Pith. sign in

REVIEW 5 major objections 4 minor 50 references

Style-Pro: Style-Guided Prompt Learning for Generalizable Vision-Language Models

T0 review · 5 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Style-Pro claims that injecting learnable style bases into a prompt-tuned CLIP, plus consistency constraints to the frozen model, prevents overfitting and improves base-to-novel, cross-dataset, and domain generalization.

desk verdict A plausible style-augmentation plus consistency recipe for prompt learning, with real benchmark work but a broken content-consistency equation that needs fixing before the claims hold. read the letter →

arxiv 2411.16018 v1 pith:JEXC3FM2 submitted 2024-11-25 cs.CV

Niloufar Alipour Talemi , Hossein Kashiani , Fatemeh Afghah This is my paper · ORCID
classification cs.CV
keywords style-guidedpromptlearningvision-languagemodelsCLIPfew-shotimagerecognitiontuningdomaingeneralizationstylebasesconsistencyregularization
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 Style-Pro, a prompt-learning method that adapts CLIP to a downstream task while trying to keep CLIP's zero-shot powers. Its central claim is that few-shot prompt tuning overfits because the training distribution is too narrow, so Style-Pro synthesizes diverse new styles with learnable style bases, maps unseen styles to weighted combinations of those bases, and adds consistency losses that keep the prompted model close to the frozen CLIP. On 11 image-classification benchmarks the method reports a 0.83% average gain over the CoPrompt baseline on novel classes, a 0.22% gain on base classes, and consistent improvements in cross-dataset transfer and domain generalization. The point of the work is to show that style-guided shifts in feature space can replace image-level augmentation for keeping prompt-tuned vision-language models general.

What carries the argument

The load-bearing mechanism is the style-shift learning module built from learnable style bases, each defined by a mean and standard deviation pair $(\mu^n_b, \sigma^n_b)$ in feature space. For every training image, Style-Pro computes the AdaIN-style statistics of the current layer's features, measures the Wasserstein distance to each basis, converts those distances into softmax weights, and mixes the bases into a mapped style that is applied to the normalized features before the next transformer layer. Two auxiliary losses shape the bases: $L_{\text{Diversity}}$ pushes the bases toward mutual orthogonality on a hypersphere, and $L_{\text{Content}}$ penalizes the deviation of $\operatorname{diag}(\Sigma_{\bar{F}^P_l,\bar{F}_l})$ from the all-ones vector. The rest of the objective is consistency-based: an MSE loss aligns prompted vision and text features with frozen CLIP features, and a KL-divergence loss aligns the prompted model's image-text similarity predictions with the frozen model's predictions.

What would settle it

Log the average diagonal of the cross-covariance between the prompted and frozen patch features during training: if it stabilizes well below 1 at convergence, the content-consistency premise is violated and any accuracy gain must be explained by another mechanism. A complementary check is to compare training-set predictions on original versus style-shifted features; if the shifted features change the predicted class for a large fraction of samples, the style shift is not preserving content.

Watch

Extended reading notes

Core claim

The central discovery the paper argues for is that a prompt-tuned CLIP can avoid overfitting if the vision encoder is pushed through a learned set of style bases during training. Style-Pro extracts channel-wise mean and standard deviation statistics from intermediate vision features, treats a collection of learnable style bases as points in that style space, and re-styles each training feature with a weighted combination of the bases, where the weights come from a Wasserstein-distance similarity between the current style and each basis. It also imposes two constraints on the style bases themselves: a diversity loss that spreads the bases orthogonally on a hypersphere, and a content-consistency loss that tries to keep the diagonal of the cross-covariance between prompted and frozen patch features equal to one. The paper reports that this style-shift module, combined with feature-level and cross-modality consistency losses against the frozen CLIP, raises novel-class accuracy to a 78.06% average and the base-novel harmonic mean to 80.98% across the 11 datasets, with the largest gains on unseen classes.

