Pith. sign in

REVIEW 5 major objections 6 minor 59 references

Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion

T0 review · 5 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read pFedDC claims that fusing global and local text and vision prompts through per-client cross-attention consistently outperforms prior federated prompt-learning methods on nine heterogeneous datasets.

desk verdict Useful incremental FL prompt method, but the unexplained pseudo-label regeneration step in Sec. 4.1 is a real confound that should be resolved before the SOTA claims are taken at face value. read the letter →

arxiv 2506.21144 v1 pith:PVEQZLEA submitted 2025-06-26 cs.LG cs.CV

Yuguang Zhang , Kuangpu Guo , Zhihe Lu , Yunbo Wang , Jian Liang This is my paper
classification cs.LGcs.CV
keywords federatedlearningpersonalizedpromptvision-languagemodelsdual-promptoptimizationcross-attentionfusionlabelshiftdomain
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 proposes pFedDC, a personalized federated learning method for pretrained vision-language models in which each client maintains four sets of tunable prompts—global and local text prompts plus global and local vision prompts—and uses client-owned cross-attention modules to fuse the global and local versions before feeding them into the frozen CLIP encoders. The claim is that this dual-prompt design handles data heterogeneity better than existing federated prompt-learning methods, which tune text prompts alone or share prompts without personalization. The paper backs the claim with experiments on nine datasets covering label shift, domain shift, and both at once, where pFedDC reports the highest average accuracy in nearly every configuration. If true, the method offers a communication-light way to personalize foundation models under privacy constraints when clients have different labels and different visual domains.

What carries the argument

The load-bearing object is the four-prompt tuple plus two per-client cross-attention modules. The prompts are initialized from the aggregated global state each round: $P = [P^t_g, P^t_l, P^v_g, P^v_l]$, with global text and vision prompts sent to the server and locally retained local prompts. The cross-attention modules compute keys, queries, and values from the global and local prompt pairs, producing fused text and vision prompts that are fed into the frozen CLIP encoders; they are updated locally and never communicated. This design lets the same shared global prompt coexist with a client-specific interpretation of it, addressing label shift through the text side and domain shift through the vision side.

What would settle it

Run the same nine-dataset comparison with the pseudo-label regeneration step either removed from pFedDC or added to every baseline; if the accuracy gap shrinks to near zero, the dual-prompt cross-fusion design is not what carries the result.

Watch

Extended reading notes

Core claim

On its own terms, pFedDC's central discovery is that the failure mode of federated prompt learning under simultaneous label and domain shift is the mismatch between globally shared text prompts and client-specific visual features, and that the fix is to personalize both modalities. Each client optimizes a global text prompt, a local text prompt, a global vision prompt, and a local vision prompt; only the global prompts are sent to the server for FedAvg aggregation, while the local prompts and the cross-attention modules stay on the client. A text cross-attention module fuses global and local text prompts, and a vision cross-attention module fuses global and local vision prompts, so each client's encoder input is a personalized blend of federation-wide consensus and client-specific semantics. Across the reported settings the fused dual-prompt model outperforms the prior single-modality or non-personalized baselines, and the ablation shows that either attention module alone helps while both together give the best accuracy.

Load-bearing premise

The reported gains assume the pseudo-label regeneration step described in Section 4.1 either is applied equally to all baselines or has no material effect on accuracy, since the method section itself never mentions it.

Editorial extensions

If this is right

  • If the reported accuracy holds, federated clients can personalize a frozen vision-language model by exchanging only prompt vectors, keeping communication cost small while still capturing client-specific semantics.
  • The vision-prompt branch gives a direct handle on domain shift: clients in different visual domains no longer rely on a single frozen image-encoder representation.
  • Personalized cross-attention modules add no communication overhead because they never leave the client, so the design scales to larger client populations without extra upload cost.
  • The ablation results indicate that each cross-attention module alone improves accuracy and that both together give the best result, making the fusion step the active ingredient in the reported gains.

