Pith. sign in

REVIEW 4 major objections 8 minor 122 references

From Noise to Nuance: Advances in Deep Generative Image Models

T0 review · 4 major / 8 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This survey argues that deep generative image models shifted decisively from GANs to diffusion and transformer architectures after 2021, driven by latent-space efficiency and control, with resource-conscious and interpretable systems…

desk verdict A broad but unreliable survey of deep generative image models; the equation errors and copy-paste problems mean it should not be trusted as a reference until extensively revised. read the letter →

arxiv 2412.09656 v1 pith:ZO3RBTBD submitted 2024-12-12 cs.CV cs.AI

Benji Peng , Chia Xin Liang , Ziqian Bi , Ming Liu , Yichao Zhang , Tianyang Wang , Keyu Chen , Xinyuan Song
show 1 more author
Pohsun Feng
This is my paper · ORCID
classification cs.CVcs.AI
keywords diffusionmodelsgenerativeimagelatentconsistencyvisiontransformersparameter-efficientfine-tuningtext-to-imagegenerationsurvey
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

This survey tries to establish that deep learning image generation changed course around 2021, with generative adversarial networks giving way to diffusion and transformer architectures. It claims the key drivers are latent-space diffusion, cross-attention conditioning, and parameter-efficient training methods such as low-rank adaptation and quantization, plus distillation techniques that cut inference to a few steps. It also argues that control mechanisms improved precision and customization, turning text-to-image systems into practical tools. The payoff is a structured map of where the field is and where it is blocked: computational scalability, quality-speed trade-offs, and interpretable, resource-conscious systems remain open.

What carries the argument

The paper's organizing object is the latent-space diffusion process with cross-attention conditioning: an autoencoder compresses images into a latent, a denoising network operates there, and cross-attention lets text or other signals guide each step. For efficiency, the other central object is the consistency function, which collapses the probability-flow trajectory so that every point on it maps to the same clean image, enabling one-step or few-step generation when learned or distilled. Low-rank adapters and quantization are the supporting machinery for deploying these models cheaply.

What would settle it

A reproducible audit that compares the SDXL-Lightning discriminator equations and the consistency-model loss with the original released implementations would settle the fidelity question: any mismatch in the loss or discriminator definitions would show that the survey's summaries of those models are unreliable. For the historical claim itself, an independent benchmark in which few-step distilled models consistently beat multi-step diffusion on every standard quality metric would weaken the paper's claim that quality-speed trade-offs remain a critical challenge.

Watch

Extended reading notes

Core claim

The central claim is that since 2021 image generation has shifted from GANs to diffusion models, and then to compute-efficient variants: latent diffusion in a compressed space, transformer-based diffusion, autoregressive and masked token transformers, and consistency models that map any noisy point on a trajectory to its clean origin. The paper further claims that conditioning mechanisms, including cross-attention on text, structural control inputs, and style-transfer adapters, turned these models into controllable tools. Taken together, the survey's account says the field's progress is best understood as a race between generative quality and computational cost, and that the remaining frontier is resource-conscious, interpretable architectures.

Load-bearing premise

The survey's narrative is trustworthy only if its reproductions of primary-source equations and model descriptions are faithful, and that fidelity is not independently verified.

Editorial extensions

If this is right

  • If the survey's historical account is right, future image-generation research will keep operating in learned latent spaces, making the quality of the autoencoder a first-order performance factor.
  • If consistency and distillation methods truly preserve quality at few steps, interactive and real-time image editing on consumer hardware becomes feasible.
  • If the claimed efficiency techniques work as described, quantization and low-rank adaptation will become standard deployment practice for large text-to-image models.
  • If multi-component prompt comprehension remains hard, evaluation will need to move toward compositional benchmarks rather than single-prompt quality scores.