Load-bearing premise

The load-bearing premise is that the prompted and frozen image encoders still carry identical content at the patch level, so the diagonal of their cross-covariance should be 1, even though the prompts are meant to change those features.

Editorial extensions

If this is right

  • If Style-Pro is correct, few-shot prompt tuning can keep zero-shot generalization: on unseen classes it reports a 78.06% average accuracy, a 0.83% gain over CoPrompt and a 2.83% gain over zero-shot CLIP.
  • Cross-dataset transfer improves: training prompts only on ImageNet and evaluating on 10 other datasets yields a 67.24% average, 0.24% above the previous best, suggesting the learned style mapping transfers across tasks.
  • Domain generalization on ImageNet variants improves to a 61.49% average, led by ImageNet-R at 77.98%, indicating the style bases absorb at least some of the shift between natural and sketch, adversarial, and rendered images.
  • Ablations attribute the gains to the combination of consistency and style-shift losses: the harmonic mean rises from 77.66% for the bare IVLP baseline to 80.98% when all components are active, with the largest jump coming from style diversity and content preservation.

Reading between the lines

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

  • The authors do not test it, but one extension is to visualize which style bases receive the largest weights on each dataset; the resulting weight vectors could serve as an interpretable description of a dataset's domain fingerprint.
  • A further extension would be to replace the hard diagonal-equals-one content constraint with a learned projection or a soft margin, so the content loss can be tuned not to fight the very feature changes the prompts are meant to induce.
  • If the style-bases mechanism generalizes, it could be applied to other frozen foundation models beyond CLIP, or to video and audio encoders, since it only manipulates intermediate feature statistics and never requires extra images.
  • One testable prediction is that the optimal number of style bases and the best layer for style shifting will vary with the gap between source and target domains; automatic selection of both hyperparameters could yield larger gains than the fixed N=12 and layer-2 settings reported.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 4 minor

Summary. The manuscript proposes Style-Pro, a prompt-learning method for CLIP that combines learnable style bases with two auxiliary losses (style diversity and content consistency) and two consistency constraints (feature-level and cross-modality) between the prompted and frozen CLIP branches. The authors report experiments on 11 image classification benchmarks in base-to-novel generalization, cross-dataset transfer, and domain generalization, claiming consistent improvements over CoPrompt, PromptSRC, and other state-of-the-art prompt-tuning methods. The central idea is that style shifts synthesized in feature space can reduce overfitting and preserve zero-shot generalization, and that consistency with frozen CLIP keeps the prompted model aligned during adaptation.

Significance. If the method is made precise and rigorously validated, it addresses a real problem in prompt learning: minimizing overfitting to few-shot base classes while retaining CLIP's zero-shot generalization. The style-shift mechanism, which synthesizes out-of-distribution styles with learnable bases rather than external images, is a plausible and potentially useful idea, and the paper provides broad benchmark coverage with helpful ablations (Table 4, Table 5, Fig. 3). However, the core equations that define the style distance and the content-consistency loss are currently ill-defined, and the reported average gains are small relative to the likely run-to-run variability. These issues must be resolved before the empirical claims can be accepted.