Reading between the lines

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

  • The pseudo-label regeneration described in Section 4.1 is not part of the method section's formulation; if it behaves like self-training, a natural test is to apply the same regeneration to all baselines or remove it from pFedDC to isolate the dual-prompt contribution.
  • Because the method is agnostic to the server aggregator, the same global/local dual-prompt structure could be combined with weighted or robust aggregation rules, which might matter when clients are unreliable.
  • The cross-attention fusion is prompt-specific but not architecture-specific; the design could transfer to other promptable encoders beyond CLIP, for example text-only or audio models, or to tasks beyond image classification.
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

5 major / 6 minor

Summary. The paper proposes pFedDC, a personalized federated learning method for CLIP-based prompt tuning. Each client learns global and local text and vision prompts; global prompts are aggregated with FedAvg while local prompts and per-client cross-attention modules remain local. The cross-attention modules fuse global and local prompts before feeding the frozen text and image encoders. Experiments on seven label-shift datasets and on DomainNet and Office-Caltech10 under domain shift and joint domain+label shift compare pFedDC with Zero-shot CLIP, PromptFL, Promptprox, pFedPrompt, and FedOTP, reporting mean accuracies over three trials, robustness analyses over client numbers, participation rates and communication rounds, an RN50 backbone study, and an ablation of the cross-attention modules. The paper claims that pFedDC consistently outperforms state-of-the-art methods.

Significance. If the empirical claims are reliable, pFedDC is a useful contribution to federated prompt learning: it extends prior text-only or single-prompt personalization to simultaneous vision and text prompt personalization and introduces a plausible mechanism for combining global and local knowledge through cross-attention. The paper is original relative to FedOTP and pFedPrompt, and the benchmark suite covers meaningful heterogeneity types. Strengths include the ablation isolating the two cross-attention modules, the robustness experiments under varying client counts, participation rates, and communication rounds, and the additional RN50 backbone check. However, the central empirical claim is currently undermined by an unexplained pseudo-label regeneration step in the implementation, missing variance reporting, and a demonstrable error in the Table 5 averages; these issues must be resolved before the stated conclusion can be accepted.

major comments (5)
  1. [Sec. 4.1] The implementation details state: 'We conduct 20 communication rounds for all experimental datasets, regenerating pseudo labels using the updated local model every 5 communication rounds.' This is the only occurrence of pseudo-labeling in the paper. Section 3 and Algorithm 1 define a fully supervised objective with true labels y and cross-entropy loss (Eq. (6)), and no unlabeled data, pseudo-label generation rule, or use of pseudo-labels in local training is described. Since self-training with pseudo-labels is known to improve accuracy under label shift, the reported gains (e.g., Table 1 avg 94.43 vs. FedOTP 93.06; Table 4 avg 47.80 vs. FedOTP 46.49) could be due to this unexplained mechanism rather than to the dual-prompt cross-fusion design. The paper must either remove this sentence, specify precisely how pseudo-labels are generated and used, and show that the same procedure is applied to all baselines, or provide an ablation that isolates its effect.
  2. [Table 5] The average column in Table 5 reports exactly the same values as Table 1 (e.g., pFedDC 94.43, FedOTP 93.06) even though the per-dataset entries differ substantially. Averaging the seven listed RN50 entries gives approximately 91.58 for pFedDC and 90.07 for FedOTP, not the reported 94.43 and 93.06. This suggests the averages were copied from the ViT-B/16 table and invalidates the claim that pFedDC 'consistently outperforms' with the RN50 backbone. Please recompute the averages, or explain if the average is computed over a different set of runs.
  3. [Tables 1-4 and Fig. 3] All tables report only the mean of three trials, with no standard deviations, and the curves in Fig. 3 show no error bars. Several reported advantages are small (e.g., Table 4 DomainNet average 47.80 vs. 46.49; Office-Caltech10 Amazon 27.36 vs. 25.54), so without variance or significance testing the 'consistently outperforms' claim is not statistically supported. Please report per-trial results or standard deviations/error bars and, where appropriate, significance tests.
  4. [Sec. 2.2 and Sec. 4.1] The baseline set is too narrow for the 'state-of-the-art' claim. The Related Work cites recent personalized federated prompt-learning methods that are not compared, including pFedMoAP [57], FedAPT [32], FedCLIP [30], FedPR [53], pFedPG [10], and DiPrompt [18]. At least the methods specifically designed for joint label and domain shift (e.g., pFedMoAP, DiPrompt) should be included; without them the experiments establish only an advantage over PromptFL, Promptprox, pFedPrompt, and FedOTP.
  5. [Table 1] On DTD under Dirichlet-based label shift with beta=0.1, pFedDC (91.42) is below FedOTP (92.77), so 'consistently outperforms state-of-the-art methods' is not literally accurate even within the reported baselines. Please either revise the claim or provide an explanation for this exception.
minor comments (6)
  1. [Table 2] The DTD entry '91.9587.03' appears to be missing a separator and should read '91.95 87.03'.
  2. [Fig. 3] The legend uses 'FedOPT' while the text and tables use 'FedOTP'; please unify the notation.
  3. [Eq. (1)] Equation (1) has an unmatched parenthesis: 'exp(sim(f^v(x),f^t_c))/τ)' should be 'exp(sim(f^v(x),f^t_c)/τ)'.
  4. [Algorithm 1] Algorithm 1 line 8 initializes P^r_k <- P^{r-1}_k, but Section 3.2 says global prompts are replaced by the server aggregate while local prompts are retained; please clarify the exact local state update to avoid ambiguity.
  5. [Table captions] Several table captions contain typographical artifacts, such as 'T able', 'EuroSATA vg.', and 'A vg.'; please clean these up.
  6. [Sec. 4.1] If the pseudo-label sentence is retained, the phrase 'using the updated local model' is unclear because only prompts and attention modules are trained while the CLIP encoders are frozen; please specify what 'local model' means.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central claim rests on direct benchmark comparisons against external datasets; the minor self-citations are contextual and not load-bearing.