Reading between the lines

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

  • An unstated consequence of the survey's framing is that further advances in generation quality are likely to come from better learned autoencoders and noise schedules rather than from scaling the denoising network alone.
  • Because the survey presents equations without code or independent verification, a reader should treat its reproduced formulas as pointers to the original papers rather than as authoritative derivations.
  • A testable extension the survey leaves implicit: a systematic comparison of consistency losses with and without the stop-gradient rule across data scales could reveal whether the consistency property itself, rather than the time-sampling schedule, drives the reported efficiency.
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 / 8 minor

Summary. This paper is a survey of deep generative image models, tracing the transition from GANs to diffusion- and transformer-based architectures. It covers DDPMs, latent diffusion, Stable Diffusion variants (SD1.x–SD3, SDXL-Turbo, SDXL-Lightning), consistency models, Imagen, DALL-E series, and practical topics such as quantization, PEFT, inpainting/outpainting, ControlNet, style transfer, and evaluation metrics. The authors argue that the field has undergone a paradigm shift since 2021 and that, despite advances in quality and efficiency, resource constraints and interpretability remain open problems. The paper concludes by outlining future research directions including neural architecture optimization and explainable generation frameworks.

Significance. As a survey, the paper's value depends on the fidelity of its technical summaries and the comprehensiveness of its coverage. The standard DDPM equations (Eqs. 2–8) and the general LDM framework (Eqs. 10–18) are reproduced correctly, and the paper covers an impressively broad set of topics, including efficiency methods, control, and evaluation. However, the survey offers no original experiments, analysis, or falsifiable predictions, and its reliability is undermined by a copied first-person fragment, a duplicated Consistency Models section, and at least two load-bearing equation inaccuracies. If corrected, the survey could serve as a useful entry point, but in its current form its technical descriptions cannot be fully trusted.

major comments (4)
  1. [Section II.B.2.d, Eq. (19)] The LDM training objective is written as E[||ε − ε_θ(z_t, t)||²], omitting the conditioning input c that appears in the same paper's Eqs. (13), (17), and (18) and in the original LDM objective (Rombach et al., Eq. 3). Because conditioning is the mechanism that makes text-to-image synthesis possible, this misstates the core model and is internally inconsistent. The equation should read E[||ε − ε_θ(z_t, t, c)||²] (or, in original notation, with the text-encoder embedding τ_θ(y) as the conditioning input).
  2. [Section II.C, Eq. (24)] The consistency loss as written evaluates f_θ on the same input x at two times t1 and t2, which does not reflect the consistency-model training procedure. In Song et al. [49], the loss compares the model outputs at two points x_t and x_t′ that are noisy versions of the same data sample along the probability-flow ODE, and it uses a target network (e.g., an EMA of f_θ). The current formula also omits the target network and the sampling of noisy points, so it does not enforce the self-consistency property described in Eq. (23). Please replace it with the actual consistency distillation or consistency training objective and state the appropriate sampling procedure for x, t1, and t2.
  3. [Section II.B.3.b] The sentence 'Our models are available as both LoRA and full UNet weights' is a first-person statement that belongs to the SDXL-Lightning authors, not to this survey. It appears without quotation marks or attribution, which is a verbatim copying problem. This casts doubt on the reliability of the surrounding technical content (Eqs. 21–22); the authors should rewrite all such material in third-person form and verify every equation against the source.
  4. [Section II.B.4 and Section II.C] Consistency Models are presented twice: first as a short subsection under 'Diffusion Model Breakthroughs' (II.B.4) and then as a full section 'Consistency Models for Efficient Image Generation' (II.C). The two treatments are inconsistent: II.B.4 defines the consistency function via the probability-flow ODE, while II.C.1 uses an arbitrary-image formulation, and Eq. (24) in II.C.2 conflicts with the II.B.4 definition. This duplication suggests an unfinished editorial pass and should be resolved by merging the two accounts into one accurate presentation.
minor comments (8)
  1. [Abstract / Section I] The abstract states that the field has undergone a 'paradigm shift since 2021,' but the paper does not provide a dated timeline or evidence that 2021 is the correct boundary; please support this claim with a brief historical analysis or soften the phrasing.
  2. [References] Several references are self-citations to the authors' prior work (e.g., [20], [30], [47], [98], [102], [117]) and are used for efficiency, evaluation, and safety claims; please identify these as the authors' own work or replace them with independent sources.
  3. [Section II.B.3.b] The SDXL-Lightning paragraph does not cite [44] in the text; the reference appears only in the bibliography. Please add the citation and ensure that all equations and claims in that subsection are attributed to the source.
  4. [Section I.A] There are grammatical slips, e.g., 'Image generation models has been through significant changes' and 'These foundation models marks a significant milestone'; they should be corrected.
  5. [Section V.A] The sentence '[98] also proposed proposed CLIP-based metrics' contains a duplicated word 'proposed'.
  6. [Section I] The survey does not describe its literature selection methodology, so a reader cannot assess completeness or bias; I recommend adding a short scope and selection-criteria paragraph.
  7. [Section II.C.2] The claim that the time sampling strategy T 'focus[es] on regions where the model's outputs are most sensitive to noise' is vague and unreferenced; please define or cite the weighting scheme.
  8. [Tables I and II] Tables I and II summarize model series but do not include citations to the primary sources in the table captions or rows; please add them.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity; the survey's narrative rests on external literature, with only minor non-load-bearing self-citations.

