Pith. sign in

REVIEW 3 major objections 6 minor 41 references

Slot Attention-based Feature Filtering for Few-Shot Learning

T0 review · 3 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read SAFF claims slot attention filtering of ViT patch embeddings improves few-shot classification on four benchmarks, beating CPEA.

desk verdict A clean, incremental few-shot method whose reported gains are small and partly unsupported; deserves refereeing, but the evidence needs strengthening. read the letter →

arxiv 2508.09699 v1 pith:VWTJDFMR submitted 2025-08-13 cs.CV

Javier Rodenas , Eduardo Aguilar , Petia Radeva This is my paper
classification cs.CV
keywords few-shotlearningslotattentionfeaturefilteringVisionTransformerclasstokenpatchembeddingsmetricimageclassification
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

The paper proposes a feature-filtering step for few-shot image classification: slot attention is applied to the patch embeddings of a Vision Transformer, initialized with the class token, and refined over several iterations. Slots whose cosine similarity to the class token falls below the mean are under-weighted, and the surviving slots' attention maps form a soft mask that reweights the patch embeddings before classification. On CIFAR-FS, FC100, miniImageNet and tieredImageNet, this filtering improves accuracy over the CPEA baseline, most clearly in 5-shot settings (e.g., 90.30% vs 89.50% on CIFAR-FS), while also reducing variance across training seeds. The paper's claim is that slot attention is a better attention mechanism for this task than dot-product or cross-attention alternatives.

What carries the argument

Slot attention (Locatello et al.) configured with 5 slots, 5 iterations, initialized with the ViT class token instead of random values. The filtering step uses L2-normalized cosine similarity between refined slots and the class token, min-max normalization, a binary mask at threshold 0.5, and a weighted combined attention mask applied to the patch embeddings, followed by a class-aware addition of the class token. This combined mask is the mechanism that under-weights irrelevant patches while preserving contextual information.

What would settle it

Run SAFF on CIFAR-FS 5-shot with the filtering step ablated (use the average of all five slot attention maps instead of only those above the 0.5 threshold) across at least 10 seeds; if the ablated version matches or exceeds 90.30%, then the class-token-based mask is not what produces the reported gain.

Watch

Extended reading notes

Core claim

The central claim is that slot attention, seeded with the class token and filtered by class-token similarity, can decompose a ViT's patch embeddings into a few class-aware attention slots, and that the weighted combination of the most class-aligned slots produces a refined representation that improves few-shot classification. Concretely, SAFF L2-normalizes the refined slots and class token, computes cosine similarity, min-max scales it to [0,1], keeps only slots with similarity above 0.5, averages their attention maps, multiplies the patch embeddings by this combined mask, and adds $\lambda = 2$ times the class token. A dense query–support similarity matrix processed by an MLP produces the c

Load-bearing premise

The method assumes the ViT class token is a dependable class-aware reference, so that slots with below-average cosine similarity to it are safely treated as irrelevant and can be down-weighted without losing discriminative information.

Editorial extensions

If this is right

  • SAFF improves 5-way 5-shot accuracy over CPEA on all four tested benchmarks, with the largest gain on CIFAR-FS (90.30% vs 89.50%) and smaller but consistent gains in 1-shot.
  • Slot attention outperforms dot-product attention, cross-attention query-support, and cross-attention patch-token when used in the same filtering role.
  • Weighted masking beats binary masking, indicating that softly reducing irrelevant patch weights preserves useful context in few-shot settings.
  • The configuration of 5 slots and 5 iterations is optimal; increasing either parameter does not yield further gains.
  • The improvements over CPEA are statistically significant at p ≤ 0.001 on tieredImageNet, miniImageNet, and FC100, but not on CIFAR-FS (p = 0.16).