major comments (5)
  1. [§3.2, Eq. (5)] The quantity dcur is written as ||µcur − µn_b||^2_2 plus (σ_cur^2 + σ_n_b^2 − 2 σ_cur σ_n_b). Since µ and σ are D-dimensional vectors in Eqs. (3)-(4), the first term is a scalar while the second term is a D-dimensional vector, so the equality is not well-formed and dcur is not a scalar distance. This is load-bearing because Eq. (6) applies a scalar softmax over 1/(1+dcur) and Eq. (7) forms scalar-weighted sums of the style bases; if dcur is a vector, the weighting is undefined. The paper should define dcur as, for example, dcur = ||µcur − µn_b||^2_2 + ||σcur − σn_b||^2_2, or explicitly state an elementwise/summed convention.
  2. [§3.2, Eqs. (10)-(11)] The content-consistency loss is not well defined. Both FbarP_l and Fbar_l are (P^2+1)×D token matrices, so the quantity E[FbarP_l · (Fbar_l)^T] is a (P^2+1)×(P^2+1) matrix, not a cross-covariance over feature dimensions; the expectation is over no explicitly specified distribution. The claim that the diagonal equals 1 when content is identical requires the token vectors to be unit-normalized and exactly aligned, which is neither stated nor proved. If a D×D Barlow-Twins-type matrix is intended, the correct construction is E[(FbarP_l)^T Fbar_l] with an additional batch normalization and column normalization. Without a precise definition, LContent is ambiguous, and the ablation in Table 4 cannot be interpreted as validating content preservation.
  3. [§3.2, Eqs. (13)-(14)] The cross-modality alignment loss is written as D_KL(P_re, P_rep), where P_re = sim(˜f, ˜g) and P_rep = sim(˜fp, ˜gp). As written, these are scalar cosine similarities between a single image feature and a single text feature, but the KL divergence is only defined between probability distributions over classes. The equations must show the softmax over the C class-text features used in Eq. (1), or the loss term is undefined. This is a fixable notational gap, but it affects a term in the final objective in Eq. (15).
  4. [§4, Tables 1-3] The central claims rest on small average differences (0.83, 0.24, and 0.69 percentage points in the three settings), and some individual datasets go against the trend, e.g., ImageNet base accuracy is 77.58 for Style-Pro versus 77.67 for CoPrompt, and EuroSAT base accuracy is 94.52 versus 94.60. The paper states that results are averaged over three runs but reports no standard deviations, confidence intervals, or significance tests. Without these, the assertion that Style-Pro 'consistently surpasses' prior methods is not supported; the authors should report per-run results and appropriate statistical comparisons.
  5. [§4.1 and §4.5] Several design choices appear to be tuned using the same benchmarks that are later reported as results: N=12 style bases, the style-shift layer (layer 2), and the prompt-depth settings (3 vs 9 layers) are selected with reference to Tables 1-3 and Fig. 3. The paper does not describe a validation split or a nested selection procedure. If these hyperparameters were chosen by inspecting test-set accuracy, the reported generalization numbers are optimistically biased. The authors should either document a validation-based selection protocol or present the layer/N analysis as a separate sensitivity study that is not used to select the final configuration.
minor comments (4)
  1. [§3.1, Eq. (1)] As written, the equation gives an average softmax probability, not a cross-entropy loss; it should include a negative logarithm over the softmax term.
  2. [§3.2, Eqs. (3)-(4)] The double sums use the same summation index k twice, and the input F'_l has no explicit indices. Please specify that the mean and standard deviation are computed over spatial positions for each channel, and state whether the CLS token is included in the style statistics.
  3. [Table 4] The checkmark/column layout does not allow the reader to determine which constraints (Feature alignment, Cross-modality alignment, Content loss, Diversity loss, Style Shift) are active in each row. Please reformat the table or spell out each configuration in the caption.
  4. [§4.1] The paper uses prompts in the first three transformer layers for domain generalization and cross-dataset evaluation but in the first nine layers for base-to-novel generalization. This difference changes the number of learnable parameters and should be justified, especially because comparison methods may use a fixed depth.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Style-Pro's benchmark claims are empirical results from held-out classes and datasets, and its regularizers are ordinary learnable-parameter constraints rather than fitted values renamed as predictions.

full rationale

Style-Pro's central claims are empirical benchmark results, not derivations from fitted values. The style bases are ordinary learnable parameters optimized with a supervised cross-entropy loss plus regularizers, and the reported base-to-novel, cross-dataset, and domain-generalization numbers are evaluated on held-out classes and datasets not used to fit the prompts or style bases. The content-consistency loss in Eqs. 10-11 contains a definitional and dimensional gap (the cross-covariance product's shapes and the 'diagonal equals 1' claim are not fully specified), but that is a correctness or validity concern, not circularity: the loss is a regularizer, and the paper does not define the predicted accuracy in terms of that loss. There are no author self-citations in the reference list, and no uniqueness theorem or central premise is imported from the authors' prior work. The ablations and choices of hyperparameters such as N and layer location are standard empirical tuning, and they do not make any prediction reduce by construction. Therefore no circular step is present.