full rationale

This is a survey paper, not a derivation paper. Its central claim—that image generation has undergone a paradigm shift since 2021 and that efficiency and interpretability challenges remain—is a synthesis of externally cited results (e.g., DDPM [3], LDM [16], SDXL [39], Consistency Models [49]), not a prediction derived from fitted parameters or from the authors' own prior work. I checked the seven self-citations ([20], [23], [30], [47], [98], [102], [117]): they support auxiliary points such as efficient architectures, UI/UX applications, LLM safety, multimodal embeddings, and CLIP-based metrics. None is load-bearing for the paradigm-shift narrative; removing them would not collapse the survey's main argument, which is independently supported by primary external sources. There is no uniqueness theorem imported from the authors' prior work, no ansatz smuggled in via self-citation, and no fitted quantity renamed as a prediction. The in-scope textual anomalies are accuracy and fidelity issues rather than circularity: Eq. (19) omits the conditioning c that appears in Eq. (18), Eq. (24) simplifies the Consistency Models objective away from the probability-flow ODE sampling used in [49], and Section II.B.3.b contains the verbatim first-person sentence 'Our models are available as both LoRA and full UNet weights,' indicating copied source text. These problems undermine the survey's reliability as a reproduction of primary equations, but they do not make any claim reduce to its own inputs. Accordingly, the circularity score is 1: minor self-citation exists, but the central claim has independent content and no step is circular by construction.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

The survey introduces no new axioms, parameters, or entities. It relies entirely on the correctness of external sources. No free parameters or invented entities are needed.

assumptions (2)
  • domain assumption The reproduced DDPM and LDM equations (Eqs. 2-20) accurately represent the original formulations of Ho et al. and Rombach et al.
    The survey's explanations of diffusion and latent diffusion rest on these equations; the authors do not derive or verify them but present them as established.
  • domain assumption The descriptions of model capabilities (SDXL-Lightning, Consistency Models, DALL-E series) faithfully reflect the claims of the cited source papers.
    The survey's synthesis of the field is valid only if its representation of primary sources is accurate. The copied 'Our models' passage in Section II.B.3.b undermines confidence in this fidelity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of From Noise to Nuance: Advances in Deep Generative Image Models." pith.science (2026). https://pith.science/paper/ZO3RBTBD

@misc{pith2026241209656,
  author       = {Pith},
  title        = {Pith review of: From Noise to Nuance: Advances in Deep Generative Image Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZO3RBTBD}},
  note         = {Machine review of arXiv:2412.09656}
}
read the original abstract

