Pith. sign in

REVIEW 4 major objections 6 minor 48 references

Adaptive Part Learning for Fine-Grained Generalized Category Discovery: A Plug-and-Play Enhancement

T0 review · 4 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 This paper claims that replacing the global CLS token with pooled, adaptively learned part features resolves the discriminability-versus-generalization trade-off in fine-grained generalized category discovery, and supports this with…

desk verdict A solid plug-and-play part-level enhancement for GCD with plausible gains, but reproducibility gaps and a vague negative-sampling rule keep the central claim from being fully verified. read the letter →

arxiv 2507.06928 v1 pith:LVPS7O6L submitted 2025-07-09 cs.CV

Qiyuan Dai , Hanzhuo Huang , Yu Wu , Sibei Yang This is my paper · ORCID
classification cs.CV
keywords generalizedcategorydiscoveryfine-grainedimagerecognitionobjectpartall-mincontrastivelearningqueriesDINOpriorsplug-and-playrepresentationself-supervisedvisiontransformers
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

Generalized Category Discovery (GCD) asks a model to classify images from both known and novel classes using only a small labeled subset, and existing methods lean on a self-supervised DINO transformer whose global CLS token feature must balance telling similar classes apart against transferring knowledge to new classes. This paper claims that the CLS token forces a trade-off: pushing global features apart to separate similar species also pushes their shared local parts apart, hurting transfer, while superclass-level sharing blurs novel classes into similar known ones. The proposed Adaptive Part Learning (APL) discovers object parts without annotations, using shared learnable part queries steered by DINO attention priors, and learns part features with an all-min contrastive loss that pulls all corresponding parts of positive pairs together but pushes apart only the least similar corresponding part of negative pairs. Pooled part features then replace the CLS token inside existing GCD frameworks. If the claim holds, a plug-in part-level representation improves both known/novel separation and knowledge transfer on fine-grained benchmarks without retraining or re-architecting the host method.

What carries the argument

The load-bearing object is the set of shared learnable part queries $Q \in \mathbb{R}^{T \times C}$, which act as bridges that give the same index to corresponding parts in different images. The DINO prior $F_{\text{prior}}$ from thresholded attention-head features (Eq. 3) steers these queries; the Gumbel-Softmax hard assignment $\mathbf{H}_{\text{part}}$ (Eq. 4) converts patch-query similarities into binary part masks while preserving gradients; the all-min contrastive loss (Eq. 5) performs the adaptive selection of discriminative parts; and the diversity loss (Eq. 6) prevents query collapse. Together these turn an image into $T$ corresponding parts whose pooled features replace the CLS token.

What would settle it

Take a set of fine-grained images, apply strong random cropping or occlusion, and compare APL's hard assignment $\mathbf{H}_{\text{part}}$ of the same image across augmentations: if a large fraction of patches switch part queries, the claimed consistency and correspondence are not present. The same test can be run as an ablation over the threshold $\epsilon$ in Eq. (3) and the query count $T$; if no setting both stabilizes assignments and preserves accuracy, the part prior is not carrying the gain.

Watch

Extended reading notes

Core claim

The paper's central claim is that object parts, rather than global image semantics, are the right unit for GCD: discriminative parts such as the head separate similar species, while shared parts such as legs keep common structure transferable. Concretely, the method extracts DINO part priors by thresholding the final-layer attention maps, transforms a set of $T$ shared learnable part queries into image-specific queries by cross-attention, hard-assigns image patches to queries through a differentiable Gumbel-Softmax step, and treats the patches assigned to each query as one consistent part. An all-min contrastive loss (Eq. 5) makes all corresponding parts of positive pairs similar and only the least-similar corresponding part of negative pairs dissimilar, so discriminative parts are sharpened while the rest stay shareable; a diversity loss (Eq. 6) keeps queries from collapsing. The pooled part feature replaces the CLS token in SimGCD, SPTNet, and CMS, yielding average gains of 3.7, 2.9, and 3.7 accuracy points across four fine-grained datasets, with larger gains generally on novel classes.

