Pith. sign in

REVIEW 3 major objections 6 minor 116 references

DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks

T0 review · 3 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read The paper claims that a frozen, pretrained layout-to-image diffusion model can be inverted into an object detector that is competitive with a basic Faster R-CNN baseline on COCO, and that the same inversion speeds up generative…

desk verdict The classification speedup is real and useful, but the detection claim rests on an asserted objective and a thin eval, so treat the headline as provisional. read the letter →

arxiv 2504.17253 v1 pith:NXIYSCE7 submitted 2025-04-24 cs.CV cs.MM

classification cs.CVcs.MM
keywords diffusionmodelgenerativemodelingdiscriminativetaskobjectdetectionvisualrecognitioninversionlayout-to-imagegenerationBayes'rule
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

DIVE studies whether a purely generative model — a frozen diffusion model trained to draw images from a layout description — can do discriminative vision tasks without finetuning or a separately trained head. The paper claims to be the first to show this works for object detection, by reversing the generation direction with Bayes' rule: instead of enumerating all possible labels as prior diffusion classifiers do, DIVE directly optimizes the layout embedding with gradients, and a learned prior over layouts supplies the non-uniform $p(y)$ that detection requires. On a 500-image COCO subset, the inverted generator reaches AP 7.1, slightly above the reproduced Faster R-CNN R50 baseline at 256×256 (6.8 AP). For classification, the same optimization scheme matches the enumeration-based diffusion classifier (77.2 vs 77.3 accuracy) while cutting per-image cost by a factor of about 14.

What carries the argument

The load-bearing identity is Bayes' posterior maximization, $\arg\max_y p_\theta(x|y)p(y)$, converted into an optimization over a learnable layout embedding. The carrying mechanism is gradient-based discrete optimization inside the frozen embedding vocabulary: the current vector $v$ is replaced by its nearest vocabulary embedding via a straight-through gradient copy, so both the layout-to-image denoiser $\epsilon_\theta$ and the prior layout denoiser $\epsilon_\phi$ backprop into $v$ while the optimized value is forced to decode to a legal box/class sequence. The prior layout diffusion model plays a necessary role: it supplies the non-uniform $p(y)$, since layouts, unlike classes, cannot be assumed uniformly distributed.

What would settle it

Estimate the true conditional log-likelihoods of DIVE's predicted layouts with a proper likelihood estimator (for example, annealed importance sampling or an ELBO with many noise samples), and compare the ranking induced by the sum-of-MSE objective with the ranking induced by the estimated likelihoods on a set of correct and incorrect layouts; a mismatch would show the objective is not the posterior. A cheaper check is to re-run detection while scaling the prior loss by a free weight; if AP changes substantially with that weight, the unweighted combination in Eq. (4) is not the unique Bayes objective.

Watch

Extended reading notes

Core claim

The central discovery, stated on the paper's own terms, is that the discriminative information needed to detect objects is already present in a frozen layout-to-image diffusion model and can be extracted by optimizing the conditional input. Substituting the two diffusion training objectives into Bayes' rule gives a single inversion objective (Eq. 4): minimize the sum of the denoising error of the image model with respect to the layout and the denoising error of a separately trained layout-prior model. Because the layout is represented by discrete vocabulary embeddings, DIVE uses a nearest-embedding replacement with gradient copying, so the optimizer stays in the model's vocabulary and the final prediction is read off as nearest-neighbor indices. With this machinery the paper reports AP 7.1 versus 6.8 for Faster R-CNN R50 at the same low resolution, better AP75 than AP50 (indicating precise but incomplete detections), and a 14x speedup over enumeration for classification with nearly identical accuracy.

Load-bearing premise

The load-bearing premise is that adding the conditional image model's denoising loss and the prior layout model's denoising loss equals the log-posterior in Bayes' rule; if that equality is only approximate or mis-weighted, the optimized layout is not a MAP estimate and the reported AP is not evidence of the frozen generator's discriminative power.

Editorial extensions

If this is right

  • A frozen generative model can act as a zero-training detector: no detector head, no finetuning, and no enumeration of candidate labels are required.
  • The Bayes-inversion recipe transfers from classification to detection, so any conditional diffusion model with a differentiable condition encoder is a candidate discriminative model.
  • Detection quality tracks the model's generative prior: the paper's layout prior removes redundant objects and illegal boxes that appear when it is ablated.
  • DIVE offers a self-contained evaluation metric for conditional diffusion models that does not depend on external classifiers; its rankings agree with FID, with the caveat that diversity rather than realism tracks discriminative ability.
  • Optimization-based inversion is the main cost bottleneck; early stopping on the monitor loss could roughly halve the average 2000-step budget without, the authors argue, hurting performance.

Reading between the lines

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

  • The objective in Eq. (4) simply adds two unweighted denoising losses; a natural extension is to learn or tune a weighting between the image-conditional and prior terms, and the resulting AP would reveal how much of the detection quality owes to the prior.
  • The poor small- and medium-object AP is attributed to 8x latent downsampling; testing the same inversion on a pixel-space or higher-resolution latent model would isolate that cause and might raise small-object AP without changing the algorithm.
  • Because DIVE's advantage over Faster R-CNN appears at AP75, the method may be most useful as a precise proposal generator or verifier rather than a full detector; combining DIVE candidates with a cheap discriminative ranker is a testable hybrid.
  • The same inversion could be pointed at dense tasks such as semantic segmentation, but the optimization space would explode and per-pixel gradient contributions would be small; coarse-to-fine or grouped optimization would be needed before that becomes feasible.
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

3 major / 6 minor

Summary. This paper proposes DIVE, a method for repurposing pretrained conditional diffusion models for discriminative tasks, specifically object detection. Following the Bayes-rule inversion paradigm of prior 'diffusion classifier' work, DIVE treats the detection label (a layout sequence of class and bounding-box tokens) as learnable parameters and optimizes it by gradient descent on the sum of two diffusion denoising losses: one from a frozen layout-to-image model (the likelihood term) and one from an auxiliary prior layout diffusion model (the prior term). To handle the discrete label vocabulary, the method uses a straight-through nearest-neighbor embedding replacement. Experiments on COCO 2017 report AP 7.1 on a 500-image subset, comparable to a 256x256 from-scratch Faster R-CNN R50 (AP 6.8); ablations on 100 images show that both the prior model and the in-vocabulary optimization contribute. On ImageNet classification, DIVE matches the enumeration-based Diffusion Classifier (77.2 vs 77.3) at roughly 14x lower per-image cost. The paper additionally proposes DIVE as an evaluation metric for conditional diffusion models.