Deep learning-based image generation has undergone a paradigm shift since 2021, marked by fundamental architectural breakthroughs and computational innovations. Through reviewing architectural innovations and empirical results, this paper analyzes the transition from traditional generative methods to advanced architectures, with focus on compute-efficient diffusion models and vision transformer architectures. We examine how recent developments in Stable Diffusion, DALL-E, and consistency models have redefined the capabilities and performance boundaries of image synthesis, while addressing persistent challenges in efficiency and quality. Our analysis focuses on the evolution of latent space representations, cross-attention mechanisms, and parameter-efficient training methodologies that enable accelerated inference under resource constraints. While more efficient training methods enable faster inference, advanced control mechanisms like ControlNet and regional attention systems have simultaneously improved generation precision and content customization. We investigate how enhanced multi-modal understanding and zero-shot generation capabilities are reshaping practical applications across industries. Our analysis demonstrates that despite remarkable advances in generation quality and computational efficiency, critical challenges remain in developing resource-conscious architectures and interpretable generation systems for industrial applications. The paper concludes by mapping promising research directions, including neural architecture optimization and explainable generation frameworks.

Figures

Figures reproduced from arXiv: 2412.09656 by the authors.

Figure 1
Figure 1. Stable Diffusion Model structure for training and in [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

122 extracted references · 34 canonical work pages

  1. [49]

    Consis tency models,

    Y . Song, P . Dhariwal, M. Chen, and I. Sutskever, “Consis tency models,” arXiv preprint arXiv:2303.01469 , 2023

    arXiv 2023

  2. [1]

    Generative adversar ial net- works,

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Ward e-Farley, S. Ozair, A. Courville, and Y . Bengio, “Generative adversar ial net- works,” Communications of the ACM , vol. 63, no. 11, pp. 139–144, 2020

    2020

  3. [2]

    Unsupervised repr esentation learning with deep convolutional generative adversarial n etworks,

    A. Radford, L. Metz, and S. Chintala, “Unsupervised repr esentation learning with deep convolutional generative adversarial n etworks,” in International Conference on Learning Representations , 2016

    2016

  4. [3]

    Denoising diffusion proba bilistic mod- els,

    J. Ho, A. Jain, and P . Abbeel, “Denoising diffusion proba bilistic mod- els,” in Advances in Neural Information Processing Systems , pp. 6840– 6851, 2020

    2020

  5. [4]

    Score-based generative modeling through sto chastic differential equations,

    Y . Song, J. Sohl-Dickstein, D. P . Kingma, A. Kumar, S. Erm on, and B. Poole, “Score-based generative modeling through sto chastic differential equations,” arXiv preprint arXiv:2011.13456 , 2020

    arXiv 2011

  6. [5]

    Diffusion models beat gans on image synthesis,

    P . Dhariwal and A. Nichol, “Diffusion models beat gans on image synthesis,” Advances in neural information processing systems , vol. 34, pp. 8780–8794, 2021

    2021

  7. [6]

    Imagenet: A large-scale hierarchical image database,

    J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei , “Imagenet: A large-scale hierarchical image database,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition , pp. 248–255, IEEE, 2009

    2009

  8. [7]

    Laion- 400m: Open dataset of clip-filtered 400 million image-text pairs,

    C. Schuhmann, R. V encu, R. Beaumont, R. Kaczmarczyk, C. M ullis, A. Katta, T. Coombes, J. Jitsev, and A. Komatsuzaki, “Laion- 400m: Open dataset of clip-filtered 400 million image-text pairs, ” arXiv preprint arXiv:2111.02114, 2021

    arXiv 2021

Show all 122 references
  1. [8]

    In-datacenter performance analysis of a tensor processing unit,

    N. P . Jouppi, C. Y oung, N. Patil, D. Patterson, G. Agrawal , R. Bajwa, S. Bates, S. Bhatia, N. Boden, A. Borchers, et al. , “In-datacenter performance analysis of a tensor processing unit,” in Proceedings of the 44th annual international symposium on computer archit ecture, ...