Load-bearing premise

The method depends on DINO's attention maps pointing at coherent, reusable object parts, and on the shared queries keeping those parts aligned across different images even when parts are occluded or cropped out; if that alignment fails, the hard part assignments are noise and the all-min loss is constraining random clusters.

Editorial extensions

If this is right

  • Plugging APL into SimGCD raises average accuracy by 3.7 points across CUB-200, Stanford Cars, FGVC-Aircraft, and Herbarium-19, with the largest single gain of 6.3 points on Stanford Cars.
  • Novel-class accuracy improves more than known-class accuracy in most fine-grained settings, for example +6.2 novel versus +5.7 known on Stanford Cars with SimGCD, which is the signature of better knowledge transfer from known to novel classes.
  • Because only the pooled part feature is swapped in, the same APL module attaches to parametric frameworks (SimGCD, SPTNet) and to the mean-shift framework (CMS) without changing their host losses beyond adding the all-min and diversity losses.
  • On generic datasets the gains shrink, for example +0.7 novel points on ImageNet-100, consistent with the claim that part sharing matters most when classes are visually similar and share components.

Reading between the lines

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

  • If the all-min mechanism is the real driver, the same 'hardest corresponding part only' negative constraint could transfer to other part- or prototype-based fine-grained recognition methods, which typically push all negative parts apart.
  • The paper does not measure part consistency quantitatively; a natural extension is to test APL's hard assignments across augmented or occluded views and check whether accuracy tracks the assignment-stability rate.
  • Because DINO part attention is treated as a frozen prior, a domain shift away from natural images could break the part prior; per-domain adaptation of the queries or the threshold $\epsilon$ in Eq. (3) is a concrete testable extension.
  • The interpretability claim could be validated directly by evaluating the discovered part masks against human part annotations, such as CUB part keypoints, even though training is unsupervised; the paper's t-SNE visualizations suggest semantic parts but do not quantify alignment.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes APL, a plug-and-play part-learning module for Generalized Category Discovery (GCD). It uses DINO's final-layer attention heads as a part prior, a set of shared learnable part queries to obtain consistent image-specific part features, and a hard-assignment mechanism with Gumbel-Softmax to decompose each image into parts. The part features replace the CLS token representation in existing GCD frameworks. A new "all-min" contrastive loss is introduced to constrain only the least similar part of negative pairs while keeping all corresponding parts of positive pairs similar, and a diversity loss prevents part-query collapse. Experiments on four fine-grained benchmarks and three generic datasets report consistent improvements over SimGCD, SPTNet, and CMS, with ablations attributing the gains to learnable queries, hard assignment, and the all-min loss.

Significance. If the method is correctly specified, APL addresses a real limitation of global-CLS-token GCD methods and offers a practically valuable plug-and-play enhancement. The empirical evidence is coherent and consistent: Table 1 shows gains across all four fine-grained datasets for three different host frameworks, and Table 4 attributes the improvement to the proposed all-min loss rather than to the host GCD loss alone. The idea of sharing generalizable parts while highlighting discriminative parts is well motivated. However, the implementation details of the central equations and the unlabeled negative-sampling rule are not yet precise enough to validate the claimed mechanism or to reproduce the experiments.