Assumptions & free parameters 7 free parameters · 4 assumptions · 1 invented entities

The method relies on several hand-set weights and design choices, plus a new parametric construct (style bases). The core novelty rests on assumptions about feature statistics being a sufficient style descriptor and about cross-covariance diagonals being 1 for identical content. These assumptions are not validated independently.

free parameters (7)
  • lambda_f (vision feature alignment weight) = 15
    Weight on the MSE feature-alignment term for the vision encoder in Eq. 12; chosen by hand without reported sensitivity analysis.
  • lambda_g (text feature alignment weight) = 25
    Weight on the MSE feature-alignment term for the text encoder in Eq. 12; chosen by hand.
  • lambda_1 (diversity loss weight) = 0.005
    Weight on L_Diversity in Eq. 9 and Eq. 15; chosen by hand.
  • lambda_2 (content loss weight) = 0.2
    Weight on L_Content in Eq. 11 and Eq. 15; chosen by hand.
  • N (number of style bases) = 12
    Selected based on the ablation in Fig. 3(b) on the same test benchmarks; no separate validation set is reported.
  • style_shift_layer = 2
    Layer of the vision encoder at which style shift is applied; selected based on Fig. 3(a) on the same test benchmarks.
  • prompt_depth = 9 layers (base-to-novel), 3 layers (cross-dataset and domain generalization)
    Number of transformer layers receiving prompts; differs by experimental setting without reported justification.
assumptions (4)
  • domain assumption Style information of an image is fully captured by the channel-wise mean and standard deviation of patch features (AdaIN assumption).
    Section 3.2 adopts this from Huang and Belongie (AdaIN); if false, the style-shift mechanism has no meaningful basis.
  • domain assumption Identical content in two feature maps implies a diagonal of ones in their cross-covariance after normalization.
    Eqs. 10-11 assume this Barlow Twins style intuition; the paper does not verify that prompted and frozen features satisfy it.
  • ad hoc to paper A set of 12 learnable style bases spans the space of natural domain shifts for the target benchmarks.
    Introduced for this method; N is tuned on test data (Fig. 3b), so the spanning assumption is not independently validated.
  • standard math The 2-Wasserstein distance between Gaussian style statistics is an appropriate similarity measure for style bases.
    The closed-form distance is standard, but the paper writes it incorrectly in Eq. 5; the intended use is a standard scalar distance.
invented entities (1)
  • Learnable style bases (mu_b, sigma_b)
    purpose: Synthesize diverse distribution shifts and map unseen styles as weighted combinations.
    A new parametric construct introduced in this paper; no external evidence that 12 bases suffice or that the style manifold is representable this way.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Style-Pro: Style-Guided Prompt Learning for Generalizable Vision-Language Models." pith.science (2026). https://pith.science/paper/JEXC3FM2

@misc{pith2026241116018,
  author       = {Pith},
  title        = {Pith review of: Style-Pro: Style-Guided Prompt Learning for Generalizable Vision-Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JEXC3FM2}},
  note         = {Machine review of arXiv:2411.16018}
}
read the original abstract