Significance. Assuming the central derivation holds, this is the first non-enumeration inversion of a frozen conditional diffusion model for a structured discriminative task with a non-uniform label space, and the learned prior-layout model is a sensible response to the non-uniform layout distribution. The paper has real strengths: code and models are promised; the ablations isolate both proposed components; the classification experiments validate the optimization-based inversion against the enumeration baseline at matched accuracy with a 14x speedup; and the detection baselines are reproduced at matching resolution, which is transparent. The paper also candidly acknowledges its limitations, including compute cost and a concluding caveat about dense-task scalability. The result, however, rests on the asserted identification in Eq. (4) between a sum of two denoising-MSE losses and the log-posterior, and the detection evaluation is thin: a 500-image subset without selection details, no error bars, and a 0.3-AP gap to the 'competitive' Faster R-CNN baseline. These issues are addressable in revision but are load-bearing for the paper's central claims.

major comments (3)
  1. [Section III-B2, Eq. (4)] The identification of the unweighted sum of two denoising-MSE expectations with log p_theta(x|y) + log p_phi(y) is asserted rather than derived. Eq. (1) itself is only approximate (the simplified epsilon-prediction loss is a reweighted version of the ELBO, with per-timestep coefficients and variance-schedule constants dropped), and Eq. (4) compounds this by summing two losses that live in different spaces (the 32x32x4 latent image space vs. the 100xd layout-embedding space) with independent noise and timestep samplings and different intrinsic scales. The default weight of 1 between the two terms is therefore not justified by any common log-density. This matters because the abstract and Section III-B frame DIVE as making 'more accurate use of the Bayes' rule' and obtaining the posterior maximum. Please either derive the relationship explicitly (e.g., via a weighted ELBO decomposition with the appropriate per-timestep coefficients), or reframe the objective as a heuristic surrogate objective; and add a validity check: on a set of held-out images, compare the objective value at the ground-truth layout against the value at corrupted or random layouts, and sweep the relative weight between the two loss terms to test whether the unit weight is near-optimal. If AP depends strongly on a non-unit weight, the method is not optimizing the stated posterior.
  2. [Section IV-A1a and Table II] The detection evaluation is currently too thin to support the headline comparison. The 500-image subset of COCO val is not described (random, seeded, or filtered), no error bars or significance tests are given, and the main claim of being on par with Faster R-CNN R50 rests on AP 7.1 vs 6.8, a 0.3-point gap that is likely within noise at this sample size; on AP50 DIVE is actually lower (11.0 vs 12.7). In addition, the optimization-step hyperparameter K is swept on the same 500 images in Table I before being fixed at 2000, so the reported AP is a test-set-selected result. Please report bootstrap confidence intervals for AP, AP50, and AP75, describe the subset selection procedure explicitly, and either temper the statement about 'competitive performance compared to discriminative methods' or support it with variance-aware comparisons.
  3. [Section IV-A1b and Section IV-A2b] The manuscript should clarify what 'frozen pretrained image generation model' means in this setup. The layout-to-image LDM-8 (363M) is retrained by the authors on COCO itself because no public model exists (footnote 1), and the prior layout model is trained on the same COCO layout distribution, so both the likelihood and the prior are in-distribution for the test task. This is not a deficiency of the inversion idea, since the Diffuser Classifier line of work has the same property, but the claim in Section IV-A2b ('the first to show that using a frozen pretrained image generation model can successfully tackle the challenging object detection task') should be qualified: the model is frozen at inference, but it is not an externally pretrained model transferring to a new task, and the prior is learned from the target label distribution. The abstract's wording about 'leveraging pretrained diffusion models' should be adjusted to match this setup.
minor comments (6)
  1. [Section IV-A1a] Please state explicitly how the 500-image subset was selected (for example, a random draw with a fixed seed) so that the reader can judge its representativeness; the current sentence 'we evaluate DIVE on a subset of 500 images' is underspecified.
  2. [Algorithm 1 / Section III-B3a] In Algorithm 1, the eval function drops none-value and illegal boxes before computing monitor losses; please clarify how the sequence is re-padded to the fixed length of 100 for the loss computation, since both diffusion models expect a fixed-length input.
  3. [Section III-B3a] The timestep set T is shared between the image model and the prior model, but the two models add noise to different spaces (image latents vs. layout embeddings) with different scales; please state the noise schedules used in each model and justify the shared T, or report sensitivity to this choice.
  4. [Section III-B2, Eqs. (5)-(7)] The notation is confusing because v is used both for the learnable continuous parameters and for the output of the frozen embedding mapping v_theta; consider distinguishing the soft and discretized versions (for example, v_soft and v_discrete) to make the straight-through operation in Eq. (6) and the decoding in Eq. (7) easier to follow.
  5. [Table I] The time-per-image cells are formatted inconsistently ('18min, 45min, 1.5h, 3h'); please unify the units and state in the caption that these AP numbers are measured on the same 500-image subset used in Table II.
  6. [Section V (Conclusion)] The concluding caveat about dense tasks, namely that the optimization space becomes large and the per-pixel objective contribution is small 'which may be problematic and needs to be addressed', is an important self-identified limitation; it should be reflected in the introduction or abstract so that the scope of the claimed contribution is clear from the start.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Eq. (4) is an explicit approximation, and detection results are benchmarked externally.

full rationale

Walking the derivation chain: DIVE starts from Bayes' rule (Eq. 2) and replaces the two log-densities with denoising-MSE surrogates (Eqs. 1 and 3), giving the combined objective in Eq. (4). The paper explicitly labels Eq. (1) as an approximation ('≈'), and for Eq. (4) it says 'we simply add the losses of the two diffusion models', so the combination is a stated modeling choice rather than a definitional identity. The predicted layout y* is obtained by optimizing this objective with fixed pretrained models; it is not fit to ground-truth detection labels, and the monitor only selects among noised/timestep evaluations using the same objective, again without label information. The central empirical claims are checked against external discriminative baselines (Faster R-CNN, DETR, DiffDet) and against the previous Diffusion Classifier, so the main result is falsifiable outside the paper's fitted values. The paper contains no self-citations and imports no uniqueness theorem from the authors' prior work. The COCO-trained layout-to-image model and COCO-trained prior do weaken the 'frozen pretrained' transfer narrative, but this is a data-distribution overlap, not an algebraic reduction of the prediction to the training input; the detection AP is still an independent function of the test image. No circular step meeting the quoted-evidence bar was found.