major comments (4)
  1. [§3.2, Eq. (3)] Equation (3) is dimensionally inconsistent as written. Fprior is defined as [ϵ(A1)*Fpatch, ..., ϵ(AM)*Fpatch], so it has shape M×C, and Q is T×C. The expression ProjQ(Q) * ProjK(Fpart)^⊤ / √C then yields a T×M matrix if Fpart=Fprior, or a T×N matrix if Fpart=Fpatch, not the claimed T×C image-specific query QI. In addition, the symbol Fpart is undefined. Since this equation is the foundation of the part-query construction, the authors should give the correct formula with explicit tensor shapes.
  2. [§3.2, Eq. (4)] The hard-assignment mechanism in Eq. (4) selects a single patch for each query: one_hot(arg max_N(...)) returns exactly one nonzero entry per query in the forward pass, so each part feature Ppart equals the feature of one patch token. This contradicts the text's description that "patches within the same query constitute a potential object part" and the visualization in Figure 3c, which shows multi-patch regions. Unless a soft weighted aggregation is intended, the method does not discover multi-patch object parts as claimed. Please clarify whether each part is a single patch or a region, and if a region, specify how the binary assignment is formed.
  3. [§3.3, Eq. (5)] The negative-sampling rule for unlabeled images is underspecified. The text states that for unlabeled Ia the authors "randomly sample some images from B as negative samples, ensuring that not all parts of these images are similar to each corresponding part of Ia, and use the similarity of all positive pairs in the batch as the threshold for sampling negative images." This is not a precise algorithm: it does not define the threshold mathematically, does not state how many negatives are sampled, does not specify how to handle cases where no image passes the threshold, and does not state whether augmentations of the same image are excluded. Because Table 4 attributes a gain of about 2.4–2.7 points to the all-min loss over its alternatives, this missing specification is load-bearing for the paper's central claim and must be provided.
  4. [§3.2–§3.3, Tables 3–4] The paper reports no sensitivity analysis over the key hyperparameters of the proposed module: the number of part queries T, the threshold filter ϵ in Eq. (3), the Gumbel temperature τ in Eq. (4), and the negative-sampling threshold in Eq. (5). The authors acknowledge that DINO's part perception "is too sensitive to variations among different images" and claim that the learnable queries stabilize it, but no quantitative consistency or correspondence metric is reported, and no ablation shows how performance varies with T or ϵ. Without these, the central mechanism of adaptive part discovery is asserted rather than directly demonstrated.
minor comments (6)
  1. [§4.2, Table 1] The main results are reported for a single run; please add standard deviations or at least a multi-seed evaluation to assess the significance of the reported improvements.
  2. [§4.3, Table 4] The comparison method "min-max contrastive loss" in row 3 is never defined in the main text; provide its equation or a precise reference so the ablation is interpretable.
  3. [§3.3, Eq. (6)] In the diversity loss, the use of max(0, sim(Pt, P̂t)) means that negative cosine similarities do not contribute; please clarify whether this is intended and whether the part features are normalized before computing the cosine similarity.
  4. [References] References [10] and [11] are the same paper (Dosovitskiy et al.); merge them into one entry to avoid duplication.
  5. [Throughout] There are several typographical errors, including "unlabled" in §3.1, "created auklet" in §1, and "gneralized" in §4.2; these should be corrected.
  6. [Appendix] The manuscript refers to an Appendix for dataset details and additional ablations, but no appendix is included in the submitted text; please ensure the appendix is present and that the main text is self-contained regarding the negative-sampling algorithm.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: APL is benchmarked against external GCD frameworks on standard held-out splits, and the reported gains do not reduce to any fitted or self-cited quantity.

full rationale

The paper's derivation chain is self-contained relative to external benchmarks. APL augments existing GCD frameworks (SimGCD, SPTNet, CMS) by discovering object parts via DINO priors and shared learnable queries, then replacing the CLS-token feature with averaged part features (Sec. 3.2, Sec. 3.4). The part-learning objectives in Eq. 5 and Eq. 6 are trained on the same labeled/unlabeled GCD splits used by the host frameworks, and evaluation is performed on standard benchmarks (SSB, Herbarium-19, CIFAR, ImageNet-100) using published comparison numbers. The reported improvements are measured against external methods, not against any quantity fitted to the test labels. The only self-referential element is the negative-sampling heuristic in Sec. 3.3, where the threshold for unlabeled negatives is derived from in-batch positive similarities; this calibrates the contrastive loss but does not define the evaluation target or manufacture the reported accuracy. Even though this sampling rule is underspecified and could be a correctness or reproducibility risk, it is not a circular reduction of the central claim. There are no apparent author self-citations, no imported uniqueness theorems, and no fitted parameter renamed as a prediction. The central claim therefore has independent content verified against external benchmarks.