Pre-trained Vision-language (VL) models, such as CLIP, have shown significant generalization ability to downstream tasks, even with minimal fine-tuning. While prompt learning has emerged as an effective strategy to adapt pre-trained VL models for downstream tasks, current approaches frequently encounter severe overfitting to specific downstream data distributions. This overfitting constrains the original behavior of the VL models to generalize to new domains or unseen classes, posing a critical challenge in enhancing the adaptability and generalization of VL models. To address this limitation, we propose Style-Pro, a novel style-guided prompt learning framework that mitigates overfitting and preserves the zero-shot generalization capabilities of CLIP. Style-Pro employs learnable style bases to synthesize diverse distribution shifts, guided by two specialized loss functions that ensure style diversity and content integrity. Then, to minimize discrepancies between unseen domains and the source domain, Style-Pro maps the unseen styles into the known style representation space as a weighted combination of style bases. Moreover, to maintain consistency between the style-shifted prompted model and the original frozen CLIP, Style-Pro introduces consistency constraints to preserve alignment in the learned embeddings, minimizing deviation during adaptation to downstream tasks. Extensive experiments across 11 benchmark datasets demonstrate the effectiveness of Style-Pro, consistently surpassing state-of-the-art methods in various settings, including base-to-new generalization, cross-dataset transfer, and domain generalization.

Figures

Figures reproduced from arXiv: 2411.16018 by the authors.

Figure 1
Figure 1. Illustration of the proposed style shift learning approach. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of the proposed Style-Pro framework. Style-Pro introduces a style-guided prompt learning framework, incorporating [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. (a) Ablation study on style shift learning at different [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 33 canonical work pages

  1. [1]

    Stylip: Multi-scale style- conditioned prompt learning for clip-based domain gener- alization

    Shirsha Bose, Ankit Jha, Enrico Fini, Mainak Singha, Elisa Ricci, and Biplab Banerjee. Stylip: Multi-scale style- conditioned prompt learning for clip-based domain gener- alization. In Proceedings of the IEEE/CVF Winter Confer- ence on Applications of Computer Vision, pages 5542–5552,

    work page

  2. [2]

    Food-101–mining discriminative components with random forests

    Lukas Bossard, Matthieu Guillaumin, and Luc Van Gool. Food-101–mining discriminative components with random forests. In European Conference on Computer Vision, pages 446–461. Springer, 2014. 6

    work page 2014

  3. [3]

    LASP: Text-to- text optimization for language-aware soft prompting of vi- sion & language models

    Adrian Bulat and Georgios Tzimiropoulos. LASP: Text-to- text optimization for language-aware soft prompting of vi- sion & language models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 23232–23241, 2023. 1

    work page 2023

  4. [4]

    PLOT: Prompt learning with optimal transport for vision-language models

    Guangyi Chen, Weiran Yao, Xiangchen Song, Xinyue Li, Yongming Rao, and Kun Zhang. PLOT: Prompt learning with optimal transport for vision-language models. arXiv preprint arXiv:2210.01253, 2022. 2

    arXiv 2022

  5. [5]

    TokenMixup: Efficient attention-guided token-level data augmentation for transformers

    Hyeong Kyu Choi, Joonmyung Choi, and Hyunwoo J Kim. TokenMixup: Efficient attention-guided token-level data augmentation for transformers. Advances in Neural Infor- mation Processing Systems, 35:14224–14235, 2022. 3

    work page 2022

  6. [6]

    Describing textures in the wild

    Mircea Cimpoi, Subhransu Maji, Iasonas Kokkinos, Sammy Mohamed, and Andrea Vedaldi. Describing textures in the wild. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 3606–3613,

  7. [7]

    RandAugment: Practical automated data augmenta- tion with a reduced search space

    Ekin D Cubuk, Barret Zoph, Jonathon Shlens, and Quoc V Le. RandAugment: Practical automated data augmenta- tion with a reduced search space. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pages 702–703, 2020. 2, 4, 8

    work page 2020

  8. [8]

    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 Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 248–255. Ieee, 2009. 6, 7

    work page 2009

Show all 50 references
  1. [9]

    Bayesian prompt learn- ing for image-language model generalization

    Mohammad Mahdi Derakhshani, Enrique Sanchez, Adrian Bulat, Victor G Turrisi da Costa, Cees GM Snoek, Georgios Tzimiropoulos, and Brais Martinez. Bayesian prompt learn- ing for image-language model generalization. In Proceed- ings of the IEEE/CVF International Conference on Com...