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

The central claim relies on several unproven surrogates: the sum of two denoising losses as a posterior, the prior model trained on the same label distribution as evaluation, and a subset-based evaluation. No new physical entities are introduced.

free parameters (5)
  • Optimization steps K = 2000
    Selected by comparing AP on the same COCO evaluation subset (Table I); no separate validation split, so this is tuning on the evaluation set.
  • Loss weighting lambda (image vs prior) = 1:1 equal weights
    The two diffusion losses in Eq. (5) are added with equal weight without ablation; the balance between reconstruction and prior plausibility is not studied.
  • Learning rate = 0.01
    AdamW learning rate for the per-image embedding optimization; a fixed choice, not ablated.
  • Timestep interval = 5 (timesteps {2,7,...,997})
    Evenly spaced timesteps following [1]; affects the noise-weighting of the loss and was not varied.
  • Monitor noise sets Eforθ, Eforϕ = random draws, seeds not specified
    Fixed noises used to compare losses across optimization steps; without specified seeds, the selection of v* via monitor values is not exactly reproducible.
assumptions (5)
  • standard math The conditional diffusion training loss in Eq. (1) is an unbiased surrogate for log pθ(x|y) up to a constant.
    Standard in diffusion literature, but for inversion the constant is assumed to be independent of y; not verified.
  • ad hoc to paper The sum of the two diffusion losses in Eq. (4) is proportional to log p(y|x).
    The paper presents this as 'simply adding the losses' without derivation; there may be weighting and normalization issues.
  • domain assumption The prior layout diffusion model pϕ(y) trained by Eq. (3) accurately models the prior distribution of object layouts.
    Trained on COCO train label sequences; assumes the training distribution matches the test layout distribution.
  • ad hoc to paper Nearest-neighbor embedding replacement with straight-through gradient (Eq. 6) yields a good discrete solution.
    Borrowed from discrete autoencoders; no convergence guarantee for this inversion objective.
  • ad hoc to paper The COCO validation subset of 500 images is representative of the full validation set.
    The paper uses a subset for compute reasons but does not describe the selection procedure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks." pith.science (2026). https://pith.science/paper/NXIYSCE7

@misc{pith2026250417253,
  author       = {Pith},
  title        = {Pith review of: DIVE: Inverting Conditional Diffusion Models for Discriminative Tasks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NXIYSCE7}},
  note         = {Machine review of arXiv:2504.17253}
}
read the original abstract

Diffusion models have shown remarkable progress in various generative tasks such as image and video generation. This paper studies the problem of leveraging pretrained diffusion models for performing discriminative tasks. Specifically, we extend the discriminative capability of pretrained frozen generative diffusion models from the classification task to the more complex object detection task, by "inverting" a pretrained layout-to-image diffusion model. To this end, a gradient-based discrete optimization approach for replacing the heavy prediction enumeration process, and a prior distribution model for making more accurate use of the Bayes' rule, are proposed respectively. Empirical results show that this method is on par with basic discriminative object detection baselines on COCO dataset. In addition, our method can greatly speed up the previous diffusion-based method for classification without sacrificing accuracy. Code and models are available at https://github.com/LiYinqi/DIVE .

Figures

Figures reproduced from arXiv: 2504.17253 by the authors.

Figure 1
Figure 1. Repurposing pretrained conditional diffusion models for [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the training framework of layout-conditional image generation model (left) and prior layout model (right). Bounding boxes in image x are for visualization only. models, the inversion can be done through a generalized de￾terministic forward diffusion process proposed in DDIM [83]. Following improvement works include exacter and faster in￾versions [84], [85] and extension to text-conditional mod￾els [8… view at source ↗
Figure 3
Figure 3. Using trained layout-conditional image generation model and the prior layout model for object detection. diffusion model trained on balanced classification datasets like ImageNet-1k [92], it is reasonable to disregard p(y) in Eq. (2) when converting the model to a “Diffusion Classifier” by calculating each class’s posterior [1], because the prior is a uniform distribution (i.e., the same for each possible class). Un… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Visualization of DIVE detection results with corresponding convergence optimization steps shown below, yielded from the monitor introduced in Section III-B3. The maximum optimization step for all images is set as a fixed number (2000) here for simplicity. The average c…
Figure 5
Figure 5. Figure 5: Visualization of the object detection results. Besides comparing DIVE with other generative baselines that use the same pretrained diffusion model as ours, we also show the influences of the prior model and in-vocabulary discrete optimization method. For these ablation…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

116 extracted references · 62 canonical work pages

  1. [1]

    Your diffusion model is secretly a zero-shot classifier

    Alexander C Li, Mihir Prabhudesai, Shivam Duggal, Ellis Brown, and Deepak Pathak. Your diffusion model is secretly a zero-shot classifier. In IEEE ICCV, pages 2206–2217, 2023

  2. [2]

    Text-to-image diffusion models are zero-shot classifiers

    Kevin Clark and Priyank Jaini. Text-to-image diffusion models are zero-shot classifiers. NeurIPS, 36, 2023

  3. [3]

    Deep unsupervised learning using nonequilibrium thermody- namics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsupervised learning using nonequilibrium thermody- namics. In ICML, pages 2256–2265, 2015

  4. [4]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. NeurIPS, 33:6840–6851, 2020