Reading between the lines

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

  • The class-token-similarity filter is only as good as the class token itself; on tasks with heavy background or distractor cues, the same threshold could discard discriminative patches, so the method's robustness is an empirical question the paper does not fully settle.
  • Because the filter is applied as a module on top of a frozen ViT, it may transfer to other transformer backbones or self-supervised features without retraining the backbone; the paper only demonstrates ViT-S/16 with one pretraining strategy.
  • The 1-shot gains are much smaller than 5-shot gains, suggesting the slot filter exploits within-class consistency that a single support sample rarely provides; testing with more support shots or class-wise slots would clarify the mechanism.
  • A direct ablation that removes the filtering step (keeping all slots) and compares against the full SAFF would isolate how much of the gain comes from filtering rather than from slot attention itself.
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

3 major / 6 minor

Summary. The paper proposes SAFF (Slot Attention-based Feature Filtering for Few-Shot Learning), a module inserted between a ViT feature extractor and the classification head. Slots are initialized from the class token, refined iteratively, and then filtered by cosine similarity to the class token: after min-max normalization, slots above the 0.5 threshold are averaged into a combined attention mask that weights patch embeddings; a weighted class-token term is added before building a dense similarity matrix that is fed to an MLP. The method is evaluated on CIFAR-FS, FC100, miniImageNet, and tieredImageNet in 5-way 1-shot and 5-shot settings, with comparisons concentrated on CPEA under the same ViT-S/16 backbone. The reported gains over CPEA are small (roughly 0.15–0.8 percentage points across settings), with 1-shot significance claimed for three of four datasets via McNemar tests.

Significance. If the gains are real, SAFF is a modest but reasonably clean plug-in contribution: it adapts slot attention to feature filtering in few-shot classification, provides ablations over slot counts/iterations and binary-vs-weighted masks, and reports three-seed median/mean/std results on four standard benchmarks. The strongest positive aspects are the controlled comparison against CPEA with the same backbone and the explicit attempt to quantify statistical significance on 1-shot results. However, the contribution is incremental, the headline 5-shot improvements are not statistically supported, and the hyperparameter-selection procedure does not rule out test-set fitting as an alternative explanation for the small margins.

major comments (3)
  1. [§4.3, Tables 2–3 and Table 5] The paper's central claim covers both 1-shot and 5-shot accuracy, but the only statistical test reported (Table 5) is for 5-way 1-shot. For 5-shot, the gains over CPEA are small (0.26–0.80 percentage points) and the 95% confidence intervals overlap in every case, e.g., CIFAR-FS 90.30±0.41 vs 89.50±0.42 and tieredImageNet 88.97±0.43 vs 88.71±0.46. With only three training seeds and no paired test across episodes, the current evidence does not substantiate the 5-shot improvements. Please report a paired significance test (e.g., McNemar over the 1,000 test episodes) or confidence intervals for the accuracy difference, and clearly separate 1-shot from 5-shot claims in the text.
  2. [§4.4, Eq. (4), §3.3.2] The hyperparameters (5 slots, 5 iterations, threshold 0.5, λ=2) are selected without a described validation protocol. Section 4.4 states that Table 6 identifies the “optimal configuration” on CIFAR-FS and miniImageNet, but does not say whether this was done on a held-out validation split or on the test benchmarks themselves. The threshold in Eq. (4) is justified qualitatively as “a good balance” and λ is taken from CPEA. Given that the margins over CPEA are below one percentage point, test-set selection of these values is a plausible alternative explanation. Please specify the validation procedure, or provide a sensitivity analysis on validation splits that does not use the final test episodes.
  3. [§3.3.2, Eq. (4)] The filter assumes that the ViT class token is a reliable class-aware reference; slots whose normalized cosine similarity to the class token falls below 0.5 are discarded from the combined mask. This is a load-bearing assumption for the method's claim to “filter irrelevant features.” The paper does not test this assumption directly. A concrete experiment would be to compare the selected slots/mask against ground-truth foreground annotations (where available) or to replace the class-token reference with an alternative (e.g., the mean support embedding) and report accuracy. Without such evidence, the risk remains that the mask discards useful features when the class token is not class-discriminative.
minor comments (6)
  1. [Abstract] The abstract contains an incomplete sentence: “Irrelevant features can significantly degrade few-shot learning performance. This problem is used to match queries and support images...” The second sentence should be rewritten.
  2. [Tables 2–3] Header typo: “5-show” should be “5-shot”.
  3. [Table 5] The last dataset is labeled “CIFAR100,” but the paper uses CIFAR-FS and FC100. This is inconsistent and confusing; presumably it should be CIFAR-FS or FC100.
  4. [§3.3.2] Duplicate word: “following the strategy strategy proposed in [9].” Also, the dimensions in Eq. (3) are not defined; specify that min/max are taken over the N slots per sample.
  5. [§4.5, Table 7] The binary-vs-weighted mask comparison is only reported for 5-shot. It would help to see 1-shot results, since the motivation about few-shot information preservation is strongest there.
  6. [References] Reference [41] is iBOT (“Image BERT pre-training with online tokenizer”); the paper would benefit from naming it explicitly when describing the pretraining strategy, rather than only citing the arXiv title.