    2023

  2. [10]

    De- coupling zero-shot semantic segmentation

    Jian Ding, Nan Xue, Gui-Song Xia, and Dengxin Dai. De- coupling zero-shot semantic segmentation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11583–11592, 2022. 1

    2022

  3. [11]

    Learning gener- ative visual models from few training examples: An incre- mental bayesian approach tested on 101 object categories

    Li Fei-Fei, Rob Fergus, and Pietro Perona. Learning gener- ative visual models from few training examples: An incre- mental bayesian approach tested on 101 object categories. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pages ...

    2004

  4. [12]

    Prompt- Det: Towards open-vocabulary detection using uncurated images

    Chengjian Feng, Yujie Zhong, Zequn Jie, Xiangxiang Chu, Haibing Ren, Xiaolin Wei, Weidi Xie, and Lin Ma. Prompt- Det: Towards open-vocabulary detection using uncurated images. In European Conference on Computer Vision, pages 701–717. Springer, 2022. 1

    2022

  5. [13]

    StyleAdv: Meta style adversarial training for cross-domain few-shot learning

    Yuqian Fu, Yu Xie, Yanwei Fu, and Yu-Gang Jiang. StyleAdv: Meta style adversarial training for cross-domain few-shot learning. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 24575–24584, 2023. 4

    2023

  6. [14]

    CLIP-adapter: Better vision-language models with fea- ture adapters

    Peng Gao, Shijie Geng, Renrui Zhang, Teli Ma, Rongyao Fang, Yongfeng Zhang, Hongsheng Li, and Yu Qiao. CLIP-adapter: Better vision-language models with fea- ture adapters. International Journal of Computer Vision , 132(2):581–595, 2024. 1

    2024

  7. [15]

    CLIP-S4: Language-guided self-supervised semantic seg- mentation

    Wenbin He, Suphanut Jamonnak, Liang Gou, and Liu Ren. CLIP-S4: Language-guided self-supervised semantic seg- mentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11207– 11216, 2023. 1

    2023

  8. [16]

    EuroSAT: A novel dataset and deep learn- ing benchmark for land use and land cover classification

    Patrick Helber, Benjamin Bischke, Andreas Dengel, and Damian Borth. EuroSAT: A novel dataset and deep learn- ing benchmark for land use and land cover classification. IEEE Journal of Selected Topics in Applied Earth Observa- tions and Remote Sensing, 12(7):2217–2226, 2019. 6

    2019

  9. [17]

    The many faces of robust- ness: A critical analysis of out-of-distribution generalization

    Dan Hendrycks, Steven Basart, Norman Mu, Saurav Kada- vath, Frank Wang, Evan Dorundo, Rahul Desai, Tyler Zhu, Samyak Parajuli, Mike Guo, et al. The many faces of robust- ness: A critical analysis of out-of-distribution generalization. In Proceedings of the IEEE/CVF Internation...

    2021

  10. [18]

    Natural adversarial examples

    Dan Hendrycks, Kevin Zhao, Steven Basart, Jacob Stein- hardt, and Dawn Song. Natural adversarial examples. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 15262–15271, 2021. 6

    2021

  11. [19]

    StyleMix: Sep- arating content and style for enhanced data augmentation

    Minui Hong, Jinwoo Choi, and Gunhee Kim. StyleMix: Sep- arating content and style for enhanced data augmentation. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 14862–14870, 2021. 2, 4

    2021

  12. [20]

    Arbitrary style transfer in real-time with adaptive instance normalization

    Xun Huang and Serge Belongie. Arbitrary style transfer in real-time with adaptive instance normalization. In Proceed- ings of the IEEE/CVF International Conference on Com- puter Vision, pages 1501–1510, 2017. 4

    2017

  13. [21]

    Scaling up visual and vision-language representa- tion learning with noisy text supervision

    Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc Le, Yun-Hsuan Sung, Zhen Li, and Tom Duerig. Scaling up visual and vision-language representa- tion learning with noisy text supervision. In International Conference on Machine Learning, pages 4904–4916. PMLR,