  5. [5]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Ku- mar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. In ICLR, 2021

  6. [6]

    Diffusion models beat GANs on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat GANs on image synthesis. NeurIPS, 34:8780–8794, 2021

  7. [7]

    High-resolution image synthesis with latent diffu- sion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffu- sion models. In IEEE CVPR, pages 10684–10695, 2022

  8. [8]

    Photorealistic text-to-image diffusion models with deep language understanding

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, Jonathan Ho, David J Fleet, and Moham- mad Norouzi. Photorealistic text-to-image diffusion models with deep language understanding. NeurIPS, 35:36479–36494, 2022

Show all 116 references
  1. [9]

    Neural discrete representation learning

    Aaron van den Oord, Oriol Vinyals, and Koray Kavukcuoglu. Neural discrete representation learning. NeurIPS, 30, 2017

  2. [10]

    Generating diverse high-fidelity images with VQ-V AE-2

    Ali Razavi, Aaron van den Oord, and Oriol Vinyals. Generating diverse high-fidelity images with VQ-V AE-2. NeurIPS, 32, 2019

  3. [11]

    Taming transformers for high-resolution image synthesis

    Patrick Esser, Robin Rombach, and Bjorn Ommer. Taming transformers for high-resolution image synthesis. In IEEE CVPR , pages 12873– 12883, 2021

  4. [12]

    Zero-shot text- to-image generation

    Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea V oss, Alec Radford, Mark Chen, and Ilya Sutskever. Zero-shot text- to-image generation. In ICML, pages 8821–8831, 2021

  5. [13]

    Generative adversarial nets

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial nets. NeurIPS, 27, 2014

  6. [14]

    Large scale GAN training for high fidelity natural image synthesis

    Andrew Brock, Jeff Donahue, and Karen Simonyan. Large scale GAN training for high fidelity natural image synthesis. In ICLR, 2019

  7. [15]

    A style-based generator architecture for generative adversarial networks

    Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks. In IEEE CVPR, pages 4401–4410, 2019

  8. [16]

    Is synthetic data from generative models ready for image recognition? In ICLR, 2023

    Ruifei He, Shuyang Sun, Xin Yu, Chuhui Xue, Wenqing Zhang, Philip Torr, Song Bai, and Xiaojuan Qi. Is synthetic data from generative models ready for image recognition? In ICLR, 2023

  9. [17]

    Fake it till you make it: Learning transferable represen- tations from synthetic imagenet clones

    Mert B ¨ulent Sarıyıldız, Karteek Alahari, Diane Larlus, and Yannis Kalantidis. Fake it till you make it: Learning transferable represen- tations from synthetic imagenet clones. In IEEE CVPR, pages 8011– 8021, 2023

  10. [18]

    Shekoofeh Azizi, Simon Kornblith, Chitwan Saharia, Mohammad Norouzi, and David J. Fleet. Synthetic data from diffusion models improves imagenet classification. TMLR, 2023

  11. [19]

    DiffuMask: Synthesizing images with pixel-level annotations for semantic segmentation using diffusion models

    Weijia Wu, Yuzhong Zhao, Mike Zheng Shou, Hong Zhou, and Chun- hua Shen. DiffuMask: Synthesizing images with pixel-level annotations for semantic segmentation using diffusion models. In IEEE ICCV , pages 1206–1217, 2023

  12. [20]

    Dataset diffusion: Diffusion-based synthetic data generation for pixel-level semantic segmentation

    Quang Nguyen, Truong Vu, Anh Tran, and Khoi Nguyen. Dataset diffusion: Diffusion-based synthetic data generation for pixel-level semantic segmentation. NeurIPS, 36, 2023

  13. [21]

    GeoDiffusion: Text-prompted geometric control for object detection data generation

    Kai Chen, Enze Xie, Zhe Chen, Yibo Wang, Lanqing Hong, Zhenguo Li, and Dit-Yan Yeung. GeoDiffusion: Text-prompted geometric control for object detection data generation. In ICLR, 2024

  14. [22]

    Data augmentation for object detection via controllable diffusion models

    Haoyang Fang, Boran Han, Shuai Zhang, Su Zhou, Cuixiong Hu, and Wen-Ming Ye. Data augmentation for object detection via controllable diffusion models. In IEEE WACV, pages 1257–1266, 2024

  15. [23]

    Label-efficient semantic segmentation with diffusion models

    Dmitry Baranchuk, Andrey V oynov, Ivan Rubachev, Valentin Khrulkov, and Artem Babenko. Label-efficient semantic segmentation with diffusion models. In ICLR, 2022

  16. [24]

    Open-vocabulary panoptic segmentation with text-to-image diffusion models

    Jiarui Xu, Sifei Liu, Arash Vahdat, Wonmin Byeon, Xiaolong Wang, and Shalini De Mello. Open-vocabulary panoptic segmentation with text-to-image diffusion models. In IEEE CVPR , pages 2955–2966, 2023

  17. [25]

    Diffusion model as representation learner

    Xingyi Yang and Xinchao Wang. Diffusion model as representation learner. In IEEE ICCV, pages 18938–18949, 2023

  18. [26]

    DreamTeacher: Pretraining image backbones with deep generative models

    Daiqing Li, Huan Ling, Amlan Kar, David Acuna, Seung Wook Kim, Karsten Kreis, Antonio Torralba, and Sanja Fidler. DreamTeacher: Pretraining image backbones with deep generative models. In IEEE ICCV, pages 16698–16708, 2023

  19. [27]

    Unleashing text-to-image diffusion models for visual perception

    Wenliang Zhao, Yongming Rao, Zuyan Liu, Benlin Liu, Jie Zhou, and Jiwen Lu. Unleashing text-to-image diffusion models for visual perception. In IEEE ICCV, pages 5729–5739, 2023

  20. [28]

    Text-image alignment for diffusion- based perception

    Neehar Kondapaneni, Markus Marks, Manuel Knott, Rogerio Guimaraes, and Pietro Perona. Text-image alignment for diffusion- based perception. In IEEE CVPR, pages 13883–13893, 2024

  21. [29]

