3107

  8. [16]

    Federated learning: Challenges, methods, and future directions,

    Tian Li and et al., “Federated learning: Challenges, methods, and future directions,” IEEE signal processing magazine, vol. 37, no. 3, pp. 50–60, 2020

  9. [17]

    Cross-silo feature space alignment for federated learning on clients with imbalanced data,

    Zhuang Qi and et al., “Cross-silo feature space alignment for federated learning on clients with imbalanced data,” 2025

  10. [18]

    Improving global generalization and local personalization for federated learning,

    Lei Meng and et al., “Improving global generalization and local personalization for federated learning,” IEEE Transactions on Neural Networks and Learning Systems , 2024

  11. [19]

    Meta-causal feature learning for out-of- distribution generalization,

    Yuqing Wang and et al., “Meta-causal feature learning for out-of- distribution generalization,” in ECCV. Springer, 2022, pp. 530–545

  12. [20]

    Causal inference with sample balancing for out-of-distribution detection in visual classification,

    Yuqing Wang and et al., “Causal inference with sample balancing for out-of-distribution detection in visual classification,” in CICAI. Springer, 2022, pp. 572–583

  13. [21]

    Contrastive-enhanced domain generalization with federated learning,

    Xinhui Yu and et al., “Contrastive-enhanced domain generalization with federated learning,” IEEE Transactions on Artificial Intelligence , vol. 5, no. 4, pp. 1525–1532, 2023

  14. [22]

    Learning across domains and devices: Style- driven source-free domain adaptation in clustered federated learning,

    Donald Shenaj and et al., “Learning across domains and devices: Style- driven source-free domain adaptation in clustered federated learning,” in WACV, 2023, pp. 444–454

  15. [23]

    Rethinking federated learning with domain shift: A prototype view,

    Huang Wenke and et al., “Rethinking federated learning with domain shift: A prototype view,” in CVPR. IEEE, 2023, pp. 16312–16322

  16. [24]

    Disentangled federated learning for tackling attributes skew via invariant aggregation and diversity transferring,

    Zhengquan Luo and et al., “Disentangled federated learning for tackling attributes skew via invariant aggregation and diversity transferring,” arXiv preprint arXiv:2206.06818 , 2022

  17. [25]

    Federated domain generalization: A secure and robust framework for intelligent fault diagnosis,

    Chao Zhao and et al., “Federated domain generalization: A secure and robust framework for intelligent fault diagnosis,” IEEE Transactions on Industrial Informatics, 2023

  18. [26]

    Efficiently assemble normalization layers and regularization for federated domain generalization,

    Khiem Le and et al., “Efficiently assemble normalization layers and regularization for federated domain generalization,” in CVPR, 2024, pp. 6027–6036

  19. [27]

    Fedsr: A simple and effective domain generalization method for federated learning,

    A Tuan Nguyen and et al., “Fedsr: A simple and effective domain generalization method for federated learning,” NeurIPS, vol. 35, pp. 38831–38843, 2022

  20. [28]

    Dafkd: Domain-aware federated knowledge distillation,

    Haozhao Wang and et al., “Dafkd: Domain-aware federated knowledge distillation,” in CVPR, 2023, pp. 20412–20421

  21. [29]

    Stablefdg: style and attention based learning for federated domain generalization,

    Jungwuk Park and et al., “Stablefdg: style and attention based learning for federated domain generalization,” NeurIPS, vol. 36, 2024

  22. [30]

    Federated domain generalization for image recognition via cross-client style transfer,

    Junming Chen and et al., “Federated domain generalization for image recognition via cross-client style transfer,” in WACV, 2023, pp. 361–370

  23. [31]

    Federated adversarial domain hallucination for privacy-preserving domain generalization,

    Qinwei Xu and et al., “Federated adversarial domain hallucination for privacy-preserving domain generalization,” IEEE Transactions on Multimedia, vol. 26, pp. 1–14, 2023

  24. [32]

    Randaugment: Practical automated data augmentation with a reduced search space,

    Ekin D Cubuk and et al., “Randaugment: Practical automated data augmentation with a reduced search space,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, 2020, pp. 702–703

  25. [33]

    Feddg: Federated domain generalization on medical image segmentation via episodic learning in continuous frequency space,

    Quande Liu and et al., “Feddg: Federated domain generalization on medical image segmentation via episodic learning in continuous frequency space,” in CVPR, 2021, pp. 1013–1023

  26. [34]

    Collaborative optimization and aggregation for decentralized domain generalization and adaptation,

    Guile Wu and et al., “Collaborative optimization and aggregation for decentralized domain generalization and adaptation,” in ICCV, 2021, pp. 6484–6493

  27. [35]

    Canny edge detection enhancement by scale multipli- cation,

    Bao and et al., “Canny edge detection enhancement by scale multipli- cation,” TPAMI, vol. 27, no. 9, pp. 1485–1490, 2005

  28. [36]

    A simple pooling-based design for real-time salient object detection,

    Jiang-Jiang Liu and et al., “A simple pooling-based design for real-time salient object detection,” in CVPR, 2019, pp. 3917–3926

  29. [37]

    Causal attention for unbiased visual recognition,

    Tan Wang and et al., “Causal attention for unbiased visual recognition,” in CVPR, 2021, pp. 3091–3100

  30. [38]