full rationale

The paper's central claim is that pFedDC consistently outperforms state-of-the-art methods on heterogeneous federated prompt learning. The evidence for this claim is direct empirical measurement on public benchmarks (Caltech101, Flowers102, OxfordPets, DTD, CUB, UCF101, EuroSAT, DomainNet, Office-Caltech10) against external baselines (Zero-shot CLIP, PromptFL, Promptprox, pFedPrompt, FedOTP). No derivation is offered that would reduce a prediction to a fitted input. The method is described through a supervised objective (Eq. 6) using cross-entropy loss and true labels, and the reported accuracies are benchmark outcomes, not predictions derived from fitted parameters. The only self-citations are contextual references in the introduction and related work (Refs. [8] and [49] include co-authors), and they do not carry the argument; removing them would not change any result. The pseudo-label regeneration step mentioned in Sec. 4.1 ('regenerating pseudo labels using the updated local model every 5 communication rounds') is a genuine reproducibility and fairness-of-comparison concern, because it is not described in the method section and no ablation isolates it. However, this is a potential confound rather than a circular dependency: the paper does not define a result in terms of the pseudo-label procedure or rename a fitted quantity as a prediction. Under the specified circularity patterns, no step reduces to its own inputs. The score reflects only the presence of minor non-load-bearing self-citations, not any circular reasoning.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on a small number of tuned hyperparameters, the frozen-CLIP transfer assumption, FedAvg behavior on prompts, and the fairness of an underdescribed pseudo-label procedure. No new physical entities are introduced. The main unexamined burden is the pseudo-label regeneration, which could act as an uncontrolled self-training boost.

free parameters (5)
  • Learning rate = 0.01
    Set uniformly for all prompt and attention updates in Sec. 4.1 with no sensitivity analysis or schedule.
  • Communication rounds = 20
    All experiments use 20 rounds; no convergence criterion or round sensitivity for the central results is provided.
  • Pseudo-label regeneration interval = every 5 rounds
    Introduced in Sec. 4.1 without a method-section description or an ablation controlling for its effect.
  • Cross-attention architecture hyperparameters = not reported
    Number of heads, layers, dimensions, and initialization of the attention modules are unspecified in Sec. 3.3.
  • Prompt length and initialization = not reported
    The length and initialization of global and local text and vision prompts are not given, which affects model capacity and outcomes.
assumptions (4)
  • domain assumption Pretrained CLIP encoders are frozen and provide sufficiently transferable features for heterogeneous federated tasks.
    Assumed throughout the method in Sec. 3.1; the paper provides no analysis of when this assumption fails.
  • domain assumption FedAvg aggregation of global prompts preserves common knowledge across clients.
    Used in Eqs. (2) and (3) with no convergence or bias analysis for prompt spaces.
  • standard math Cross-entropy loss on local data is the correct training signal for personalization under combined label and domain shift.
    Standard supervised objective in Eq. (6), but its adequacy for the stated heterogeneity is not examined.
  • ad hoc to paper Pseudo-label regeneration every 5 rounds improves or does not bias the comparison.
    Introduced in Sec. 4.1 without details or a baseline control; the central comparison depends on this step being fair.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion." pith.science (2026). https://pith.science/paper/PVEQZLEA