    Diffusion models trained with large data are transferable visual models

    Guangkai Xu, Yongtao Ge, Mingyu Liu, Chengxiang Fan, Kangyang Xie, Zhiyue Zhao, Hao Chen, and Chunhua Shen. Diffusion models trained with large data are transferable visual models. arXiv preprint arXiv: 2403.06090, 2024

  22. [30]

    On discriminative vs

    Andrew Ng and Michael Jordan. On discriminative vs. generative clas- sifiers: A comparison of logistic regression and naive bayes. NeurIPS, 14, 2001

  23. [31]

    Microsoft COCO: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Per- ona, Deva Ramanan, Piotr Doll ´ar, and C Lawrence Zitnick. Microsoft COCO: Common objects in context. In ECCV, pages 740–755, 2014

  24. [32]

    Faster R-CNN: Towards real-time object detection with region proposal networks

    Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster R-CNN: Towards real-time object detection with region proposal networks. NeurIPS, 28, 2015

  25. [33]

    AnimeDiff: Customized image generation of anime characters using diffusion model

    Yuqi Jiang, Qiankun Liu, Dongdong Chen, Lu Yuan, and Ying Fu. AnimeDiff: Customized image generation of anime characters using diffusion model. IEEE TMM, pages 1–13, 2024

  26. [34]

    SGDM: An adaptive style-guided diffusion model for personalized text to image generation

    Yifei Xu, Xiaolong Xu, Honghao Gao, and Fu Xiao. SGDM: An adaptive style-guided diffusion model for personalized text to image generation. IEEE TMM, 26:9804–9813, 2024

  27. [35]

    SDEdit: Guided image synthesis and editing with stochastic differential equations

    Chenlin Meng, Yutong He, Yang Song, Jiaming Song, Jiajun Wu, Jun- Yan Zhu, and Stefano Ermon. SDEdit: Guided image synthesis and editing with stochastic differential equations. In ICLR, 2022

  28. [36]

    DiffFashion: Reference-based fashion design with structure-aware transfer by diffusion models

    Shidong Cao, Wenhao Chai, Shengyu Hao, Yanting Zhang, Hangyue Chen, and Gaoang Wang. DiffFashion: Reference-based fashion design with structure-aware transfer by diffusion models. IEEE TMM, 26:3962–3975, 2024

  29. [37]

    MMGInpainting: Multi-modality guided image inpainting based on diffusion models

    Cong Zhang, Wenxia Yang, Xin Li, and Huan Han. MMGInpainting: Multi-modality guided image inpainting based on diffusion models. IEEE TMM, 26:8811–8823, 2024

  30. [38]

    Video diffusion models

    Jonathan Ho, Tim Salimans, Alexey Gritsenko, William Chan, Moham- mad Norouzi, and David J Fleet. Video diffusion models. NeurIPS, 35:8633–8646, 2022

  31. [39]

    Conditional video diffusion network for fine-grained temporal sentence grounding

    Daizong Liu, Jiahao Zhu, Xiang Fang, Zeyu Xiong, Huan Wang, Renfu Li, and Pan Zhou. Conditional video diffusion network for fine-grained temporal sentence grounding. IEEE TMM, 26:5461–5476, 2024

  32. [40]

    TA2V: Text-audio guided video generation

    Minglu Zhao, Wenmin Wang, Tongbao Chen, Rui Zhang, and Ruochen Li. TA2V: Text-audio guided video generation. IEEE TMM, 26:7250– 7264, 2024

  33. [41]

    Imagi- naryNet: Learning object detectors without real images and annotations

    Minheng Ni, Zitong Huang, Kailai Feng, and Wangmeng Zuo. Imagi- naryNet: Learning object detectors without real images and annotations. In ICLR, 2023

  34. [42]

    Diffusion models and semi-supervised learners benefit mutually with few labels

    Zebin You, Yong Zhong, Fan Bao, Jiacheng Sun, Chongxuan Li, and Jun Zhu. Diffusion models and semi-supervised learners benefit mutually with few labels. NeurIPS, 36, 2023

  35. [43]

    Diversify your vision datasets with automatic diffusion-based augmentation

    Lisa Dunlap, Alyssa Umino, Han Zhang, Jiezhi Yang, Joseph E Gonzalez, and Trevor Darrell. Diversify your vision datasets with automatic diffusion-based augmentation. NeurIPS, 36, 2023

  36. [44]

    FreeMask: Synthetic images with dense annotations make stronger segmentation models

    Lihe Yang, Xiaogang Xu, Bingyi Kang, Yinghuan Shi, and Hengshuang Zhao. FreeMask: Synthetic images with dense annotations make stronger segmentation models. NeurIPS, 36, 2023

  37. [45]

    Diffusion models for open-vocabulary segmentation

    Laurynas Karazija, Iro Laina, Andrea Vedaldi, and Christian Rupprecht. Diffusion models for open-vocabulary segmentation. In ECCV, 2024

  38. [46]

    Mosaicfusion: Diffusion models as data aug- menters for large vocabulary instance segmentation

    Jiahao Xie, Wei Li, Xiangtai Li, Ziwei Liu, Yew Soon Ong, and Chen Change Loy. Mosaicfusion: Diffusion models as data aug- menters for large vocabulary instance segmentation. arXiv preprint arXiv:2309.13042, 2023

  39. [47]

    Gen2Det: Generate to detect

    Saksham Suri, Fanyi Xiao, Animesh Sinha, Sean Culatana, Raghu- raman Krishnamoorthi, Chenchen Zhu, and Abhinav Shrivastava. Gen2Det: Generate to detect. In IEEE CVPRW, 2024. PREPRINT 11

  40. [48]

    Open-vocabulary object segmentation with diffusion models

    Ziyi Li, Qinye Zhou, Xiaoyun Zhang, Ya Zhang, Yanfeng Wang, and Weidi Xie. Open-vocabulary object segmentation with diffusion models. In IEEE ICCV, pages 7667–7676, 2023

  41. [49]

    Do text-free diffusion models learn discriminative visual representations? In ECCV, 2024