    The mnist database of handwritten digits,

    Yann LeCun, “The mnist database of handwritten digits,” http://yann. lecun. com/exdb/mnist/, 1998

  31. [39]

    Model-contrastive federated learning,

    Qinbin Li and et al., “Model-contrastive federated learning,” in CVPR, 2021, pp. 10713–10722

  32. [40]

    Deep residual learning for image recognition,

    Kaiming He and et al., “Deep residual learning for image recognition,” in CVPR, 2016, pp. 770–778

  33. [41]

    Communication-efficient learning of deep networks from decentralized data,

    Brendan McMahan and et al., “Communication-efficient learning of deep networks from decentralized data,” in Artificial intelligence and statistics. PMLR, 2017, pp. 1273–1282

  34. [42]

    Federated optimization in heterogeneous networks,

    Tian et al. Li, “Federated optimization in heterogeneous networks,” Proceedings of Machine learning and systems , vol. 2, pp. 429–450, 2020

  35. [43]

    Out-of-distribution generalization of federated learning via implicit invariant relationships,

    Yaming Guo and et al., “Out-of-distribution generalization of federated learning via implicit invariant relationships,” in ICML. PMLR, 2023, pp. 11905–11933

  36. [44]

    Fair federated learning under domain skew with local consistency and domain diversity,

    Yuhang Chen and et al., “Fair federated learning under domain skew with local consistency and domain diversity,” inCVPR, 2024, pp. 12077– 12086

  37. [45]

    Multi-source collaborative gradient discrepancy minimization for federated domain generalization,

    Yikang Wei and et al., “Multi-source collaborative gradient discrepancy minimization for federated domain generalization,” in AAAI, 2024, vol. 38, pp. 15805–15813

  38. [46]

    Grad-cam: Visual explanations from deep networks via gradient-based localization,

    Ramprasaath R Selvaraju and et al., “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in ICCV, 2017, pp. 618–626

  39. [47]

    Cross-modal content inference and feature enrichment for cold-start recommendation,

    Haokai Ma and et al., “Cross-modal content inference and feature enrichment for cold-start recommendation,” in IJCNN, 2023, pp. 1–8

  40. [48]

    Cross-modal learning using privileged information for long-tailed image classification,

    Xiangxian Li and et al., “Cross-modal learning using privileged information for long-tailed image classification,” CVM, vol. 10, no. 5, pp. 981–992, 2024

  41. [49]

    Multi-channel attentive weighting of visual frames for multimodal video classification,

    Yuqing Wang and et al., “Multi-channel attentive weighting of visual frames for multimodal video classification,” in IJCNN, 2023, pp. 1–8

  42. [50]

    Machine learning empowering drug discovery: Applications, opportunities and challenges,

    Xin Qi and et al, “Machine learning empowering drug discovery: Applications, opportunities and challenges,” Molecules, vol. 29, no. 4, pp. 903, 2024

  43. [51]

    Heterogeneous fusion of semantic and collaborative information for visually-aware food recommendation,

    Lei Meng and et al, “Heterogeneous fusion of semantic and collaborative information for visually-aware food recommendation,” in MM, 2020, pp. 3460–3468

  44. [52]

    Plug-in diffusion model for sequential recom- mendation,

    Haokai Ma and et al., “Plug-in diffusion model for sequential recom- mendation,” in AAAI, 2024, pp. 8886–8894

  45. [53]

    Triple sequence learning for cross-domain recommendation,

    Haokai Ma and et al., “Triple sequence learning for cross-domain recommendation,” ACM Trans. Inf. Syst., vol. 42, no. 4, pp. 91:1–91:29, 2024

  46. [54]

    Unsupervised contrastive masking for visual haze classification,

    Jingyu Li and et al., “Unsupervised contrastive masking for visual haze classification,” in ICMR, 2022, pp. 426–434

  47. [55]

    Modeling event-level causal representation for video classification,

    Yuqing Wang and et al., “Modeling event-level causal representation for video classification,” in MM, 2024, pp. 3936–3944

  48. [56]

    Towards online and personalized daily activity recognition, habit modeling, and anomaly detection for the solitary el- derly through unobtrusive sensing,

    Lei Meng and et al., “Towards online and personalized daily activity recognition, habit modeling, and anomaly detection for the solitary el- derly through unobtrusive sensing,” Multimedia Tools and Applications, vol. 76, pp. 10779–10799, 2017

  49. [57]

    Comparative study of adversarial training methods for long-tailed classification,

    Xiangxian Li and et al., “Comparative study of adversarial training methods for long-tailed classification,” in ADVM, 2021, pp. 1–7

  50. [58]

    Prompt learning with cross-modal feature align- ment for visual domain adaptation,

    Jinxing Liu, Junjin Xiao, Haokai Ma, Xiangxian Li, Zhuang Qi, Xiangxu Meng, and Lei Meng, “Prompt learning with cross-modal feature align- ment for visual domain adaptation,” in CAAI International Conference on Artificial Intelligence . Springer, 2022, pp. 416–428

  51. [59]

    Robust visual tracking via iterative gradient descent and threshold selection,

    Zhuang Qi, Junlin Zhang, and Xin Qi, “Robust visual tracking via iterative gradient descent and threshold selection,” arXiv preprint arXiv:2406.00589, 2024

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.