@misc{pith2026250621144,
  author       = {Pith},
  title        = {Pith review of: Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PVEQZLEA}},
  note         = {Machine review of arXiv:2506.21144}
}
read the original abstract

Federated learning (FL) enables collaborative model training across decentralized clients without sharing local data, but is challenged by heterogeneity in data, computation, and communication. Pretrained vision-language models (VLMs), with their strong generalization and lightweight tuning via prompts, offer a promising solution. However, existing federated prompt-learning methods rely only on text prompts and overlook joint label-domain distribution shifts. In this paper, we propose a personalized FL framework based on dual-prompt learning and cross fusion, termed pFedDC. Specifically, each client maintains both global and local prompts across vision and language modalities: global prompts capture common knowledge shared across the federation, while local prompts encode client-specific semantics and domain characteristics. Meanwhile, a cross-fusion module is designed to adaptively integrate prompts from different levels, enabling the model to generate personalized representations aligned with each client's unique data distribution. Extensive experiments across nine datasets with various types of heterogeneity show that pFedDC consistently outperforms state-of-the-art methods.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

59 extracted references · 44 canonical work pages

  1. [57]

    Mixture of experts made personalized: Federated prompt learning for vision-language models,

    J. Luo, C. Chen, and S. Wu, “Mixture of experts made personalized: Federated prompt learning for vision-language models,” in Proc. ICLR, 2025

    work page 2025

  2. [32]

    Federated adaptive prompt tuning for multi-domain collaborative learning,

    S. Su, M. Yang, B. Li, and X. Xue, “Federated adaptive prompt tuning for multi-domain collaborative learning,” Proc. AAAI, 2024

    work page 2024

  3. [30]

    Fedclip: Fast generalization and personalization for clip in federated learning,

    W. Lu, X. Hu, J. Wang, and X. Xie, “Fedclip: Fast generalization and personalization for clip in federated learning,” arXiv:2302.13485, 2023

    arXiv 2023

  4. [53]

    Learning federated visual prompt in null space for mri reconstruction,

    C.-M. Feng, B. Li, X. Xu, Y. Liu, H. Fu, and W. Zuo, “Learning federated visual prompt in null space for mri reconstruction,” in Proc. CVPR, 2023

    work page 2023

  5. [10]

    Efficient model person- alization in federated learning via client-specific prompt generation,

    F.-E. Yang, C.-Y. Wang, and Y.-C. F. Wang, “Efficient model person- alization in federated learning via client-specific prompt generation,” in Proc. CVPR, 2023

    work page 2023

  6. [18]

    Diprompt: Disentangled prompt tuning for multiple latent domain generalization in federated learning,

    S. Bai, J. Zhang, S. Guo, S. Li, J. Guo, J. Hou, T. Han, and X. Lu, “Diprompt: Disentangled prompt tuning for multiple latent domain generalization in federated learning,” in Proc. CVPR, 2024. Springer Nature 2021 LATEX template Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion 17

    work page 2024

  7. [1]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark et al. , “Learning transferable visual models from natural language supervision,” in Proc. ICML, 2021

    work page 2021

  8. [2]

    Promptfl: Let feder- ated participants cooperatively learn prompts instead of models-federated learning in age of foundation model,

    T. Guo, S. Guo, J. Wang, X. Tang, and W. Xu, “Promptfl: Let feder- ated participants cooperatively learn prompts instead of models-federated learning in age of foundation model,” IEEE Transactions on Mobile Computing, pp. 5179–5194, 2023

    work page 2023

Show all 59 references
  1. [3]

    Federated optimization in heterogeneous networks,

    T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V. Smith, “Federated optimization in heterogeneous networks,” Proceedings of Machine learning and systems , pp. 429–450, 2020

    2020

  2. [4]

    Global and local prompts coop- eration via optimal transport for federated learning,

    H. Li, W. Huang, J. Wang, and Y. Shi, “Global and local prompts coop- eration via optimal transport for federated learning,” in Proc. CVPR , 2024

    2024

  3. [5]

    Communication-efficient learning of deep networks from decentralized data,

    B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Proc. AISTATS, 2017

    2017

  4. [6]

    Federated foundation mod- els: Privacy-preserving and collaborative learning for large models,