    Soumik Mukhopadhyay, Matthew Gwilliam, Yosuke Yamaguchi, Vat- sal Agarwal, Namitha Padmanabhan, Archana Swaminathan, Tianyi Zhou, and Abhinav Shrivastava. Do text-free diffusion models learn discriminative visual representations? In ECCV, 2024

  42. [50]

    Bridging generative and discriminative models for unified visual perception with diffusion priors

    Shiyin Dong, Mingrui Zhu, Kun Cheng, Nannan Wang, and Xinbo Gao. Bridging generative and discriminative models for unified visual perception with diffusion priors. In IJCAI, pages 740–748, 2024

  43. [51]

    ECoDepth: Effective conditioning of diffusion models for monocular depth estimation

    Suraj Patni, Aradhye Agarwal, and Chetan Arora. ECoDepth: Effective conditioning of diffusion models for monocular depth estimation. In IEEE CVPR, pages 28285–28295, 2024

  44. [52]

    SegDiff: Image segmentation with diffusion probabilistic models

    Tomer Amit, Tal Shaharbany, Eliya Nachmani, and Lior Wolf. SegDiff: Image segmentation with diffusion probabilistic models. arXiv preprint arXiv:2112.00390, 2021

  45. [53]

    A generalist framework for panoptic segmentation of images and videos

    Ting Chen, Lala Li, Saurabh Saxena, Geoffrey Hinton, and David J Fleet. A generalist framework for panoptic segmentation of images and videos. In IEEE ICCV, pages 909–919, 2023

  46. [54]

    DiffusionDet: Diffusion model for object detection

    Shoufa Chen, Peize Sun, Yibing Song, and Ping Luo. DiffusionDet: Diffusion model for object detection. In IEEE ICCV , pages 19830– 19843, 2023

  47. [55]

    DiffusionDepth: Diffusion denoising approach for monocular depth estimation

    Yiqun Duan, Xianda Guo, and Zheng Zhu. DiffusionDepth: Diffusion denoising approach for monocular depth estimation. arXiv preprint arXiv:2303.05021, 2023

  48. [56]

    DDP: Diffusion model for dense visual prediction

    Yuanfeng Ji, Zhe Chen, Enze Xie, Lanqing Hong, Xihui Liu, Zhaoqiang Liu, Tong Lu, Zhenguo Li, and Ping Luo. DDP: Diffusion model for dense visual prediction. In IEEE ICCV, pages 21741–21752, 2023

  49. [57]

    Exploiting diffusion prior for generalizable dense prediction

    Hsin-Ying Lee, Hung-Yu Tseng, Hsin-Ying Lee, and Ming-Hsuan Yang. Exploiting diffusion prior for generalizable dense prediction. In IEEE CVPR, pages 7861–7871, 2024

  50. [58]

    Repurposing diffusion-based image generators for monocular depth estimation

    Bingxin Ke, Anton Obukhov, Shengyu Huang, Nando Metzger, Ro- drigo Caye Daudt, and Konrad Schindler. Repurposing diffusion-based image generators for monocular depth estimation. In IEEE CVPR , pages 9492–9502, 2024

  51. [59]

    DSIS- DPR: Structured instance segmentation and diffusion prior refinement for dental anatomy learning

    Xianyun Wang, Linhong Wang, Zhenchen Yang, Jiacong Zhou, Yuchen Zheng, Feng Chen, Richang Hong, Jun Yu, and Fan Yang. DSIS- DPR: Structured instance segmentation and diffusion prior refinement for dental anatomy learning. IEEE TMM, 2024

  52. [60]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. NeurIPS, 30, 2017

  53. [61]

    Diffuse attend and segment: Unsupervised zero-shot segmentation using stable diffusion

    Junjiao Tian, Lavisha Aggarwal, Andrea Colaco, Zsolt Kira, and Mar Gonzalez-Franco. Diffuse attend and segment: Unsupervised zero-shot segmentation using stable diffusion. In IEEE CVPR, pages 3554–3563, 2024

  54. [62]

    LD-ZNet: A latent diffusion approach for text-based image segmentation

    Koutilya Pnvr, Bharat Singh, Pallabi Ghosh, Behjat Siddiquie, and David Jacobs. LD-ZNet: A latent diffusion approach for text-based image segmentation. In IEEE ICCV, pages 4157–4168, 2023

  55. [63]

    Diffusion model is secretly a training-free open vocabulary semantic segmenter

    Jinglong Wang, Xiawei Li, Jing Zhang, Qingyuan Xu, Qin Zhou, Qian Yu, Lu Sheng, and Dong Xu. Diffusion model is secretly a training-free open vocabulary semantic segmenter. arXiv preprint arXiv:2309.02773, 2023

  56. [64]

    From text to mask: Localizing entities using the attention of text-to-image diffusion models

    Changming Xiao, Qi Yang, Feng Zhou, and Changshui Zhang. From text to mask: Localizing entities using the attention of text-to-image diffusion models. Neurocomputing, 610:128437, 2024

  57. [65]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. In IEEE ICCV, pages 4195–4205, 2023

  58. [66]

    Score-based generative classifiers

    Roland S Zimmermann, Lukas Schott, Yang Song, Benjamin Adric Dunn, and David A Klindt. Score-based generative classifiers. In NeurIPS Workshop on Deep Generative Models and Downstream Applications, 2021

  59. [67]

    Robust classification via a single diffusion model

    Huanran Chen, Yinpeng Dong, Zhengyi Wang, Xiao Yang, Chengqi Duan, Hang Su, and Jun Zhu. Robust classification via a single diffusion model. In ICML, 2024

  60. [68]

    Your diffusion model is secretly a certifiably robust classifier

    Huanran Chen, Yinpeng Dong, Shitong Shao, Zhongkai Hao, Xiao Yang, Hang Su, and Jun Zhu. Your diffusion model is secretly a certifiably robust classifier. arXiv preprint arXiv: 2402.02316 , 2024

  61. [69]

    Are diffusion models vision-and-language reasoners? NeurIPS, 36, 2023

    Benno Krojer, Elinor Poole-Dayan, Vikram V oleti, Christopher Pal, and Siva Reddy. Are diffusion models vision-and-language reasoners? NeurIPS, 36, 2023

  62. [70]

    SelfEval: Leveraging the discriminative nature of generative models for evaluation

    Sai Saketh Rambhatla and Ishan Misra. SelfEval: Leveraging the discriminative nature of generative models for evaluation. arXiv preprint arXiv: 2311.10708 , 2023

  63. [71]

    Generative visual manipulation on the natural image manifold

    Jun-Yan Zhu, Philipp Kr ¨ahenb¨uhl, Eli Shechtman, and Alexei A Efros. Generative visual manipulation on the natural image manifold. In ECCV, pages 597–613, 2016

  64. [72]

    GAN inversion: A survey

    Weihao Xia, Yulun Zhang, Yujiu Yang, Jing-Hao Xue, Bolei Zhou, and Ming-Hsuan Yang. GAN inversion: A survey. IEEE TPAMI , 45(3):3121–3138, 2022

  65. [73]

    Plug- and-play diffusion features for text-driven image-to-image translation

    Narek Tumanyan, Michal Geyer, Shai Bagon, and Tali Dekel. Plug- and-play diffusion features for text-driven image-to-image translation. In IEEE CVPR, pages 1921–1930, 2023

  66. [74]

    MasaCtrl: Tuning-free mutual self-attention control for consistent image synthesis and editing

    Mingdeng Cao, Xintao Wang, Zhongang Qi, Ying Shan, Xiaohu Qie, and Yinqiang Zheng. MasaCtrl: Tuning-free mutual self-attention control for consistent image synthesis and editing. In IEEE ICCV , pages 22560–22570, 2023

  67. [75]

    Inverting the generator of a generative adversarial network

    Antonia Creswell and Anil Anthony Bharath. Inverting the generator of a generative adversarial network. IEEE TNNLS , 30(7):1967–1974, 2018

  68. [76]

    Image2StyleGAN: How to embed images into the StyleGAN latent space? In IEEE ICCV, pages 4432–4441, 2019

    Rameen Abdal, Yipeng Qin, and Peter Wonka. Image2StyleGAN: How to embed images into the StyleGAN latent space? In IEEE ICCV, pages 4432–4441, 2019

  69. [77]

    Improved StyleGAN embedding: Where are the good latents? arXiv preprint arXiv:2012.09036 , 2020

    Peihao Zhu, Rameen Abdal, Yipeng Qin, John Femiani, and Peter Wonka. Improved StyleGAN embedding: Where are the good latents? arXiv preprint arXiv:2012.09036 , 2020

  70. [78]

    Encoding in style: A StyleGAN encoder for image-to-image translation

    Elad Richardson, Yuval Alaluf, Or Patashnik, Yotam Nitzan, Yaniv Azar, Stav Shapiro, and Daniel Cohen-Or. Encoding in style: A StyleGAN encoder for image-to-image translation. In IEEE CVPR , pages 2287–2296, 2021

  71. [79]

    Designing an encoder for StyleGAN image manipulation

    Omer Tov, Yuval Alaluf, Yotam Nitzan, Or Patashnik, and Daniel Cohen-Or. Designing an encoder for StyleGAN image manipulation. ACM TOG, 40(4):1–14, 2021

  72. [80]

    HyperStyle: StyleGAN inversion with hypernetworks for real image editing

    Yuval Alaluf, Omer Tov, Ron Mokady, Rinon Gal, and Amit Bermano. HyperStyle: StyleGAN inversion with hypernetworks for real image editing. In IEEE CVPR, pages 18511–18521, 2022

  73. [81]

    Unsupervised image-to- image translation via pre-trained StyleGAN2 network

    Jialu Huang, Jing Liao, and Sam Kwong. Unsupervised image-to- image translation via pre-trained StyleGAN2 network. IEEE TMM , 24:1435–1448, 2022

  74. [82]

    In-domain GAN inversion for real image editing

    Jiapeng Zhu, Yujun Shen, Deli Zhao, and Bolei Zhou. In-domain GAN inversion for real image editing. In ECCV, pages 592–608, 2020

  75. [83]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. In ICLR, 2021

  76. [84]

    EDICT: Exact diffusion inversion via coupled transformations

    Bram Wallace, Akash Gokul, and Nikhil Naik. EDICT: Exact diffusion inversion via coupled transformations. In IEEE CVPR, pages 22532– 22541, 2023

  77. [85]

    Exact diffusion inver- sion via bi-directional integration approximation

    Guoqiang Zhang and W Bastiaan Kleijn. Exact diffusion inver- sion via bi-directional integration approximation. arXiv preprint arXiv:2307.10829, 2023

  78. [86]

    Null-text inversion for editing real images using guided diffusion models

    Ron Mokady, Amir Hertz, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Null-text inversion for editing real images using guided diffusion models. In IEEE CVPR, pages 6038–6047, 2023

  79. [87]

    Negative-prompt inversion: Fast image inversion for editing with text- guided diffusion models

    Daiki Miyake, Akihiro Iohara, Yu Saito, and Toshiyuki Tanaka. Negative-prompt inversion: Fast image inversion for editing with text- guided diffusion models. arXiv preprint arXiv:2305.16807 , 2023

  80. [88]

    An image is worth one word: Personalizing text-to-image generation using textual inversion

    Rinon Gal, Yuval Alaluf, Yuval Atzmon, Or Patashnik, Amit Haim Bermano, Gal Chechik, and Daniel Cohen-or. An image is worth one word: Personalizing text-to-image generation using textual inversion. In ICLR, 2023

  81. [89]

    De-diffusion makes text a strong cross-modal interface

    Chen Wei, Chenxi Liu, Siyuan Qiao, Zhishuai Zhang, Alan Yuille, and Jiahui Yu. De-diffusion makes text a strong cross-modal interface. In IEEE CVPR, pages 13492–13503, 2024

  82. [90]

    Prompting hard or hardly prompting: Prompt inversion for text-to- image diffusion models

    Shweta Mahajan, Tanzila Rahman, Kwang Moo Yi, and Leonid Sigal. Prompting hard or hardly prompting: Prompt inversion for text-to- image diffusion models. In IEEE CVPR, pages 6808–6817, 2024

  83. [91]

    Conditional generative adversarial nets

    Mehdi Mirza and Simon Osindero. Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784 , 2014

  84. [92]

    Bernstein, Alexander C

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael S. Bernstein, Alexander C. Berg, and Li Fei-Fei. Imagenet large scale visual recognition challenge. IJCV, 115:211–252, 2015

  85. [93]

    Frido: Feature pyramid diffusion for complex scene image synthesis

    Wan-Cyuan Fan, Yen-Chun Chen, DongDong Chen, Yu Cheng, Lu Yuan, and Yu-Chiang Frank Wang. Frido: Feature pyramid diffusion for complex scene image synthesis. In AAAI, volume 37, pages 579– 587, 2023

  86. [94]

    LayoutDiffuse: Adapting foundational diffusion models for layout-to-image generation

    Jiaxin Cheng, Xiao Liang, Xingjian Shi, Tong He, Tianjun Xiao, and Mu Li. LayoutDiffuse: Adapting foundational diffusion models for layout-to-image generation. arXiv preprint arXiv: 2302.08908 , 2023

  87. [95]

    LayoutDiffusion: Controllable diffusion model for layout-to-image generation

    Guangcong Zheng, Xianpan Zhou, Xuewei Li, Zhongang Qi, Ying Shan, and Xi Li. LayoutDiffusion: Controllable diffusion model for layout-to-image generation. In IEEE CVPR, pages 22490–22499, 2023

  88. [96]

    COCO-Stuff: Thing and stuff classes in context

    Holger Caesar, Jasper Uijlings, and Vittorio Ferrari. COCO-Stuff: Thing and stuff classes in context. In IEEE CVPR, pages 1209–1218, 2018. PREPRINT 12

  89. [97]

    ReCo: Region-controlled text-to-image generation

    Zhengyuan Yang, Jianfeng Wang, Zhe Gan, Linjie Li, Kevin Lin, Chenfei Wu, Nan Duan, Zicheng Liu, Ce Liu, Michael Zeng, and Lijuan Wang. ReCo: Region-controlled text-to-image generation. In IEEE CVPR, pages 14246–14255, 2023

  90. [98]

    GLIGEN: Open-set grounded text-to-image generation

    Yuheng Li, Haotian Liu, Qingyang Wu, Fangzhou Mu, Jianwei Yang, Jianfeng Gao, Chunyuan Li, and Yong Jae Lee. GLIGEN: Open-set grounded text-to-image generation. In IEEE CVPR , pages 22511– 22521, 2023

  91. [99]

    U-Net: Con- volutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-Net: Con- volutional networks for biomedical image segmentation. In MICCAI, pages 234–241, 2015

  92. [100]

    Decoupled weight decay regulariza- tion

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regulariza- tion. In ICLR, 2019

  93. [101]

    End-to-end object detection with transformers

    Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko. End-to-end object detection with transformers. In ECCV, pages 213–229, 2020

  94. [102]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In IEEE CVPR , pages 770– 778, 2016

  95. [103]

    Feature pyramid networks for object detection

    Tsung-Yi Lin, Piotr Doll ´ar, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. Feature pyramid networks for object detection. In IEEE CVPR, pages 2117–2125, 2017

  96. [104]

    Fully convolutional instance-aware semantic segmentation

    Yi Li, Haozhi Qi, Jifeng Dai, Xiangyang Ji, and Yichen Wei. Fully convolutional instance-aware semantic segmentation. In IEEE CVPR , pages 2359–2367, 2017

  97. [105]

    Detectron2

    Yuxin Wu, Alexander Kirillov, Francisco Massa, Wan-Yen Lo, and Ross Girshick. Detectron2. https://github.com/facebookresearch/ detectron2, 2019

  98. [106]

    Adversarial attacks on image generation with made- up words

    Rapha ¨el Milli`ere. Adversarial attacks on image generation with made- up words. arXiv preprint arXiv:2208.04135 , 2022

  99. [107]

    A pilot study of query- free adversarial attack against stable diffusion

    Haomin Zhuang, Yihua Zhang, and Sijia Liu. A pilot study of query- free adversarial attack against stable diffusion. In IEEE CVPRW, pages 2385–2392, 2023

  100. [108]

    Evaluat- ing the robustness of text-to-image diffusion models against real-world attacks

    Hongcheng Gao, Hao Zhang, Yinpeng Dong, and Zhijie Deng. Evaluat- ing the robustness of text-to-image diffusion models against real-world attacks. arXiv preprint arXiv:2306.13103 , 2023

  101. [109]

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

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weis- senborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition ...

  102. [110]

    ImageNet classification with deep convolutional neural networks

    Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. ImageNet classification with deep convolutional neural networks. NeurIPS, 25, 2012

  103. [111]

    Girshick, Piotr Doll ´ar, Z

    Saining Xie, Ross B. Girshick, Piotr Doll ´ar, Z. Tu, and Kaiming He. Aggregated residual transformations for deep neural networks. In IEEE CVPR, 2016

  104. [112]

    Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and B. Guo. Swin transformer: Hierarchical vision transformer using shifted windows. In IEEE ICCV, 2021

  105. [113]

    Rethinking the inception architecture for computer vision

    Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jon Shlens, and Zbigniew Wojna. Rethinking the inception architecture for computer vision. In IEEE CVPR, pages 2818–2826, 2016

  106. [114]

    GANs trained by a two time-scale update rule converge to a local nash equilibrium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. GANs trained by a two time-scale update rule converge to a local nash equilibrium. NeurIPS, 30, 2017

  107. [115]

    Improved techniques for training GANs

    Tim Salimans, Ian Goodfellow, Wojciech Zaremba, Vicki Cheung, Alec Radford, and Xi Chen. Improved techniques for training GANs. NeurIPS, 29, 2016

  108. [116]

    Improved precision and recall metric for assessing generative models

    Tuomas Kynk ¨a¨anniemi, Tero Karras, Samuli Laine, Jaakko Lehtinen, and Timo Aila. Improved precision and recall metric for assessing generative models. NeurIPS, 32, 2019

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.