    2017

  2. [9]

    Scaling l aws for neural language models,

    J. Kaplan, S. McCandlish, T. Henighan, T. B. Brown, B. Che ss, R. Child, S. Gray, A. Radford, J. Wu, and D. Amodei, “Scaling l aws for neural language models,” arXiv preprint arXiv:2001.08361 , 2020

    2001 arXiv

  3. [10]

    Megatron-lm: Training multi-billion parameter lan guage models using model parallelism,

    M. Shoeybi, M. Patwary, R. Puri, P . LeGresley, J. Casper , and B. Catan- zaro, “Megatron-lm: Training multi-billion parameter lan guage models using model parallelism,” arXiv preprint arXiv:1909.08053 , 2019

    1909 arXiv

  4. [11]

    Scaling language models: Methods, analysis & insights from training gopher,

    J. W. Rae, S. Borgeaud, T. Cai, K. Millican, J. Hoffmann, F. Song, J. Aslanides, S. Henderson, R. Ring, S. Y oung, et al., “Scaling language models: Methods, analysis & insights from training gopher, ” arXiv preprint arXiv:2112.11446, 2021

    2021 arXiv

  5. [12]

    Attention is all you need,

    A. V aswani, “Attention is all you need,” Advances in Neural Informa- tion Processing Systems , 2017

    2017

  6. [13]

    H ierarchi- cal text-conditional image generation with clip latents,

    A. Ramesh, P . Dhariwal, A. Nichol, C. Chu, and M. Chen, “H ierarchi- cal text-conditional image generation with clip latents,” arXiv preprint arXiv:2204.06125, vol. 1, no. 2, p. 3, 2022

    2022 arXiv

  7. [14]

    Photorealistic text-to-image diffusion models with dee p lan- guage understanding,

    C. Saharia, W. Chan, S. Saxena, L. Li, J. Whang, E. L. Dent on, K. Ghasemipour, R. Gontijo Lopes, B. Karagol Ayan, T. Salima ns, et al. , “Photorealistic text-to-image diffusion models with dee p lan- guage understanding,” Advances in neural information processing systems, vol...

    2022

  8. [15]

    Zero-shot text-to-image generation,

    A. Ramesh, M. Pavlov, G. Goh, S. Gray, C. V oss, A. Radford , M. Chen, and I. Sutskever, “Zero-shot text-to-image generation,” i n International conference on machine learning , pp. 8821–8831, Pmlr, 2021

    2021

  9. [16]

    High-resolution image synthesis with latent diffusion mo dels,

    R. Rombach, A. Blattmann, D. Lorenz, P . Esser, and B. Omm er, “High-resolution image synthesis with latent diffusion mo dels,” in Proceedings of the IEEE/CVF conference on computer vision a nd pattern recognition, pp. 10684–10695, 2022

    2022

  10. [17]

    Improved denoising diffu sion prob- abilistic models,

    A. Q. Nichol and P . Dhariwal, “Improved denoising diffu sion prob- abilistic models,” in International conference on machine learning , pp. 8162–8171, PMLR, 2021

    2021

  11. [18]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Ag arwal, G. Sastry, A. Askell, P . Mishkin, J. Clark, et al., “Learning transferable visual models from natural language supervision,” in International conference on machine learning , pp. 8748–8763, PMLR, 2021

    2021

  12. [19]

    Alias-free generative adversarial networks,

    T. Karras, M. Aittala, S. Laine, E. H¨ ark¨ onen, J. Hells ten, J. Lehtinen, and T. Aila, “Alias-free generative adversarial networks, ” Advances in neural information processing systems , vol. 34, pp. 852–863, 2021

    2021

  13. [20]

    Surveying the mllm landscape: A meta- review of current surveys,

    M. Li, K. Chen, Z. Bi, M. Liu, B. Peng, Q. Niu, J. Liu, J. Wan g, S. Zhang, X. Pan, et al. , “Surveying the mllm landscape: A meta- review of current surveys,” arXiv preprint arXiv:2409.18991 , 2024

    2024

  14. [21]