    S. Yu, J. P. Mu˜ noz, and A. Jannesari, “Federated foundation mod- els: Privacy-preserving and collaborative learning for large models,” arXiv:2305.11414, 2023. Springer Nature 2021 LATEX template 16 Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion

    2023 arXiv

  5. [7]

    Feddisco: Fed- erated learning with discrepancy-aware collaboration,

    R. Ye, M. Xu, J. Wang, C. Xu, S. Chen, and Y. Wang, “Feddisco: Fed- erated learning with discrepancy-aware collaboration,” in Proc. ICML , 2023

    2023

  6. [8]

    Not all minori- ties are equal: Empty-class-aware distillation for heterogeneous federated learning,

    K. Guo, Y. Ding, J. Liang, R. He, Z. Wang, and T. Tan, “Not all minori- ties are equal: Empty-class-aware distillation for heterogeneous federated learning,” in Proc. AAAI, 2025

    2025

  7. [9]

    Harmonizing generalization and personalization in federated prompt learning,

    T. Cui, H. Li, J. Wang, and Y. Shi, “Harmonizing generalization and personalization in federated prompt learning,” in Proc. ICML, 2024

    2024

  8. [11]

    Scaling up visual and vision-language representation learning with noisy text supervision,

    C. Jia, Y. Yang, Y. Xia, Y.-T. Chen, Z. Parekh, H. Pham, Q. Le, Y.- H. Sung, Z. Li, and T. Duerig, “Scaling up visual and vision-language representation learning with noisy text supervision,” inProc. ICML, 2021

    2021

  9. [12]

    Reduce communication costs and preserve privacy: Prompt tuning method in federated learning,

    H. Zhao, W. Du, F. Li, P. Li, and G. Liu, “Reduce communication costs and preserve privacy: Prompt tuning method in federated learning,” arXiv:2208.12268, 2022

    2022 arXiv

  10. [13]

    Pre- train, prompt, and predict: A systematic survey of prompting methods in natural language processing,

    P. Liu, W. Yuan, J. Fu, Z. Jiang, H. Hayashi, and G. Neubig, “Pre- train, prompt, and predict: A systematic survey of prompting methods in natural language processing,” ACM Computing Surveys , pp. 1–35, 2023

    2023

  11. [14]

    Fedbn: Federated learning on non-iid features via local batch normalization,

    X. Li, M. Jiang, X. Zhang, M. Kamp, and Q. Dou, “Fedbn: Federated learning on non-iid features via local batch normalization,” inProc. ICLR, 2021

    2021

  12. [15]

    Model-contrastive federated learning,

    Q. Li, B. He, and D. Song, “Model-contrastive federated learning,” in Proc. CVPR, 2021

    2021

  13. [16]

    Pfedprompt: Learning personalized prompt for vision-language models in federated learning,

    T. Guo, S. Guo, and J. Wang, “Pfedprompt: Learning personalized prompt for vision-language models in federated learning,” in Proceedings of the ACM Web Conference 2023 , 2023

    2023

  14. [17]

    Dual prompt tuning for domain-aware federated learning,

    G. Wei, F. Wang, A. Shah, and R. Chellappa, “Dual prompt tuning for domain-aware federated learning,” arXiv:2310.03103, 2023

    2023 arXiv

  15. [19]

    Learning to prompt for vision- language models,

    K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Learning to prompt for vision- language models,” International Journal of Computer Vision , pp. 2337– 2348, 2022

    2022

  16. [20]

    Visual prompt tuning,

    M. Jia, L. Tang, B.-C. Chen, C. Cardie, S. Belongie, B. Hariharan, and S.-N. Lim, “Visual prompt tuning,” in Proc. ECCV, 2022

    2022

  17. [21]

    Dual modality prompt tuning for vision-language pre-trained model,

    Y. Xing, Q. Wu, D. Cheng, S. Zhang, G. Liang, P. Wang, and Y. Zhang, “Dual modality prompt tuning for vision-language pre-trained model,” in Proc. ACM-MM, 2023

    2023

  18. [22]

    Personalized federated learning with feature alignment and classifier collaboration,

    J. Xu, X. Tong, and S.-L. Huang, “Personalized federated learning with feature alignment and classifier collaboration,” arXiv:2306.11867, 2023

    2023 arXiv

  19. [23]

    Federated learning with personalization layers,

    M. G. Arivazhagan, V. Aggarwal, A. K. Singh, and S. Choudhary, “Federated learning with personalization layers,”arXiv:1912.00818, 2019