  14. [22]

    MaPLe: Multi-modal prompt learning

    Muhammad Uzair Khattak, Hanoona Rasheed, Muhammad Maaz, Salman Khan, and Fahad Shahbaz Khan. MaPLe: Multi-modal prompt learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 19113–19122, 2023. 1, 2, 3, 6, 7, 8

    2023

  15. [23]

    Self-regulating prompts: Foundational model adaptation without forgetting

    Muhammad Uzair Khattak, Syed Talal Wasim, Muzam- mal Naseer, Salman Khan, Ming-Hsuan Yang, and Fa- had Shahbaz Khan. Self-regulating prompts: Foundational model adaptation without forgetting. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 15...

    2023

  16. [24]

    How to adapt your large-scale vision-and-language model

    Konwoo Kim, Michael Laskin, Igor Mordatch, and Deepak Pathak. How to adapt your large-scale vision-and-language model. 2021. 1

    2021

  17. [25]

    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/CVF International Conference on Computer Vision Workshops, pages 554–561, 2013. 6

    2013

  18. [26]

    Cross- domain ensemble distillation for domain generalization

    Kyungmoon Lee, Sungyeon Kim, and Suha Kwak. Cross- domain ensemble distillation for domain generalization. In European Conference on Computer Vision , pages 1–20. Springer, 2022. 3

    2022

  19. [27]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017. 3

    2017 arXiv

  20. [28]

    Prompt distribution learning

    Yuning Lu, Jianzhuang Liu, Yonggang Zhang, Yajing Liu, and Xinmei Tian. Prompt distribution learning. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5206–5215, 2022. 2, 6, 7

    2022

  21. [29]

    Fine-grained visual classi- fication of aircraft

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

    2013 arXiv

  22. [30]

    I2DFormer: Learning image to docu- ment attention for zero-shot image classification

    Muhammad Ferjad Naeem, Yongqin Xian, Luc V Gool, and Federico Tombari. I2DFormer: Learning image to docu- ment attention for zero-shot image classification. Advances in Neural Information Processing Systems, 35:12283–12294,

  23. [31]

    Automated flower classification over a large number of classes

    Maria-Elena Nilsback and Andrew Zisserman. Automated flower classification over a large number of classes. In 2008 Sixth Indian Conference on Computer Vision, Graphics & Image Processing, pages 722–729. IEEE, 2008. 6

    2008

  24. [32]

    Cats and dogs

    Omkar M Parkhi, Andrea Vedaldi, Andrew Zisserman, and CV Jawahar. Cats and dogs. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 3498–3505. IEEE, 2012. 6

    2012

  25. [33]

    Learn- ing transferable visual models from natural language super- vision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learn- ing transferable visual models from natural language super- vision. In International Conference on Machine Learning...

    2021

  26. [34]

    DenseCLIP: Language-guided dense prediction with context-aware prompting

    Yongming Rao, Wenliang Zhao, Guangyi Chen, Yansong Tang, Zheng Zhu, Guan Huang, Jie Zhou, and Jiwen Lu. DenseCLIP: Language-guided dense prediction with context-aware prompting. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 18082–...

    2022

  27. [35]

    Fine-tuned CLIP models are efficient video learners

    Hanoona Rasheed, Muhammad Uzair Khattak, Muhammad Maaz, Salman Khan, and Fahad Shahbaz Khan. Fine-tuned CLIP models are efficient video learners. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6545–6554, 2023. 3, 6

    2023

  28. [36]

    Do ImageNet classifiers generalize to Im- ageNet? In International Conference on Machine Learning, pages 5389–5400

    Benjamin Recht, Rebecca Roelofs, Ludwig Schmidt, and Vaishaal Shankar. Do ImageNet classifiers generalize to Im- ageNet? In International Conference on Machine Learning, pages 5389–5400. PMLR, 2019. 6

    2019

  29. [37]