Assumptions & free parameters 5 free parameters · 5 assumptions · 1 invented entities

The ledger shows a typical empirical-methods profile: no physics-style invented entities, five hand-chosen hyperparameters deferred to an appendix, and five assumptions, of which the DINO part-prior assumption and the negative-sampling heuristic are load-bearing. The part queries are internal parameters whose only evidence is the reported benchmarks. The central claim does not depend on any fitted free parameter in the physics sense, but the magnitude of the reported gains depends on untested choices of T, tau, epsilon, and the negative-sampling threshold.

free parameters (5)
  • Number of part queries T
    Hand-chosen granularity of the part decomposition; determines how many parts each image is hard-assigned to and the capacity for cross-image correspondence. Value deferred to the Appendix.
  • Gumbel-Softmax temperature tau (Eq. 4)
    Controls the sharpness of the hard-assignment approximation; affects assignment stability and gradient quality. Not specified in the main text.
  • Attention threshold filter epsilon (Eq. 3)
    Binarizes DINO attention maps into part priors; load-bearing for whether the part priors are semantically coherent. Not specified in the main text.
  • Negative-sample similarity threshold (Eq. 5)
    Unlabeled negatives are sampled using positive-pair similarities as the threshold; the rule governs which negatives the min-over-parts term acts on and is only heuristically described.
  • Loss balance factor lambda (Eq. 8)
    Inherited from SimGCD as the balance between labeled and unlabeled losses; kept from the host framework rather than newly introduced.
assumptions (5)
  • domain assumption DINO final-layer MSA attention heads localize semantically meaningful object parts
    Invoked in Sec 3.1 to 3.2 to build the part prior Fprior in Eq. 3. If false, part discovery inherits noise regardless of the learnable queries.
  • domain assumption Hard assignment of patches to T shared queries yields consistent, corresponding parts across images, with missing parts absorbed by degenerate queries
    The consistency and correspondence claims in Sec 3.2 depend on this. No quantitative consistency metric is reported to verify it.
  • standard math Gumbel-Softmax straight-through estimator is a faithful differentiable surrogate of the hard assignment
    Standard technique used in Eq. 4; accepted background from [13, 26].
  • domain assumption Cluster accuracy with Hungarian matching on the unlabeled set faithfully measures GCD performance
    Standard protocol inherited from [40, 43] and used in Sec 4.1; it assumes predicted clusters align with true classes.
  • domain assumption Positive-pair-similarity thresholding yields informative negative samples for the all-min loss
    The negative sampling rule in Eq. 5 is load-bearing for the min-over-parts term and is only heuristically described in Sec 3.3.
invented entities (1)
  • Shared learnable part queries Q in R^(T x C)
    purpose: Bridge for one-to-one part correspondence across images and seed for object-part discovery
    Internal model parameters with no falsifiable handle outside the paper; their validity is evidenced only by the benchmark numbers reported in the paper, so they carry the same evidentiary weight as any unverified internal mechanism.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Adaptive Part Learning for Fine-Grained Generalized Category Discovery: A Plug-and-Play Enhancement." pith.science (2026). https://pith.science/paper/LVPS7O6L

@misc{pith2026250706928,
  author       = {Pith},
  title        = {Pith review of: Adaptive Part Learning for Fine-Grained Generalized Category Discovery: A Plug-and-Play Enhancement},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LVPS7O6L}},
  note         = {Machine review of arXiv:2507.06928}
}
read the original abstract

Generalized Category Discovery (GCD) aims to recognize unlabeled images from known and novel classes by distinguishing novel classes from known ones, while also transferring knowledge from another set of labeled images with known classes. Existing GCD methods rely on self-supervised vision transformers such as DINO for representation learning. However, focusing solely on the global representation of the DINO CLS token introduces an inherent trade-off between discriminability and generalization. In this paper, we introduce an adaptive part discovery and learning method, called APL, which generates consistent object parts and their correspondences across different similar images using a set of shared learnable part queries and DINO part priors, without requiring any additional annotations. More importantly, we propose a novel all-min contrastive loss to learn discriminative yet generalizable part representation, which adaptively highlights discriminative object parts to distinguish similar categories for enhanced discriminability while simultaneously sharing other parts to facilitate knowledge transfer for improved generalization. Our APL can easily be incorporated into different GCD frameworks by replacing their CLS token feature with our part representations, showing significant enhancements on fine-grained datasets.

Figures

Figures reproduced from arXiv: 2507.06928 by the authors.