    1912 arXiv

  20. [24]

    Exploiting shared representations for personalized federated learning,

    L. Collins, H. Hassani, A. Mokhtari, and S. Shakkottai, “Exploiting shared representations for personalized federated learning,” in Proc. ICML, 2021

    2021

  21. [25]

    Fedbabu: Towards enhanced representa- tion for federated image classification,

    J. Oh, S. Kim, and S.-Y. Yun, “Fedbabu: Towards enhanced representa- tion for federated image classification,” arXiv:2106.06042, 2021

    2021 arXiv

  22. [26]

    Fedtp: Federated learning by transformer personalization,

    H. Li, Z. Cai, J. Wang, J. Tang, W. Ding, C.-T. Lin, and Y. Shi, “Fedtp: Federated learning by transformer personalization,” IEEE Transactions on Neural Networks and Learning Systems , pp. 13 426 – 13 440, 2023

    2023

  23. [27]

    Partialfed: Cross-domain per- sonalized federated learning via partial initialization,

    B. Sun, H. Huo, Y. Yang, and B. Bai, “Partialfed: Cross-domain per- sonalized federated learning via partial initialization,” Proc. NeurIPS , 2021

    2021

  24. [28]

    Text-driven prompt generation for vision-language models in federated learning,

    C. Qiu, X. Li, C. K. Mummadi, M. R. Ganesh, Z. Li, L. Peng, and W.- Y. Lin, “Text-driven prompt generation for vision-language models in federated learning,” arXiv:2310.06123, 2023

    2023 arXiv

  25. [29]

    Position paper: Assessing robustness, privacy, and fairness in federated learning integrated with foundation models,

    X. Li and J. Wang, “Position paper: Assessing robustness, privacy, and fairness in federated learning integrated with foundation models,” arXiv:2402.01857, 2024

    2024

  26. [31]

    Hepco: Data-free heterogeneous prompt consolidation for continual federated learning,

    S. Halbe, J. S. Smith, J. Tian, and Z. Kira, “Hepco: Data-free heterogeneous prompt consolidation for continual federated learning,” arXiv:2306.09970, 2023. Springer Nature 2021 LATEX template 18 Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion

    2023 arXiv

  27. [33]

    Squeeze-and-excitation networks,

    J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in Proc. CVPR, 2018

    2018

  28. [34]

    Eca-net: Effi- cient channel attention for deep convolutional neural networks,

    Q. Wang, B. Wu, P. Zhu, P. Li, W. Zuo, and Q. Hu, “Eca-net: Effi- cient channel attention for deep convolutional neural networks,” in Proc. CVPR, 2020

    2020

  29. [35]

    Stand-alone self-attention in vision models,

    P. Ramachandran, N. Parmar, A. Vaswani, I. Bello, A. Levskaya, and J. Shlens, “Stand-alone self-attention in vision models,” Proc. NeurIPS, 2019

    2019

  30. [36]

    Exploring self-attention for image recognition,

    H. Zhao, J. Jia, and V. Koltun, “Exploring self-attention for image recognition,” in Proc. CVPR, 2020

    2020

  31. [37]

    Learning generative visual models from few training examples: An incremental bayesian approach tested on 101 object categories,

    L. Fei-Fei, R. Fergus, and P. Perona, “Learning generative visual models from few training examples: An incremental bayesian approach tested on 101 object categories,” in Proc. CVPR Workshops , 2004

    2004

  32. [38]

    Automated flower classification over a large number of classes,

    M.-E. Nilsback and A. Zisserman, “Automated flower classification over a large number of classes,” in Proc. ICVGIP, 2008

    2008

  33. [39]

    Cats and dogs,

    O. M. Parkhi, A. Vedaldi, A. Zisserman, and C. Jawahar, “Cats and dogs,” in Proc. CVPR, 2012

    2012

  34. [40]

    Eurosat: A novel dataset and deep learning benchmark for land use and land cover classifica- tion,

    P. Helber, B. Bischke, A. Dengel, and D. Borth, “Eurosat: A novel dataset and deep learning benchmark for land use and land cover classifica- tion,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , pp. 2217–2226, 2019

    2019

  35. [41]

    Ucf101: A dataset of 101 human actions classes from videos in the wild,

    K. Soomro, “Ucf101: A dataset of 101 human actions classes from videos in the wild,” arXiv:1212.0402, 2012

    2012 arXiv

  36. [42]

    Describ- ing textures in the wild,

    M. Cimpoi, S. Maji, I. Kokkinos, S. Mohamed, and A. Vedaldi, “Describ- ing textures in the wild,” in Proc. CVPR, 2014

    2014

  37. [43]

    The caltech-ucsd birds-200-2011 dataset,

    C. Wah, S. Branson, P. Welinder, P. Perona, and S. Belongie, “The caltech-ucsd birds-200-2011 dataset,” Technical Report CNS-TR-2011- 001, California Institute of Technology , 2011

    2011

  38. [44]

    Learning multiple layers of features from tiny images,

    A. Krizhevsky, G. Hinton et al., “Learning multiple layers of features from tiny images,” Technical report, University of Toronto , 2009

    2009

  39. [45]

    Moment matching for multi-source domain adaptation,

    X. Peng, Q. Bai, X. Xia, Z. Huang, K. Saenko, and B. Wang, “Moment matching for multi-source domain adaptation,” in Proc. ICCV, 2019. Springer Nature 2021 LATEX template Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion 19

    2019

  40. [46]

    Geodesic flow kernel for unsupervised domain adaptation,

    B. Gong, Y. Shi, F. Sha, and K. Grauman, “Geodesic flow kernel for unsupervised domain adaptation,” in Proc. CVPR, 2012

    2012

  41. [47]

    Non-local neural networks,

    X. Wang, R. Girshick, A. Gupta, and K. He, “Non-local neural networks,” in Proc. CVPR, 2018

    2018

  42. [48]

    No fear of heterogeneity: Classifier calibration for federated learning with non-iid data,

    M. Luo, F. Chen, D. Hu, Y. Zhang, J. Liang, and J. Feng, “No fear of heterogeneity: Classifier calibration for federated learning with non-iid data,” in Proc. NeurIPS, 2021

    2021

  43. [49]

    Towards under- standing and mitigating dimensional collapse in heterogeneous federated learning,

    Y. Shi, J. Liang, W. Zhang, V. Y. Tan, and S. Bai, “Towards under- standing and mitigating dimensional collapse in heterogeneous federated learning,” in Proc. ICLR, 2022

    2022

  44. [50]

    Ditto: Fair and robust federated learning through personalization,

    T. Li, S. Hu, A. Beirami, and V. Smith, “Ditto: Fair and robust federated learning through personalization,” in Proc. ICML, 2021

    2021

  45. [51]

    Personalized federated learning with moreau envelopes,

    C. T Dinh, N. Tran, and J. Nguyen, “Personalized federated learning with moreau envelopes,” in Proc. NeurIPS, 2020

    2020

  46. [52]

    Fedcp: Separating feature information for personalized federated learn- ing via conditional policy,

    J. Zhang, Y. Hua, H. Wang, T. Song, Z. Xue, R. Ma, and H. Guan, “Fedcp: Separating feature information for personalized federated learn- ing via conditional policy,” in Proc. KDD, 2023

    2023

  47. [54]

    Residual attention network for image classification,

    F. Wang, M. Jiang, C. Qian, S. Yang, C. Li, H. Zhang, X. Wang, and X. Tang, “Residual attention network for image classification,” in Proc. CVPR, 2017

    2017

  48. [55]

    Cbam: Convolutional block attention module,

    S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “Cbam: Convolutional block attention module,” in Proc. ECCV, 2018

    2018

  49. [56]

    Pytorch: An imperative style, high-performance deep learning library,

    A. Paszke, “Pytorch: An imperative style, high-performance deep learning library,” arXiv:1912.01703, 2019

    1912 arXiv

  50. [58]

    Dynamic heterogeneous federated learning with multi-level prototypes,

    S. Guo, H. Wang, and X. Geng, “Dynamic heterogeneous federated learning with multi-level prototypes,” Pattern Recognition, p. 110542, 2024

    2024

  51. [59]

    Addressing skewed heterogeneity via federated prototype rectification with personalization,

    S. Guo, H. Wang, S. Lin, Z. Kou, and X. Geng, “Addressing skewed heterogeneity via federated prototype rectification with personalization,” IEEE Transactions on Neural Networks and Learning Systems , pp. 8442 Springer Nature 2021 LATEX template 20 Personalized Federated Learni...

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.