    Consistency-guided prompt learning for vision-language models

    Shuvendu Roy and Ali Etemad. Consistency-guided prompt learning for vision-language models. arXiv preprint arXiv:2306.01195, 2023. 3, 6, 7, 8

    2023 arXiv

  30. [38]

    UCF101: A dataset of 101 human actions classes from videos in the wild

    Khurram Soomro, Amir Roshan Zamir, and Mubarak Shah. UCF101: A dataset of 101 human actions classes from videos in the wild. arXiv preprint arXiv:1212.0402 , 2012. 6

    2012 arXiv

  31. [39]

    Dropout: a simple way to prevent neural networks from overfitting

    Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. Dropout: a simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research, 15(1):1929–1958, 2014. 3

    1929

  32. [40]

    Multimodal few-shot learning with frozen language models

    Maria Tsimpoukelli, Jacob L Menick, Serkan Cabi, SM Es- lami, Oriol Vinyals, and Felix Hill. Multimodal few-shot learning with frozen language models. Advances in Neural Information Processing Systems, 34:200–212, 2021. 1

    2021

  33. [41]

    Calculation of the Wasserstein distance be- tween probability distributions on the line

    SS Vallender. Calculation of the Wasserstein distance be- tween probability distributions on the line. Theory of Proba- bility & Its Applications, 18(4):784–786, 1974. 4

    1974

  34. [42]

    Learning robust global representations by penalizing local predictive power.Advances in Neural Information Pro- cessing Systems, 32, 2019

    Haohan Wang, Songwei Ge, Zachary Lipton, and Eric P Xing. Learning robust global representations by penalizing local predictive power.Advances in Neural Information Pro- cessing Systems, 32, 2019. 6

    2019

  35. [43]

    Feature- based style randomization for domain generalization

    Yue Wang, Lei Qi, Yinghuan Shi, and Yang Gao. Feature- based style randomization for domain generalization. IEEE Transactions on Circuits and Systems for Video Technology, 32(8):5495–5509, 2022. 1

    2022

  36. [44]

    SUN database: Large-scale scene recognition from abbey to zoo

    Jianxiong Xiao, James Hays, Krista A Ehinger, Aude Oliva, and Antonio Torralba. SUN database: Large-scale scene recognition from abbey to zoo. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3485–3492. IEEE, 2010. 6

    2010

  37. [45]

    MMA: Multi-modal adapter for vision-language models

    Lingxiao Yang, Ru-Yuan Zhang, Yanchen Wang, and Xiao- hua Xie. MMA: Multi-modal adapter for vision-language models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 23826– 23837, 2024. 6, 7, 8

    2024

  38. [46]

    FILIP: Fine-grained interactive language- image pre-training

    Lewei Yao, Runhui Huang, Lu Hou, Guansong Lu, Minzhe Niu, Hang Xu, Xiaodan Liang, Zhenguo Li, Xin Jiang, and Chunjing Xu. FILIP: Fine-grained interactive language- image pre-training. arXiv preprint arXiv:2111.07783, 2021. 2

    2021 arXiv

  39. [47]

    Barlow Twins: Self-supervised learning via redundancy reduction

    Jure Zbontar, Li Jing, Ishan Misra, Yann LeCun, and St´ephane Deny. Barlow Twins: Self-supervised learning via redundancy reduction. In International conference on Ma- chine Learning, pages 12310–12320. PMLR, 2021. 5

    2021

  40. [48]

    Conditional prompt learning for vision-language mod- els

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. Conditional prompt learning for vision-language mod- els. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 16816–16825,

  41. [49]

    Learning to prompt for vision-language models

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. Learning to prompt for vision-language models. In- ternational Journal of Computer Vision, 130(9):2337–2348,

  42. [50]

    Domain generalization with MixStyle

    Kaiyang Zhou, Yongxin Yang, Yu Qiao, and Tao Xi- ang. Domain generalization with MixStyle. arXiv preprint arXiv:2104.02008, 2021. 1, 4, 8

    2021 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.