Figure 1
Figure 1. Comparison of discriminability and generalization between our and existing GCD methods that use global [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Part discovery by (a) K-means, (b) DINO priors, and (c) [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. t-SNE visualization of part features on CUB-200 and [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: The similarity distribution of discriminative parts be [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 28 canonical work pages

  1. [1]

    PDiscoFormer: Relaxing Part Discovery Constraints with Vision Transformers

    Ananthu Aniraj, Cassio F Dantas, Dino Ienco, and Diego Marcos. Pdiscoformer: Relaxing part discovery constraints with vision transformers. arXiv preprint arXiv:2407.04538,

    work page Pith review arXiv

  2. [2]

    k-means++: The advantages of careful seeding

    David Arthur, Sergei Vassilvitskii, et al. k-means++: The advantages of careful seeding. In Soda, pages 1027–1035,

    work page

  3. [3]

    Open-world semi-supervised learning

    Kaidi Cao, Maria Brbic, and Jure Leskovec. Open-world semi-supervised learning. arXiv preprint arXiv:2102.03526,

    arXiv

  4. [4]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. In Pro- ceedings of the IEEE/CVF international conference on com- puter vision, pages 9650–9660, 2021. 1, 3, 4

    work page 2021

  5. [5]

    This looks like that: deep learn- ing for interpretable image recognition

    Chaofan Chen, Oscar Li, Daniel Tao, Alina Barnett, Cynthia Rudin, and Jonathan K Su. This looks like that: deep learn- ing for interpretable image recognition. Advances in neural information processing systems, 32, 2019. 3

    work page 2019

  6. [6]

    Parametric information max- imization for generalized category discovery

    Florent Chiaroni, Jose Dolz, Ziko Imtiaz Masud, Amar Mitiche, and Ismail Ben Ayed. Parametric information max- imization for generalized category discovery. InProceedings of the IEEE/CVF International Conference on Computer Vi- sion, pages 1729–1739, 2023. 7

    work page 2023

  7. [7]

    Contrastive Mean-Shift Learning for Generalized Category Discovery

    Sua Choi, Dahyun Kang, and Minsu Cho. Contrastive mean- shift learning for generalized category discovery. arXiv preprint arXiv:2404.09451, 2024. 3, 6, 7

    work page Pith review arXiv 2024

  8. [8]

    Unsupervised part discovery from con- trastive reconstruction

    Subhabrata Choudhury, Iro Laina, Christian Rupprecht, and Andrea Vedaldi. Unsupervised part discovery from con- trastive reconstruction. Advances in Neural Information Pro- cessing Systems, 34:28104–28118, 2021. 3, 4

    work page 2021

Show all 48 references
  1. [9]

    Imagenet: A large-scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009. 1, 6

    2009

  2. [11]

    An image is worth 16x16 words: Trans- formers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, et al. An image is worth 16x16 words: Trans- formers for image recognition at scale. arXiv preprint a...

    2010 arXiv

  3. [12]

    A unified objective for novel class discovery

    Enrico Fini, Enver Sangineto, St ´ephane Lathuili `ere, Zhun Zhong, Moin Nabi, and Elisa Ricci. A unified objective for novel class discovery. InProceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pages 9284–9292,

  4. [13]

    Statistical theory of extreme values and some practical applications: a series of lectures

    Emil Julius Gumbel. Statistical theory of extreme values and some practical applications: a series of lectures . US Gov- ernment Printing Office, 1954. 5

    1954

  5. [14]

    Noise-contrastive estimation: A new estimation principle for unnormalized statistical models

    Michael Gutmann and Aapo Hyv ¨arinen. Noise-contrastive estimation: A new estimation principle for unnormalized statistical models. In Proceedings of the thirteenth inter- national conference on artificial intelligence and statistics , pages 297–304. JMLR Workshop and Conferen...