Circularity Check

1 steps flagged · score 6.0 of 10

5-shot accuracy on CIFAR-FS/miniImageNet is the result of selecting slot/iteration hyperparameters on those same test benchmarks; otherwise the method is an empirical pipeline with no equation-level circularity.

  1. fitted input called prediction [Section 4.4 (Table 6); results reported in Tables 1-3]
    "The experimental results in Table 6 show that the optimal configuration uses 5 slots with 5 iterations achieving 90.30±0.41 accuracy on CIFAR-FS and 87.19±0.36 accuracy on miniImageNet."

    Table 6 enumerates 3x3 slot/iteration configurations and their 5-shot accuracies on the same test benchmarks, then the paper adopts the highest-accuracy configuration (5 slots, 5 iterations) as the reported SAFF result. Thus the headline accuracies are the maximum over a test-set search, not out-of-sample predictions of a pre-specified model. With 3 slots/3 iterations SAFF obtains 89.40 on CIFAR-FS 5-shot, below CPEA's 89.50; the claimed improvement over CPEA is created by selecting 5/5 on the same test data. This is a fitted input (number of slots and iterations) renamed as predictive performance.

full rationale

The paper's derivation chain is empirical rather than equation-level. The SAFF pipeline is self-contained: ViT patch embeddings are refined through slot attention, slots are filtered by cosine similarity to the class token, the resulting mask weights the embeddings, and classification uses a support-query similarity matrix through an MLP. No equation is defined in terms of the result it is claimed to predict. The self-citation [1] for McNemar's test is not load-bearing, and the key comparisons to CPEA and other methods are independent experiments. The one circular element is Section 4.4/Table 6: hyperparameters are selected on the test benchmarks whose accuracies are later presented as the method's performance, which statistically forces the reported 5-shot gains on CIFAR-FS and miniImageNet. This is a fitted-input-called-prediction step. The 1-shot results and the architectural comparisons retain independent content, so the paper is only partially circular.

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

The method relies on four manually chosen hyperparameters and three domain assumptions about the behavior of class tokens and slot attention. No new entities are introduced.

free parameters (4)
  • Number of slots = 5
    Selected via ablation on test sets (Table 6); not a learned parameter.
  • Number of iterations = 5
    Selected via ablation on test sets (Table 6).
  • Similarity threshold = 0.5
    Set ad hoc to binarize slot mask (Eq. 4); no justification beyond 'good balance'.
  • Class token weight lambda = 2
    Copied from CPEA [9] (Eq. 8); not ablated in this paper.
assumptions (4)
  • domain assumption The class token of ViT reliably represents class-specific information for filtering slots.
    Used in Section 3.3.2; if false, the filtering would discard useful slots.
  • domain assumption Slot attention can separate class-relevant from irrelevant features in patch embeddings.
    Central premise of SAFF; supported only by visualization (Fig. 1).
  • domain assumption Standard few-shot learning protocol with 1,000 episodes provides reliable accuracy estimates.
    Used in Section 4.2.
  • ad hoc to paper CPEA's training hyperparameters transfer to SAFF.
    Section 4.2 states they follow [9] without reporting the values.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Slot Attention-based Feature Filtering for Few-Shot Learning." pith.science (2026). https://pith.science/paper/VWTJDFMR

@misc{pith2026250809699,
  author       = {Pith},
  title        = {Pith review of: Slot Attention-based Feature Filtering for Few-Shot Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VWTJDFMR}},
  note         = {Machine review of arXiv:2508.09699}
}
read the original abstract

Irrelevant features can significantly degrade few-shot learn ing performance. This problem is used to match queries and support images based on meaningful similarities despite the limited data. However, in this process, non-relevant fea tures such as background elements can easily lead to confu sion and misclassification. To address this issue, we pro pose Slot Attention-based Feature Filtering for Few-Shot Learning (SAFF) that leverages slot attention mechanisms to discriminate and filter weak features, thereby improving few-shot classification performance. The key innovation of SAFF lies in its integration of slot attention with patch em beddings, unifying class-aware slots into a single attention mechanism to filter irrelevant features effectively. We intro duce a similarity matrix that computes across support and query images to quantify the relevance of filtered embed dings for classification. Through experiments, we demon strate that Slot Attention performs better than other atten tion mechanisms, capturing discriminative features while reducing irrelevant information. We validate our approach through extensive experiments on few-shot learning bench marks: CIFAR-FS, FC100, miniImageNet and tieredIma geNet, outperforming several state-of-the-art methods.

Figures

Figures reproduced from arXiv: 2508.09699 by the authors.

Figure 1
Figure 1. Visualization of the attention assigned to each slot. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. First, the attention maps for each of the 5 slots, focus on different regions of the image. Then, iterative refinement enhances [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the SAFF pipeline. First, the ViT-S/16 captures both support and query image information from the input image, [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

41 extracted references · 39 canonical work pages

  1. [1]

    Uncertainty-aware selecting for an ensem- ble of deep food recognition models

    Eduardo Aguilar, Bhalaji Nagarajan, and Petia Radeva. Uncertainty-aware selecting for an ensem- ble of deep food recognition models. Computers in Biology and Medicine, 146:105645, 2022. 6

    work page 2022

  2. [2]

    Frozen feature augmentation for few- shot image classification

    Andreas B ¨ar, Neil Houlsby, Mostafa Dehghani, and Manoj Kumar. Frozen feature augmentation for few- shot image classification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 16046–16057, 2024. 1

    work page 2024

  3. [3]

    Henriques, Philip H

    Luca Bertinetto, Jo ˜ao F. Henriques, Philip H. S. Torr, and Andrea Vedaldi. Meta-learning with differentiable closed-form solvers. In 7th International Confer- ence on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019 . OpenReview.net,

    work page 2019

  4. [4]

    Image deformation meta-networks for one-shot learning

    Zitian Chen, Yanwei Fu, Yu-Xiong Wang, Lin Ma, Wei Liu, and Martial Hebert. Image deformation meta-networks for one-shot learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019. 1

    work page 2019

  5. [5]

    Pareto self-supervised training for few-shot learning

    Zhengyu Chen, Jixie Ge, Heshen Zhan, Siteng Huang, and Donglin Wang. Pareto self-supervised training for few-shot learning. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021, pages 13663–13672. Computer Vi- sion Foundation / IEEE, 2021. 7

    work page 2021

  6. [6]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. In 9th International Conference on Learning Representations, ICLR 2021, ...

    work page 2021

  7. [7]

    Adaptive slot attention: Object discovery with dynamic slot number

    Ke Fan, Zechen Bai, Tianjun Xiao, Tong He, Max Horn, Yanwei Fu, Francesco Locatello, and Zheng Zhang. Adaptive slot attention: Object discovery with dynamic slot number. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 23062–23071, 2024. 3

    work page 2024

  8. [8]

    Model-agnostic meta-learning for fast adaptation of deep networks

    Chelsea Finn, Pieter Abbeel, and Sergey Levine. Model-agnostic meta-learning for fast adaptation of deep networks. In Proceedings of the 34th Interna- tional Conference on Machine Learning, ICML 2017, Sydney, NSW, Australia, 6-11 August 2017 , pages 1126–1135. PMLR, 2017. 1, 2

    work page 2017

Show all 41 references
  1. [9]

    Class-aware patch em- bedding adaptation for few-shot image classification

    Fusheng Hao, Fengxiang He, Liu Liu, Fuxiang Wu, Dacheng Tao, and Jun Cheng. Class-aware patch em- bedding adaptation for few-shot image classification. In Proceedings of the IEEE/CVF International Con- ference on Computer Vision (ICCV) , pages 18905– 18915, 2023. 2, 5, 6, 7

    2023

  2. [10]

    Rethinking generalization in few- shot classification

    Markus Hiller, Rongkai Ma, Mehrtash Harandi, and Tom Drummond. Rethinking generalization in few- shot classification. In Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, N...

    2022

  3. [11]

    Relational embedding for few-shot clas- sification

    Dahyun Kang, Heeseung Kwon, Juhong Min, and Minsu Cho. Relational embedding for few-shot clas- sification. In 2021 IEEE/CVF International Con- ference on Computer Vision, ICCV 2021, Montreal, QC, Canada, October 10-17, 2021, pages 8802–8813. IEEE, 2021. 7

    2021

  4. [12]

    Few-shot learning for fea- ture selection with hilbert-schmidt independence cri- terion

    Atsutoshi Kumagai, Tomoharu Iwata, Yasutoshi Ida, and Yasuhiro Fujiwara. Few-shot learning for fea- ture selection with hilbert-schmidt independence cri- terion. In Advances in Neural Information Processing Systems, pages 9577–9590. Curran Associates, Inc.,

  5. [13]

    Clustered-patch element connec- tion for few-shot learning

    Jinxiang Lai, Siqian Yang, Junhong Zhou, Wenlong Wu, Xiaochen Chen, Jun Liu, Bin-Bin Gao, and Chengjie Wang. Clustered-patch element connec- tion for few-shot learning. In Proceedings of the Thirty-Second International Joint Conference on Ar- tificial Intelligence, IJCAI 2023,...

    2023

  6. [14]

    Guided slot attention for unsupervised video object segmentation

    Minhyeok Lee, Suhwan Cho, Dogyoon Lee, Chaewon Park, Jungho Lee, and Sangyoun Lee. Guided slot attention for unsupervised video object segmentation. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR) , pages 3807–3816, 2024. 3

    2024

  7. [15]

    An adaptive plug-and-play network for few-shot learning

    Hao Li, Li Li, Yunmeng Huang, Ning Li, and Yong- tao Zhang. An adaptive plug-and-play network for few-shot learning. In IEEE International Conference on Acoustics, Speech and Signal Processing ICASSP 2023, Rhodes Island, Greece, June 4-10, 2023 , pages 1–5. IEEE, 2023. 1

    2023

  8. [16]

    SCOUTER: slot attention-based classifier for explain- able image recognition

    Liangzhi Li, Bowen Wang, Manisha Verma, Yuta Nakashima, Ryo Kawasaki, and Hajime Nagahara. SCOUTER: slot attention-based classifier for explain- able image recognition. In 2021 IEEE/CVF Interna- tional Conference on Computer Vision, ICCV 2021, Montreal, QC, Canada, October 10-...

    2021

  9. [17]

    Sabernet: Self-attention based effective relation net- work for few-shot learning

    Zijun Li, Zhengping Hu, Weiwei Luo, and Xiao Hu. Sabernet: Self-attention based effective relation net- work for few-shot learning. Pattern Recognit., 133: 109024, 2023. 3

    2023

  10. [18]

    Learning a few-shot embedding model with contrastive learning

    Chen Liu, Yanwei Fu, Chengming Xu, Siqian Yang, Jilin Li, Chengjie Wang, and Li Zhang. Learning a few-shot embedding model with contrastive learning. In Thirty-Fifth AAAI Conference on Artificial Intel- ligence, AAAI 2021, Thirty-Third Conference on In- novative Applications o...

    2021

  11. [19]

    Object-centric learning with slot attention

    Francesco Locatello, Dirk Weissenborn, Thomas Un- terthiner, Aravindh Mahendran, Georg Heigold, Jakob Uszkoreit, Alexey Dosovitskiy, and Thomas Kipf. Object-centric learning with slot attention. In Ad- vances in Neural Information Processing Systems 33: Annual Conference on Ne...

    2020

  12. [20]

    Bi-directional task-guided network for few-shot fine- grained image classification

    Zhen-Xiang Ma, Zhen-Duo Chen, Li-Jun Zhao, Zi- Chao Zhang, Tai Zheng, Xin Luo, and Xin-Shun Xu. Bi-directional task-guided network for few-shot fine- grained image classification. In Proceedings of the 32nd ACM International Conference on Multimedia , page 8277–8286, New York,...

    2024

  13. [21]

    LaFTer: Label-free tuning of zero-shot classifier using language and unlabeled image collections

    Muhammad Jehanzeb Mirza, Leonid Karlinsky, Wei Lin, Horst Possegger, Mateusz Kozinski, Rogerio Feris, and Horst Bischof. LaFTer: Label-free tuning of zero-shot classifier using language and unlabeled image collections. In Thirty-seventh Conference on Neural Information Process...

    2023

  14. [22]

    On first-order meta-learning algorithms

    Alex Nichol, Joshua Achiam, and John Schulman. On first-order meta-learning algorithms. CoRR, abs/1803.02999, 2018. 2

    2018 arXiv

  15. [23]

    Oreshkin, Pau Rodr´ıguez L´opez, and Alexan- dre Lacoste

    Boris N. Oreshkin, Pau Rodr´ıguez L´opez, and Alexan- dre Lacoste. TADAM: task dependent adaptive metric for improved few-shot learning. In Advances in Neu- ral Information Processing Systems 31: Annual Con- ference on Neural Information Processing Systems 2018, NeurIPS 2018, ...

    2018

  16. [24]

    Transductive few-shot classification on the oblique manifold

    Guodong Qi, Huimin Yu, Zhaohui Lu, and Shuzhao Li. Transductive few-shot classification on the oblique manifold. In 2021 IEEE/CVF International Con- ference on Computer Vision, ICCV 2021, Montreal, QC, Canada, October 10-17, 2021, pages 8392–8402. IEEE, 2021. 7

    2021

  17. [25]

    Tenenbaum, Hugo Larochelle, and Richard S

    Mengye Ren, Eleni Triantafillou, Sachin Ravi, Jake Snell, Kevin Swersky, Joshua B. Tenenbaum, Hugo Larochelle, and Richard S. Zemel. Meta-learning for semi-supervised few-shot classification. In 6th In- ternational Conference on Learning Representations, ICLR 2018, Vancouver, ...

    2018

  18. [26]

    Proto- typical networks for few-shot learning

    Jake Snell, Kevin Swersky, and Richard Zemel. Proto- typical networks for few-shot learning. InAdvances in Neural Information Processing Systems. Curran Asso- ciates, Inc., 2017. 2, 7

    2017

  19. [27]

    Label propagation for zero-shot classification with vision-language models

    Vladan Stojni?, Yannis Kalantidis, and Giorgos To- lias. Label propagation for zero-shot classification with vision-language models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 23209–23218, 2024. 1

    2024

  20. [28]

    Meta-transfer learning for few-shot learning

    Qianru Sun, Yaoyao Liu, Tat-Seng Chua, and Bernt Schiele. Meta-transfer learning for few-shot learning. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR) , 2019. 3

    2019

  21. [29]

    Matching networks for one shot learning

    Oriol Vinyals, Charles Blundell, Tim Lillicrap, Koray Kavukcuoglu, and Daan Wierstra. Matching networks for one shot learning. In Advances in Neural Infor- mation Processing Systems 29: Annual Conference on Neural Information Processing Systems 2016, Decem- ber 5-10, 2016, Bar...

    2016

  22. [30]

    Background-filtering feature-enhanced graph neural networks for few-shot learning

    Binbin Wang, Yuemao Wang, and Yaoqun Xu. Background-filtering feature-enhanced graph neural networks for few-shot learning. Applied Sciences, 14 (15), 2024. 3

    2024

  23. [31]

    Focus your attention when few-shot classification

    Haoqing Wang, Shibo Jie, and Zhi-Hong Deng. Focus your attention when few-shot classification. In Thirty- seventh Conference on Neural Information Processing Systems, 2023. 2

    2023

  24. [32]

    Distribu- tional modeling on a diet: One-shot word learning from text only

    Su Wang, Stephen Roller, and Katrin Erk. Distribu- tional modeling on a diet: One-shot word learning from text only. In Proceedings of the Eighth Inter- national Joint Conference on Natural Language Pro- cessing, IJCNLP 2017, Taipei, Taiwan, November 27 - December 1, 2017 - Vo...

    2017

  25. [33]

    Task-aware part mining network for few- shot learning

    Jiamin Wu, Tianzhu Zhang, Yongdong Zhang, and Feng Wu. Task-aware part mining network for few- shot learning. In 2021 IEEE/CVF International Con- ference on Computer Vision, ICCV 2021, Montreal, QC, Canada, October 10-17, 2021, pages 8413–8422. IEEE, 2021. 7

    2021

  26. [34]

    Learning dynamic alignment via meta- filter for few-shot learning

    Chengming Xu, Yanwei Fu, Chen Liu, Chengjie Wang, Jilin Li, Feiyue Huang, Li Zhang, and Xi- angyang Xue. Learning dynamic alignment via meta- filter for few-shot learning. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021 , pag...

    2021

  27. [35]

    A dual attention network with semantic embedding for few-shot learning

    Shipeng Yan, Songyang Zhang, and Xuming He. A dual attention network with semantic embedding for few-shot learning. In The Thirty-Third AAAI Con- ference on Artificial Intelligence, AAAI 2019, The Thirty-First Innovative Applications of Artificial Intel- ligence Conference, IA...

    2019

  28. [36]

    Learning embedding adaptation for few-shot learning

    Han-Jia Ye, Hexiang Hu, De-Chuan Zhan, and Fei Sha. Learning embedding adaptation for few-shot learning. CoRR, abs/1812.03664, 2018. 6

    2018 arXiv

  29. [37]

    Few-shot learning via embedding adaptation with set-to-set functions

    Han-Jia Ye, Hexiang Hu, De-Chuan Zhan, and Fei Sha. Few-shot learning via embedding adaptation with set-to-set functions. In 2020 IEEE/CVF Con- ference on Computer Vision and Pattern Recogni- tion, CVPR 2020, Seattle, WA, USA, June 13-19, 2020, pages 8805–8814. Computer Vision...

    2020

  30. [38]

    Simple semantic-aided few-shot learning

    Hai Zhang, Junzhe Xu, Shanlin Jiang, and Zhenan He. Simple semantic-aided few-shot learning. In IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, CVPR 2024, Seattle, WA, USA, June 16-22, 2024, pages 28588–28597. IEEE, 2024. 1, 3

    2024

  31. [39]

    Hospedales

    Xueting Zhang, Debin Meng, Henry Gouk, and Tim- othy M. Hospedales. Shallow bayesian meta learn- ing for real-world few-shot recognition. In 2021 IEEE/CVF International Conference on Computer Vi- sion, ICCV 2021, Montreal, QC, Canada, October 10- 17, 2021, pages 631–640. IEEE, 2021. 7

    2021

  32. [40]

    Few-shot learning based on prototype rectification with a self-attention mechanism

    Peng Zhao, Liang Wang, Xuyang Zhao, Huiting Liu, and Xia Ji. Few-shot learning based on prototype rectification with a self-attention mechanism. Expert Syst. Appl., 249:123586, 2024. 3

    2024

  33. [41]

    Yuille, and Tao Kong

    Jinghao Zhou, Chen Wei, Huiyu Wang, Wei Shen, Ci- hang Xie, Alan L. Yuille, and Tao Kong. Image BERT pre-training with online tokenizer. In The Tenth In- ternational Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022 . Open- Review.net, 2022. 6

    2022

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.