    A survey of accelerator architectures for deep neural networks,

    Y . Chen, Y . Xie, L. Song, F. Chen, and T. Tang, “A survey of accelerator architectures for deep neural networks,” Engineering, vol. 6, no. 3, pp. 264–274, 2020

    2020

  15. [22]

    The age of generative ai and ai-generated everything,

    H. Du, D. Niyato, J. Kang, Z. Xiong, P . Zhang, S. Cui, X. Sh en, S. Mao, Z. Han, A. Jamalipour, et al. , “The age of generative ai and ai-generated everything,” IEEE Network , 2024

    2024

  16. [23]

    Llms and diffusion models in ui/ux: Advancing human-computer interaction and design,

    L. Sun, M. Qin, and B. Peng, “Llms and diffusion models in ui/ux: Advancing human-computer interaction and design,” OSF Preprints , Oct 2024

    2024

  17. [24]

    Lightweight g enerative adversarial networks for text-guided image manipulation,

    B. Li, X. Qi, P . Torr, and T. Lukasiewicz, “Lightweight g enerative adversarial networks for text-guided image manipulation, ” Advances in Neural Information Processing Systems , vol. 33, pp. 22020–22031, 2020

    2020

  18. [25]

    Dpm-sol ver: A fast ode solver for diffusion probabilistic model sampling in around 10 steps,

    C. Lu, Y . Zhou, F. Bao, J. Chen, C. Li, and J. Zhu, “Dpm-sol ver: A fast ode solver for diffusion probabilistic model sampling in around 10 steps,” Advances in Neural Information Processing Systems , vol. 35, pp. 5775–5787, 2022

    2022

  19. [26]

    The emergence of deepfake technology: A review,

    M. Westerlund, “The emergence of deepfake technology: A review,” Technology innovation management review , vol. 9, no. 11, 2019

    2019

  20. [27]

    The cat and mouse game: The ongoing arms race between diffusion models and det ection methods,

    L. Laurier, A. Giulietta, A. Octavia, and M. Cleti, “The cat and mouse game: The ongoing arms race between diffusion models and det ection methods,” arXiv preprint arXiv:2410.18866 , 2024

    2024 arXiv

  21. [28]

    Easily a ccessible text-to-image generation amplifies demographic stereotyp es at large scale,

    F. Bianchi, P . Kalluri, E. Durmus, F. Ladhak, M. Cheng, D . Nozza, T. Hashimoto, D. Jurafsky, J. Zou, and A. Caliskan, “Easily a ccessible text-to-image generation amplifies demographic stereotyp es at large scale,” in Proceedings of the 2023 ACM Conference on Fairness, Accoun...

    2023

  22. [29]

    The creativity of text-to-image gen eration,

    J. Oppenlaender, “The creativity of text-to-image gen eration,” in Pro- ceedings of the 25th international academic mindtrek confe rence, pp. 192–202, 2022

    2022

  23. [30]

    Jailbreaking and mitigation of vuln erabilities in large language models,

    B. Peng, Z. Bi, Q. Niu, M. Liu, P . Feng, T. Wang, L. K. Y an, Y . Wen, Y . Zhang, and C. H. Yin, “Jailbreaking and mitigation of vuln erabilities in large language models,” arXiv preprint arXiv:2410.15236 , 2024

    2024 arXiv

  24. [31]

    Scalable diffusion models with t ransformers,

    W. Peebles and S. Xie, “Scalable diffusion models with t ransformers,” in Proceedings of the IEEE/CVF International Conference on Co m- puter Vision, pp. 4195–4205, 2023

    2023

  25. [32]

    Sd-dit: Unleashing the power of self-supervised discrimi nation in diffusion transformer,

    R. Zhu, Y . Pan, Y . Li, T. Y ao, Z. Sun, T. Mei, and C. W. Chen, “Sd-dit: Unleashing the power of self-supervised discrimi nation in diffusion transformer,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 8435–8445, 2024

    2024

  26. [33]

    Scaling autoregres- sive models for content-rich text-to-image generation,

    J. Y u, Y . Xu, J. Y . Koh, T. Luong, G. Baid, Z. Wang, V . V a- sudevan, A. Ku, Y . Y ang, B. K. Ayan, et al. , “Scaling autoregres- sive models for content-rich text-to-image generation,” arXiv preprint arXiv:2206.10789, vol. 2, no. 3, p. 5, 2022

    2022 arXiv

  27. [34]

    Muse: Text-to-image generation via masked generative transform ers,

    H. Chang, H. Zhang, J. Barber, A. Maschinot, J. Lezama, L . Jiang, M.-H. Y ang, K. Murphy, W. T. Freeman, M. Rubinstein, et al., “Muse: Text-to-image generation via masked generative transform ers,” arXiv preprint arXiv:2301.00704, 2023

    2023 arXiv

  28. [35]

    Cogview: Mastering text-to-image generation via transformers,

    M. Ding, Z. Y ang, W. Hong, W. Zheng, C. Zhou, D. Yin, J. Lin , X. Zou, Z. Shao, H. Y ang, et al., “Cogview: Mastering text-to-image generation via transformers,” Advances in neural information processing systems , vol. 34, pp. 19822–19835, 2021

    2021

  29. [36]

    Cogview2: Faste r and better text-to-image generation via hierarchical transfo rmers,

    M. Ding, W. Zheng, W. Hong, and J. Tang, “Cogview2: Faste r and better text-to-image generation via hierarchical transfo rmers,” Advances in Neural Information Processing Systems , vol. 35, pp. 16890–16902, 2022

    2022

  30. [37]

    Cogview3: Finer and faster text-to-im age generation via relay diffusion,

    W. Zheng, J. Teng, Z. Y ang, W. Wang, J. Chen, X. Gu, Y . Dong , M. Ding, and J. Tang, “Cogview3: Finer and faster text-to-im age generation via relay diffusion,” arXiv preprint arXiv:2403.05121, 2024

    2024 arXiv

  31. [38]

    Auto-encoding variational bayes,

    D. P . Kingma, “Auto-encoding variational bayes,” arXiv preprint arXiv:1312.6114, 2013

    2013 arXiv

  32. [39]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis,

    D. Podell, Z. English, K. Lacey, A. Blattmann, T. Dockho rn, J. M¨ uller, J. Penna, and R. Rombach, “Sdxl: Improving latent diffusion models for high-resolution image synthesis,” arXiv preprint arXiv:2307.01952, 2023

    2023 arXiv

  33. [40]

    Adve rsarial diffusion distillation,

    A. Sauer, D. Lorenz, A. Blattmann, and R. Rombach, “Adve rsarial diffusion distillation,” in European Conference on Computer Vision , pp. 87–103, Springer, 2025

    2025

  34. [41]

    Scaling rectified flow transformers for high-resolution image synthesis,

    P . Esser, S. Kulal, A. Blattmann, R. Entezari, J. M¨ ulle r, H. Saini, Y . Levi, D. Lorenz, A. Sauer, F. Boesel, et al. , “Scaling rectified flow transformers for high-resolution image synthesis,” in F orty-first International Conference on Machine Learning , 2024

    2024

  35. [42]

    Fast high-resolution image synthesis with latent ad versarial diffusion distillation,

    A. Sauer, F. Boesel, T. Dockhorn, A. Blattmann, P . Esser , and R. Rom- bach, “Fast high-resolution image synthesis with latent ad versarial diffusion distillation,” arXiv preprint arXiv:2403.12015 , 2024

    2024 arXiv

  36. [43]

    Progressive distillation for fa st sampling of diffusion models,

    T. Salimans and J. Ho, “Progressive distillation for fa st sampling of diffusion models,” arXiv preprint arXiv:2202.00512 , 2022

    2022 arXiv

  37. [44]

    Sdxl-lightning: Progress ive adversarial diffusion distillation,

    S. Lin, A. Wang, and X. Y ang, “Sdxl-lightning: Progress ive adversarial diffusion distillation,” arXiv preprint arXiv:2402.13929 , 2024

    2024 arXiv

  38. [45]

    Google research imagen 2 update

    Google Research, “Google research imagen 2 update.” https://blog.google/technology/ai/google-imagen-2/. Accessed: 2024-11-10

    2024

  39. [46]

    Google deepmind imagen 3 update

    Google DeepMind, “Google deepmind imagen 3 update.” https://deepmind.com/technologies/imagen-3/ . Accesse d: 2024-11-10

    2024

  40. [47]

    From word vectors to multimodal embeddings: Techniques, applications, and future directions for large language models,

    C. Zhang, B. Peng, X. Sun, Q. Niu, J. Liu, K. Chen, M. Li, P . Feng, Z. Bi, M. Liu, et al. , “From word vectors to multimodal embeddings: Techniques, applications, and future directions for large language models,” arXiv preprint arXiv:2411.05036 , 2024

    2024

  41. [48]

    Dall-e 3: Ai that can create images from text wi th improved prompt following

    OpenAI, “Dall-e 3: Ai that can create images from text wi th improved prompt following.” https://openai.com/dall-e-3. Access ed: 2024-11-10

    2024

  42. [50]

    Consi stency models made easy,

    Z. Geng, A. Pokle, W. Luo, J. Lin, and J. Z. Kolter, “Consi stency models made easy,” arXiv preprint arXiv:2406.14548 , 2024

    2024 arXiv

  43. [51]

    Q-diffusion: Quantizing diffusion models,

    X. Li, Y . Liu, L. Lian, H. Y ang, Z. Dong, D. Kang, S. Zhang, and K. Keutzer, “Q-diffusion: Quantizing diffusion models,” i n Proceed- ings of the IEEE/CVF International Conference on Computer V ision, pp. 17535–17545, 2023

    2023

  44. [52]

    Post-traini ng quantiza- tion on diffusion models,

    Y . Shang, Z. Y uan, B. Xie, B. Wu, and Y . Y an, “Post-traini ng quantiza- tion on diffusion models,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 1972–1981, 2023

    1972

  45. [53]

    Efficientdm: Efficient quantization-aware fine-tuning of low-bit diffusion model s,

    Y . He, J. Liu, W. Wu, H. Zhou, and B. Zhuang, “Efficientdm: Efficient quantization-aware fine-tuning of low-bit diffusion model s,” arXiv preprint arXiv:2310.03270, 2023

    2023 arXiv

  46. [54]

    Efficientqat: Efficient quantization-aware traini ng for large language models,

    M. Chen, W. Shao, P . Xu, J. Wang, P . Gao, K. Zhang, Y . Qiao, and P . Luo, “Efficientqat: Efficient quantization-aware traini ng for large language models,” arXiv preprint arXiv:2407.11062 , 2024

    2024 arXiv

  47. [55]

    Mixed precision training,

    P . Micikevicius, S. Narang, J. Alben, G. Diamos, E. Else n, D. Garcia, B. Ginsburg, M. Houston, O. Kuchaiev, G. V enkatesh, et al. , “Mixed precision training,” arXiv preprint arXiv:1710.03740 , 2017

    2017 arXiv

  48. [56]

    Adaptive quantization for deep neural network,

    Y . Zhou, S.-M. Moosavi-Dezfooli, N.-M. Cheung, and P . F rossard, “Adaptive quantization for deep neural network,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 32, 2018

    2018

  49. [57]

    Parameter-efficie nt transfer learning for nlp,

    N. Houlsby, A. Giurgiu, S. Jastrzebski, B. Morrone, Q. D e Laroussilhe, A. Gesmundo, M. Attariyan, and S. Gelly, “Parameter-efficie nt transfer learning for nlp,” in International conference on machine learning , pp. 2790–2799, PMLR, 2019

    2019

  50. [58]

    Lora: Low-rank adaptation of large language mo dels,

    E. J. