Pith. sign in

REVIEW 4 major objections 6 minor 53 references

Strategic Base Representation Learning via Feature Augmentations for Few-Shot Class Incremental Learning

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

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

Pith's one-line read This paper shows that mixing two augmented feature vectors to create proxy classes, together with a self-supervised contrastive loss, expands the base-session feature space so that new classes can be added in few-shot increments with less…

desk verdict The SOTA claim is undermined by an evaluation that gives FACL two votes per base class while baselines get one; the idea is a plausible regularizer but the reported margins are not trustworthy. read the letter →

arxiv 2501.09361 v1 pith:C7RKJMQD submitted 2025-01-16 cs.CV

Parinita Nema , Vinod K Kurmi This is my paper · ORCID
classification cs.CV
keywords few-shotclassincrementallearningfeatureaugmentationproxyclassesself-supervisedcontrastivebasesessionrepresentationprototypeevaluationCIFAR100miniImageNet
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 asks how to train the base session in few-shot class incremental learning so that later classes, which arrive with only a few labeled examples, can be added without collapsing into old classes. It proposes Feature Augmentation driven Contrastive Learning (FACL), which blends feature vectors from transformed images, assigns proxy labels to those mixtures, and trains with a contrastive loss to push base-class representations apart. The result, the authors report, is a base feature space with more room for future classes: average accuracy improves by 4.77% on CIFAR100, 0.6% on miniImageNet, and 1.59% on CUB200 over prior state-of-the-art methods, with the largest gains in the base session. The claim is that a deliberately expanded base representation is the key to accommodating few-shot novel classes without retraining the backbone.

What carries the argument

The load-bearing mechanism is the proxy-class feature mixture: $F^i_{\text{aug}} = \delta Z^i_{\text{aug}} + (1-\delta) Z^j_{\text{aug}}$ with $\delta = 0.5$, where $Z_{\text{aug}}$ are features of images subjected to a 180° rotation plus RGB permutation. Each mixture receives a proxy label $y_p = y\times 2 + p$, effectively doubling the label space during base training and creating placeholders for future classes. A MoCo-style self-supervised contrastive loss is applied to two augmented views of $F_{\text{comb}}$, and inference uses cosine-similarity nearest-mean classification over both original and proxy prototypes. These components together are what the paper credits for separating base classes and leaving room for incremental classes.

What would settle it

Run FACL and the strongest baselines under identical evaluation rules: evaluate FACL using only its original-class prototypes (dropping proxy predictions), and evaluate baselines with the same proxy-label aggregation (adding a second prototype per class). If FACL's average-accuracy lead over the best baseline shrinks below the reported margins or reverses, the central claim is not about the representation but about the evaluation protocol.

Watch

Extended reading notes

Core claim

FACL's central claim is that the base session, not the incremental sessions, should do the work of preparing for future classes. The authors generate an augmented feature vector $F^i_{\text{aug}} = \delta Z^i_{\text{aug}} + (1-\delta) Z^j_{\text{aug}}$ by mixing features of two differently transformed images, assign these mixtures proxy labels $y_p = y \times P + p$ (with $P=2$), and train the classifier on both original and proxy labels. A MoCo-style self-supervised contrastive loss on the combined features $F_{\text{comb}} = \mathrm{concat}(Z_{\text{ori}}, F_{\text{aug}})$ then separates base-class representations. At evaluation, the model aggregates original-label and proxy-label predictions under the nearest-mean classifier. The paper reports that this recipe outperforms current state-of-the-art FSCIL methods on all three benchmarks, with base-session accuracy rising to 86.20 on CIFAR100 relative to 79.13 for the closest prior method.

Load-bearing premise

The central claim rests on comparing FACL's aggregated original-plus-proxy predictions against baselines that use a single prototype per class, so the reported gains could be an artifact of the evaluation protocol rather than the learned representation.

Editorial extensions

If this is right

  • If FACL is right, base-session training should be judged not only by its own accuracy but by how much unused feature space it leaves for later classes.
  • Frozen-backbone FSCIL methods can improve without touching the incremental session by changing only how the base representation is shaped.
  • Proxy-class placeholders make the number of base training labels a tunable resource: doubling the label space via $P=2$ is a simple lever that other FSCIL pipelines could adopt.
  • The contrastive loss and feature mixing appear to be complementary: the ablation shows the largest gains only when both are present, suggesting the expanded space and the separation pressure reinforce each other.
  • The reported gains are largest on the base session and on CIFAR100, so the method's benefit may scale with the number of base classes and the difficulty of separating them.

Reading between the lines

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

  • One testable extension would be to apply the same proxy-label evaluation protocol to existing baselines, giving each baseline a second prototype per class, to isolate how much of FACL's gain comes from the representation versus the aggregation rule.
  • The proxy-class idea could transfer to other continual-learning settings, such as task-incremental or online continual learning, where reserving placeholder prototypes might reduce inter-task confusion without rehearsal.
  • The choice of $\delta=0.5$ and only one augmentation type (rotation plus RGB permutation) suggests a cheap optimization surface: sweeping $\delta$ and the augmentation family per dataset could yield further gains, or reveal where the mixture stops being semantically valid.
  • Because the base-session accuracy jumps roughly 11 points when feature augmentation is added, the paper implicitly claims that overfitting to base classes is a major bottleneck; an independent test would be to measure FACL's base-session calibration or feature-space coverage versus baselines.
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

4 major / 6 minor

Summary. The paper proposes FACL (Feature Augmentation driven Contrastive Learning) for few-shot class incremental learning (FSCIL). The method augments feature vectors, assigns proxy labels to the augmented features, and adds a contrastive loss to train the base-session representation, aiming to leave room for future incremental classes. Experiments on CIFAR100, miniImageNet, and CUB200 report average accuracy improvements of 4.77%, 0.6%, and 1.59% over prior state-of-the-art methods, with base-session gains as large as 86.20% on CIFAR100 (vs. 74.77% for plain cross-entropy).

Significance. If the reported gains were reliable, FACL would be a useful contribution to FSCIL by structuring the base-session feature space with feature augmentation and proxy classes. However, the central SOTA claim is undermined by an asymmetric evaluation protocol that credits proxy-label predictions to the original classes only for FACL, an underspecified inference procedure, and the absence of error bars. These issues mean the paper's headline results are not yet supported.