    2010

  6. [15]

    Automatically discov- ering and learning new visual categories with ranking statis- tics

    Kai Han, Sylvestre-Alvise Rebuffi, Sebastien Ehrhardt, An- drea Vedaldi, and Andrew Zisserman. Automatically discov- ering and learning new visual categories with ranking statis- tics. arXiv preprint arXiv:2002.05714, 2020. 7

    2002 arXiv

  7. [16]

    Cipr: An efficient framework with cross-instance positive rela- tions for generalized category discovery

    Shaozhe Hao, Kai Han, and Kwan-Yee K Wong. Cipr: An efficient framework with cross-instance positive rela- tions for generalized category discovery. arXiv preprint arXiv:2304.06928, 2023. 7

    2023 arXiv

  8. [17]

    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. 1

    2016

  9. [18]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR,

  10. [19]

    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. 3

    2020

  11. [20]

    Interpretable and accurate fine- grained recognition via region grouping

    Zixuan Huang and Yin Li. Interpretable and accurate fine- grained recognition via region grouping. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8662–8672, 2020. 3

    2020

  12. [21]

    Scops: Self-supervised co-part segmentation

    Wei-Chih Hung, Varun Jampani, Sifei Liu, Pavlo Molchanov, Ming-Hsuan Yang, and Jan Kautz. Scops: Self-supervised co-part segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 869–878, 2019. 3

    2019

  13. [22]

    3d object representations for fine-grained categorization

    Jonathan Krause, Michael Stark, Jia Deng, and Li Fei-Fei. 3d object representations for fine-grained categorization. In Proceedings of the IEEE international conference on com- puter vision workshops, pages 554–561, 2013. 6

    2013

  14. [23]

    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

  15. [24]

    Panoptic-partformer: Learning a unified model for panoptic part segmentation

    Xiangtai Li, Shilin Xu, Yibo Yang, Guangliang Cheng, Yun- hai Tong, and Dacheng Tao. Panoptic-partformer: Learning a unified model for panoptic part segmentation. In European Conference on Computer Vision , pages 729–747. Springer,

  16. [25]

    Microsoft coco: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll´ar, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceeding...

    2014

  17. [26]

    A* sam- pling

    Chris J Maddison, Daniel Tarlow, and Tom Minka. A* sam- pling. Advances in neural information processing systems , 27, 2014. 5

    2014

  18. [27]

    Fine-grained visual classi- 9 fication of aircraft

    Subhransu Maji, Esa Rahtu, Juho Kannala, Matthew Blaschko, and Andrea Vedaldi. Fine-grained visual classi- 9 fication of aircraft. arXiv preprint arXiv:1306.5151 , 2013. 6

    2013 arXiv

  19. [28]

    Gmnet: Graph matching network for large scale part semantic segmentation in the wild

    Umberto Michieli, Edoardo Borsato, Luca Rossi, and Pietro Zanuttigh. Gmnet: Graph matching network for large scale part semantic segmentation in the wild. InComputer Vision– ECCV 2020: 16th European Conference, Glasgow, UK, Au- gust 23–28, 2020, Proceedings, Part VIII 16, page...

    2020

  20. [29]

    Dinov2: Learning robust visual features without supervision

    Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023. 2, 3, 4

    2023 arXiv

  21. [30]

    Dynamic conceptional contrastive learning for generalized category discovery

    Nan Pu, Zhun Zhong, and Nicu Sebe. Dynamic conceptional contrastive learning for generalized category discovery. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 7579–7588, 2023. 1, 3, 7

    2023

  22. [31]

    Learn to categorize or categorize to learn? self-coding for general- ized category discovery

    Sarah Rastegar, Hazel Doughty, and Cees Snoek. Learn to categorize or categorize to learn? self-coding for general- ized category discovery. Advances in Neural Information Processing Systems, 36, 2024. 1, 7

    2024

  23. [32]

    Selex: Self-expertise in fine-grained generalized category discovery

    Sarah Rastegar, Mohammadreza Salehi, Yuki M Asano, Hazel Doughty, and Cees GM Snoek. Selex: Self-expertise in fine-grained generalized category discovery. arXiv preprint arXiv:2408.14371, 2024. 7

    2024 arXiv

  24. [33]

    Particle: Part discov- ery and contrastive learning for fine-grained recognition

    Oindrila Saha and Subhransu Maji. Particle: Part discov- ery and contrastive learning for fine-grained recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 167–176, 2023. 3

    2023

  25. [34]

    Very deep convo- lutional networks for large-scale image recognition

    Karen Simonyan and Andrew Zisserman. Very deep convo- lutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014. 1

    2014 arXiv

  26. [35]

    Going denser with open-vocabulary part segmentation

    Peize Sun, Shoufa Chen, Chenchen Zhu, Fanyi Xiao, Ping Luo, Saining Xie, and Zhicheng Yan. Going denser with open-vocabulary part segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 15453–15465, 2023. 3

    2023

  27. [36]

    The herbarium challenge 2019 dataset

    Kiat Chuan Tan, Yulong Liu, Barbara Ambrose, Melissa Tulig, and Serge Belongie. The herbarium challenge 2019 dataset. arXiv preprint arXiv:1906.05372, 2019. 6

    2019 arXiv

  28. [37]

    Con- trastive multiview coding

    Yonglong Tian, Dilip Krishnan, and Phillip Isola. Con- trastive multiview coding. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XI 16 , pages 776–794. Springer,

    2020

  29. [38]

    Pdisconet: Semantically consistent part discovery for fine-grained recognition

    Robert van der Klis, Stephan Alaniz, Massimiliano Mancini, Cassio F Dantas, Dino Ienco, Zeynep Akata, and Diego Mar- cos. Pdisconet: Semantically consistent part discovery for fine-grained recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision,...

    2023

  30. [39]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017. 3

    2017

  31. [40]

    Generalized category discovery

    Sagar Vaze, Kai Han, Andrea Vedaldi, and Andrew Zisser- man. Generalized category discovery. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7492–7501, 2022. 1, 3, 6, 7

    2022

  32. [41]

    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

  33. [42]

    Sptnet: An efficient alternative framework for generalized category discovery with spatial prompt tuning

    Hongjun Wang, Sagar Vaze, and Kai Han. Sptnet: An efficient alternative framework for generalized category discovery with spatial prompt tuning. arXiv preprint arXiv:2403.13684, 2024. 1, 3, 4, 6, 7

    2024 arXiv

  34. [43]

    Parametric classification for generalized category discovery: A baseline study

    Xin Wen, Bingchen Zhao, and Xiaojuan Qi. Parametric classification for generalized category discovery: A baseline study. In Proceedings of the IEEE/CVF International Con- ference on Computer Vision, pages 16590–16600, 2023. 1, 2, 3, 4, 6, 7

    2023

  35. [44]

    Decompose novel into known: Part concept learning for 3d novel class discovery

    Tingyu Weng, Jun Xiao, and Haiyong Jiang. Decompose novel into known: Part concept learning for 3d novel class discovery. Advances in Neural Information Processing Sys- tems, 36:54282–54299, 2023. 3

    2023

  36. [45]

    Prompt- cal: Contrastive affinity learning via auxiliary prompts for generalized novel category discovery

    Sheng Zhang, Salman Khan, Zhiqiang Shen, Muzammal Naseer, Guangyi Chen, and Fahad Shahbaz Khan. Prompt- cal: Contrastive affinity learning via auxiliary prompts for generalized novel category discovery. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern R...

    2023

  37. [46]

    Novel visual category discov- ery with dual ranking statistics and mutual knowledge distil- lation

    Bingchen Zhao and Kai Han. Novel visual category discov- ery with dual ranking statistics and mutual knowledge distil- lation. Advances in Neural Information Processing Systems, 34:22982–22994, 2021. 3

    2021

  38. [47]

    Learning semi- supervised gaussian mixture models for generalized category discovery

    Bingchen Zhao, Xin Wen, and Kai Han. Learning semi- supervised gaussian mixture models for generalized category discovery. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 16623–16633, 2023. 7

    2023

  39. [48]

    Multi- class part parsing with joint boundary-semantic awareness

    Yifan Zhao, Jia Li, Yu Zhang, and Yonghong Tian. Multi- class part parsing with joint boundary-semantic awareness. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 9177–9186, 2019. 3

    2019

  40. [49]

    Learn- ing multi-attention convolutional neural network for fine- grained image recognition

    Heliang Zheng, Jianlong Fu, Tao Mei, and Jiebo Luo. Learn- ing multi-attention convolutional neural network for fine- grained image recognition. In Proceedings of the IEEE inter- national conference on computer vision , pages 5209–5217,

Pith tools

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