major comments (4)
  1. [3.3.4, Eq. (11)] The evaluation aggregates original-label and proxy-label predictions, so a correct proxy prediction is counted as correct for the original class. Since each base class has P=2 labels under Eq. (5), FACL effectively has two prototypes per base class at test time, while baselines such as CEC, FACT, TEEN, ALICE, MICS, and SAVC are evaluated with one prototype per class. Tables 2–4 therefore do not provide an apples-to-apples comparison. This asymmetry likely explains the large session-0 gain (86.20 vs. 74.77 in Table 5). The authors must re-run the evaluation with a single prototype per class, or apply the same proxy-aggregation rule to all baselines, and report those results.
  2. [3.3.1–3.3.4] The inference procedure is underspecified. Training uses F_comb = concat(Z_ori, F_aug) (Eq. 4) and the classifier loss in Eq. (8) operates on augmented versions F_a of F_comb. However, evaluation Eq. (11) computes cosine similarity between f(x) (the original feature only) and prototypes c_s_y (Eq. 10). It is unclear how the classifier trained on concatenated features is used at test time, how proxy prototypes are constructed for incremental sessions, and how the 'integrated' evaluation combines classifier logits with NCM distances. Without a precise inference algorithm, the reported numbers cannot be reproduced.
  3. [Tables 2–4] No error bars, standard deviations, or number of seeds are reported. The claimed miniImageNet improvement is 0.6% average accuracy, which is within typical run-to-run variance for FSCIL benchmarks. The authors should report mean ± std over at least 3 seeds for their method and the closest baselines (e.g., MICS, SAVC) to support the claim of significant improvement.
  4. [4.4, Table 5] The ablation is confounded with the evaluation protocol. The CE baseline uses one prototype per class, while the full model (CE+SSCL+PC+FA) benefits from proxy-prediction aggregation. The +12.33% final-session gain over CE therefore conflates the effect of the method with the effect of the evaluation rule. The ablation should be performed under a single evaluation protocol, either with proxy aggregation for all rows or without it for the full model.
minor comments (6)
  1. [Table 4] The citation 'MICS [24]' is inconsistent with the reference list, where MICS is [16] (the same table also has a typo in 'SA VC'; it should be 'SAVC' or 'SA-VC').
  2. [2.3, 3.3.3] The paper calls the contrastive loss 'self-supervised' but cites [15] (Supervised Contrastive Learning) and uses MoCo [6]; please clarify which loss is actually used and provide correct references.
  3. [Eq. (7)] The notation in Eq. (7) is under-specified; please define S(j), N(j), za, zk+, and explain how the feature and label queue is used.
  4. [4.3] The t-SNE and confusion-matrix visualizations are qualitative; consider adding quantitative metrics, such as inter-class/intra-class distance or average cosine similarity, to support the claim of improved feature separation.
  5. [Discussion] The sentence 'our approach maintains the same number of proxy classes as the original ones' is confusing because Eq. (5) with P=2 doubles the label space; please rephrase.
  6. [4.1] The statement 'We follow the CEC protocol' should specify whether the same data splits, evaluation code, and hyperparameter selection procedure are used for all baselines.

Circularity Check

1 steps flagged · score 7.0 of 10

The reported SOTA gains rest on an integrated evaluation that counts proxy-label predictions as correct for original labels, giving FACL two chances per base class while baselines get one; the accuracy metric is partly defined by the method's own proxy-label construction.

  1. self definitional [Section 3.3.4 (Integrated Feature Evaluation), with Eq. (5) and P=2 in Section 3.3.2 and Eq. (11) in Section 3.3.4]
    "For evaluation in our approach, aggregate the original label predictions and proxy label predictions to compute the accuracy for the original labels. ... yp = y × P+ p ... Therefore, P = 2, so for each particular label: y = {yp}M p=0."

    Proxy labels are defined as a deterministic re-indexing of original labels: yp = y*P + p with P=2. The evaluation then counts a correct proxy prediction as a correct original-label prediction, so each base class can be correctly classified by either its original prototype or its proxy prototype. This is a definitional expansion of the accuracy metric, not a property of the learned representation alone. Baselines such as CEC, FACT, TEEN, ALICE, MICS, and SAVC are evaluated with standard nearest-mean classification using one prototype per class; the paper does not apply the same dual-vote aggregation to them.

full rationale

The paper's training framework is not circular in itself: feature augmentation, proxy-class generation, and contrastive loss are defined from the data and the stated losses, and the method is compared on standard FSCIL benchmarks. The central circularity is in the evaluation protocol. Section 3.3.4 explicitly aggregates proxy-label predictions with original-label predictions to compute accuracy, while proxy labels are generated from original labels via yp = y*P + p with P=2. This gives FACL two chances per base class at test time (original and proxy prototypes) and the paper does not give baselines the same treatment. The large session-0 gain in Table 5 (86.20 vs 74.77 CE) and the 4.77% average gain on CIFAR100 are therefore partially manufactured by the definition of the accuracy metric, not wholly by the learned representation. No other load-bearing circularity was found: the self-citation [29] is not used to justify the method's core premise, and the method's components are evaluated independently. Because the headline 'outperforms state-of-the-art' claim is substantially tied to this non-comparable, definition-expanding evaluation, the circularity score is 7 rather than a lower score.

Assumptions & free parameters 4 free parameters · 4 assumptions · 1 invented entities

The method depends on several domain assumptions about feature space geometry and on hyperparameters chosen by hand. The largest free parameter is delta=0.5, selected by ablation. Proxy classes are an invented label structure with no external evidence.

free parameters (4)
  • delta = 0.5
    Mixing coefficient in Eq. 3; chosen by ablation in Fig. 5, best at 0.5 across datasets.
  • P = 2
    Number of proxy labels per class in Eq. 5; set to the number of feature versions.
  • tau = not specified
    Temperature in contrastive loss Eq. 7; no value reported in the paper.
  • M = 1
    Number of image transformations; one augmented version per image using 180 degree rotation plus RGB permutation.
assumptions (4)
  • domain assumption Mixing augmented image features yields useful proxies for future classes.
    Core premise of Section 3.3.1; no theoretical justification, relies on empirical validation.
  • domain assumption Concatenating original and augmented features (Eq. 4) is a valid classifier input and preserves semantic information.
    Required for the integrated classifier; unstated and untested.
  • domain assumption Contrastive loss with instance-level positives improves base class separation.
    Borrowed from MoCo and SupCon; assumed to transfer to FSCIL.
  • domain assumption The feature extractor can be frozen after base training, per CEC protocol.
    Standard FSCIL protocol; not justified in this paper but accepted from prior work.
invented entities (1)
  • Proxy label space
    purpose: Expand the label space with placeholder classes to reserve room for future classes.
    Defined in Section 3.3.2; only exists inside the model and has no external falsifiable handle.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Strategic Base Representation Learning via Feature Augmentations for Few-Shot Class Incremental Learning." pith.science (2026). https://pith.science/paper/C7RKJMQD

@misc{pith2026250109361,
  author       = {Pith},
  title        = {Pith review of: Strategic Base Representation Learning via Feature Augmentations for Few-Shot Class Incremental Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C7RKJMQD}},
  note         = {Machine review of arXiv:2501.09361}
}
read the original abstract

Few-shot class incremental learning implies the model to learn new classes while retaining knowledge of previously learned classes with a small number of training instances. Existing frameworks typically freeze the parameters of the previously learned classes during the incorporation of new classes. However, this approach often results in suboptimal class separation of previously learned classes, leading to overlap between old and new classes. Consequently, the performance of old classes degrades on new classes. To address these challenges, we propose a novel feature augmentation driven contrastive learning framework designed to enhance the separation of previously learned classes to accommodate new classes. Our approach involves augmenting feature vectors and assigning proxy labels to these vectors. This strategy expands the feature space, ensuring seamless integration of new classes within the expanded space. Additionally, we employ a self-supervised contrastive loss to improve the separation between previous classes. We validate our framework through experiments on three FSCIL benchmark datasets: CIFAR100, miniImageNet, and CUB200. The results demonstrate that our Feature Augmentation driven Contrastive Learning framework significantly outperforms other approaches, achieving state-of-the-art performance.

Figures

Figures reproduced from arXiv: 2501.09361 by the authors.

Figure 1
Figure 1. The illustration of the motivation for our framework. Our [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The illustration of the proposed FACL framework. FACL consists of three main components: (a) feature augmentation, where [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Comparison of CE and FACL Confusion Matrices on the [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Accuracy values obtained during the final session for [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 37 canonical work pages

  1. [1]

    Il2m: Class in- cremental learning with dual memory

    Eden Belouadah and Adrian Popescu. Il2m: Class in- cremental learning with dual memory. In Proceedings of the IEEE/CVF international conference on computer vision, pages 583–592, 2019. 2

    work page 2019

  2. [2]

    End-to-end incre- mental learning

    Francisco M Castro, Manuel J Mar ´ın-Jim´enez, Nicol´as Guil, Cordelia Schmid, and Karteek Alahari. End-to-end incre- mental learning. In Proceedings of the European conference on computer vision (ECCV), pages 233–248, 2018. 6, 7

    work page 2018

  3. [3]

    Continual learning in low-rank orthogonal sub- spaces

    Arslan Chaudhry, Naeemullah Khan, Puneet Dokania, and Philip Torr. Continual learning in low-rank orthogonal sub- spaces. Advances in Neural Information Processing Systems, 33:9900–9911, 2020. 2

    work page 2020

  4. [4]

    Incremental few-shot learn- ing via vector quantization in deep embedded space

    Kuilin Chen and Chi-Guhn Lee. Incremental few-shot learn- ing via vector quantization in deep embedded space. In In- ternational Conference on Learning Representations , 2020. 1, 7

    work page 2020

  5. [5]

    Sharpness-aware gradient guidance for few-shot class-incremental learning

    Runhang Chen, Xiao-Yuan Jing, Fei Wu, and Haowen Chen. Sharpness-aware gradient guidance for few-shot class-incremental learning. Knowledge-Based Systems, page 112030, 2024. 6, 7

    work page 2024

  6. [6]

    Improved baselines with momentum contrastive learning

    Xinlei Chen, Haoqi Fan, Ross Girshick, and Kaiming He. Improved baselines with momentum contrastive learning. arXiv preprint arXiv:2003.04297, 2020. 5

    arXiv 2003

  7. [7]

    Metafscil: A meta-learning approach for few-shot class incremental learning

    Zhixiang Chi, Li Gu, Huan Liu, Yang Wang, Yuanhao Yu, and Jin Tang. Metafscil: A meta-learning approach for few-shot class incremental learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 14166–14175, 2022. 6, 7

    work page 2022

  8. [8]

    Few-shot class- incremental learning via relation knowledge distillation

    Songlin Dong, Xiaopeng Hong, Xiaoyu Tao, Xinyuan Chang, Xing Wei, and Yihong Gong. Few-shot class- incremental learning via relation knowledge distillation. In Proceedings of the AAAI Conference on Artificial Intelli- gence, volume 35, pages 1255–1263, 2021. 1

    work page 2021

Show all 53 references
  1. [9]

    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 International conference on machine learning , pages 1126–1135. PMLR, 2017. 2

    2017

  2. [10]

    Decision boundary optimization for few-shot class-incremental learn- ing

    Chenxu Guo, Qi Zhao, Shuchang Lyu, Binghao Liu, Chun- lei Wang, Lijiang Chen, and Guangliang Cheng. Decision boundary optimization for few-shot class-incremental learn- ing. In Proceedings of the IEEE/CVF International Confer- ence on Computer Vision, pages 3501–3511, 2023. 8

    2023

  3. [11]

    Momentum contrast for unsupervised visual rep- resentation learning

    Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xie, and Ross Girshick. Momentum contrast for unsupervised visual rep- resentation learning. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 9729–9738, 2020. 5

    2020

  4. [12]

    Mask r-cnn

    Kaiming He, Georgia Gkioxari, Piotr Doll ´ar, and Ross Gir- shick. Mask r-cnn. In Proceedings of the IEEE international conference on computer vision, pages 2961–2969, 2017. 1

    2017

  5. [13]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 6

    2016

  6. [14]

    Learning a unified classifier incrementally via rebalancing

    Saihui Hou, Xinyu Pan, Chen Change Loy, Zilei Wang, and Dahua Lin. Learning a unified classifier incrementally via rebalancing. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 831–839,

  7. [15]

    Supervised contrastive learning

    Prannay Khosla, Piotr Teterwak, Chen Wang, Aaron Sarna, Yonglong Tian, Phillip Isola, Aaron Maschinot, Ce Liu, and Dilip Krishnan. Supervised contrastive learning. Advances in neural information processing systems, 33:18661–18673,

  8. [16]

    Mics: Midpoint interpolation to learn compact and separated representations for few-shot class-incremental learning

    Solang Kim, Yuho Jeong, Joon Sung Park, and Sung Whan Yoon. Mics: Midpoint interpolation to learn compact and separated representations for few-shot class-incremental learning. In Proceedings of the IEEE/CVF Winter Confer- ence on Applications of Computer Vision, pages 2236–2245,

  9. [17]

    Learning multiple layers of features from tiny images

    Alex Krizhevsky, Geoffrey Hinton, et al. Learning multiple layers of features from tiny images. 2009. 6

    2009

  10. [18]

    Do not forget to attend to uncer- tainty while mitigating catastrophic forgetting

    Vinod K Kurmi, Badri N Patro, Venkatesh K Subramanian, and Vinay P Namboodiri. Do not forget to attend to uncer- tainty while mitigating catastrophic forgetting. In Proceed- ings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 736–745, 2021. 1

    2021

  11. [19]

    Learning without forgetting

    Zhizhong Li and Derek Hoiem. Learning without forgetting. IEEE transactions on pattern analysis and machine intelli- gence, 40(12):2935–2947, 2017. 1, 2

    2017

  12. [20]

    Rotate your networks: Better weight consolidation and less catastrophic forgetting

    Xialei Liu, Marc Masana, Luis Herranz, Joost Van de Weijer, Antonio M Lopez, and Andrew D Bagdanov. Rotate your networks: Better weight consolidation and less catastrophic forgetting. In 2018 24th International Conference on Pattern Recognition (ICPR), pages 2262–2268. IEEE, 2018. 2

    2018

  13. [21]

    Spectral normalization for generative ad- versarial networks

    Takeru Miyato, Toshiki Kataoka, Masanori Koyama, and Yuichi Yoshida. Spectral normalization for generative ad- versarial networks. arXiv preprint arXiv:1802.05957, 2018. 1

    2018 arXiv

  14. [22]

    On first-order meta-learning algorithms

    Alex Nichol, Joshua Achiam, and John Schulman. On first-order meta-learning algorithms. arXiv preprint arXiv:1803.02999, 2018. 2

    2018 arXiv

  15. [23]

    Tadam: Task dependent adaptive metric for improved few-shot learning

    Boris Oreshkin, Pau Rodr ´ıguez L ´opez, and Alexandre La- coste. Tadam: Task dependent adaptive metric for improved few-shot learning. Advances in neural information process- ing systems, 31, 2018. 2

    2018

  16. [24]

    Few-shot class-incremental learning from an open- set perspective

    Can Peng, Kun Zhao, Tianren Wang, Meng Li, and Brian C Lovell. Few-shot class-incremental learning from an open- set perspective. In European Conference on Computer Vi- sion. Springer, 2022. 2, 6, 7, 8

    2022

  17. [25]

    Low-shot learning with imprinted weights

    Hang Qi, Matthew Brown, and David G Lowe. Low-shot learning with imprinted weights. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 5822–5830, 2018. 2

    2018

  18. [26]

    icarl: Incremental classifier and representation learning

    Sylvestre-Alvise Rebuffi, Alexander Kolesnikov, Georg Sperl, and Christoph H Lampert. icarl: Incremental classifier and representation learning. In Proceedings of the IEEE con- ference on Computer Vision and Pattern Recognition, pages 2001–2010, 2017. 1, 2, 6, 7

    2001

  19. [27]

    Imagenet large scale visual recognition challenge

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, San- jeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, et al. Imagenet large scale visual recognition challenge. International journal of computer vision, 115:211–252, 2015. 6

    2015

  20. [28]

    Overcoming catastrophic for- getting in incremental few-shot learning by finding flat min- ima

    Guangyuan Shi, Jiaxin Chen, Wenlong Zhang, Li-Ming Zhan, and Xiao-Ming Wu. Overcoming catastrophic for- getting in incremental few-shot learning by finding flat min- ima. Advances in neural information processing systems , 34:6747–6761, 2021. 1, 3, 6, 7

    2021

  21. [29]

    To- wards robust few-shot class incremental learning in audio classification using contrastive representation

    Riyansha Singh, Parinita Nema, and Vinod K Kurmi. To- wards robust few-shot class incremental learning in audio classification using contrastive representation. In Proc. In- terspeech 2024, pages 5023–5027, 2024. 1

    2024

  22. [30]

    Prototypical networks for few-shot learning

    Jake Snell, Kevin Swersky, and Richard Zemel. Prototypical networks for few-shot learning. Advances in neural informa- tion processing systems, 30, 2017. 2

    2017

  23. [31]

    Learning with fantasy: Semantic-aware virtual contrastive constraint for few-shot class-incremental learning

    Zeyin Song, Yifan Zhao, Yujun Shi, Peixi Peng, Li Yuan, and Yonghong Tian. Learning with fantasy: Semantic-aware virtual contrastive constraint for few-shot class-incremental learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages...

    2023

  24. [32]

    Few-shot class- incremental learning

    Xiaoyu Tao, Xiaopeng Hong, Xinyuan Chang, Songlin Dong, Xing Wei, and Yihong Gong. Few-shot class- incremental learning. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 12183–12192, 2020. 1, 2, 6, 7

    2020

  25. [33]

    Visualiz- ing data using t-sne

    Laurens Van der Maaten and Geoffrey Hinton. Visualiz- ing data using t-sne. Journal of machine learning research, 9(11), 2008. 7

    2008

  26. [34]

    Manifold mixup: Better representations by interpolating hidden states

    Vikas Verma, Alex Lamb, Christopher Beckham, Amir Na- jafi, Ioannis Mitliagkas, David Lopez-Paz, and Yoshua Ben- gio. Manifold mixup: Better representations by interpolating hidden states. In International conference on machine learn- ing, pages 6438–6447. PMLR, 2019. 8

    2019

  27. [35]

    Matching networks for one shot learning.Ad- vances in neural information processing systems , 29, 2016

    Oriol Vinyals, Charles Blundell, Timothy Lillicrap, Daan Wierstra, et al. Matching networks for one shot learning.Ad- vances in neural information processing systems , 29, 2016. 2

    2016

  28. [36]

    The caltech-ucsd birds-200-2011 dataset

    Catherine Wah, Steve Branson, Peter Welinder, Pietro Per- ona, and Serge Belongie. The caltech-ucsd birds-200-2011 dataset. 2011. 6

    2011

  29. [37]

    Few-shot class-incremental learning via training-free prototype calibration

    Qi-Wei Wang, Da-Wei Zhou, Yi-Kai Zhang, De-Chuan Zhan, and Han-Jia Ye. Few-shot class-incremental learning via training-free prototype calibration. Advances in Neural Information Processing Systems, 36, 2024. 3, 6, 7

    2024

  30. [38]

    Batchensemble: an alternative approach to efficient ensemble and lifelong learning

    Yeming Wen, Dustin Tran, and Jimmy Ba. Batchensemble: an alternative approach to efficient ensemble and lifelong learning. arXiv preprint arXiv:2002.06715, 2020. 2

    2002 arXiv

  31. [39]

    Tkil: Tangent kernel optimization for class balanced incremental learning

    Jinlin Xiang and Eli Shlizerman. Tkil: Tangent kernel optimization for class balanced incremental learning. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3529–3539, 2023. 2

    2023

  32. [40]

    Der: Dy- namically expandable representation for class incremental learning

    Shipeng Yan, Jiangwei Xie, and Xuming He. Der: Dy- namically expandable representation for class incremental learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 3014–3023,

  33. [41]

    Neural collapse inspired feature- classifier alignment for few-shot class incremental learning

    Yibo Yang, Haobo Yuan, Xiangtai Li, Zhouchen Lin, Philip Torr, and Dacheng Tao. Neural collapse inspired feature- classifier alignment for few-shot class incremental learning. arXiv preprint arXiv:2302.03004, 2023. 2

    2023 arXiv

  34. [42]

    Incrementally learned an- gular representations for few-shot class-incremental learn- ing

    In-Ug Yoon and Jong-Hwan Kim. Incrementally learned an- gular representations for few-shot class-incremental learn- ing. IEEE Access, 2023. 2, 6

    2023

  35. [43]

    Cutmix: Regu- larization strategy to train strong classifiers with localizable features

    Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo. Cutmix: Regu- larization strategy to train strong classifiers with localizable features. In Proceedings of the IEEE/CVF international con- ference on computer vision, pages 6023–6032, 2019. 8

    2019

  36. [44]

    Few-shot incremental learning with contin- ually evolved classifiers

    Chi Zhang, Nan Song, Guosheng Lin, Yun Zheng, Pan Pan, and Yinghui Xu. Few-shot incremental learning with contin- ually evolved classifiers. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 12455–12464, 2021. 1, 4, 6, 7

    2021

  37. [45]

    mixup: Beyond empirical risk minimization

    Hongyi Zhang. mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412, 2017. 8

    2017 arXiv

  38. [46]

    Mgsvf: Multi-grained slow versus fast framework for few-shot class-incremental learning

    Hanbin Zhao, Yongjian Fu, Mintong Kang, Qi Tian, Fei Wu, and Xi Li. Mgsvf: Multi-grained slow versus fast framework for few-shot class-incremental learning. IEEE Transactions on Pattern Analysis and Machine Intelligence , 46(3):1576– 1588, 2021. 1

    2021

  39. [47]

    Few-shot class- incremental learning via class-aware bilateral distillation

    Linglan Zhao, Jing Lu, Yunlu Xu, Zhanzhan Cheng, Dashan Guo, Yi Niu, and Xiangzhong Fang. Few-shot class- incremental learning via class-aware bilateral distillation. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 11838–11847, 2023. 1, 3

    2023

  40. [48]

    Forward compatible few-shot class-incremental learning

    Da-Wei Zhou, Fu-Yun Wang, Han-Jia Ye, Liang Ma, Shil- iang Pu, and De-Chuan Zhan. Forward compatible few-shot class-incremental learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 9046–9056, 2022. 1, 2, 3, 6, 7

    2022

  41. [49]

    Few-shot class-incremental learning by sampling multi-phase tasks

    Da-Wei Zhou, Han-Jia Ye, Liang Ma, Di Xie, Shiliang Pu, and De-Chuan Zhan. Few-shot class-incremental learning by sampling multi-phase tasks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022. 6, 7

    2022

  42. [50]

    Class-incremental learning via dual augmentation

    Fei Zhu, Zhen Cheng, Xu-Yao Zhang, and Cheng-lin Liu. Class-incremental learning via dual augmentation. Advances in Neural Information Processing Systems, 34:14306–14318,

  43. [51]

    Prototype augmentation and self-supervision for incremental learning

    Fei Zhu, Xu-Yao Zhang, Chuang Wang, Fei Yin, and Cheng- Lin Liu. Prototype augmentation and self-supervision for incremental learning. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 5871–5880, 2021. 2

    2021

  44. [52]

    Self-promoted prototype refinement for few-shot class- incremental learning

    Kai Zhu, Yang Cao, Wei Zhai, Jie Cheng, and Zheng-Jun Zha. Self-promoted prototype refinement for few-shot class- incremental learning. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 6801–6810, 2021. 1, 6, 7

    2021

  45. [53]

    Enhanced few-shot class-incremental learn- ing via ensemble models

    Mingli Zhu, Zihao Zhu, Sihong Chen, Chen Chen, and Baoyuan Wu. Enhanced few-shot class-incremental learn- ing via ensemble models. arXiv preprint arXiv:2401.07208,

    arXiv

Pith tools

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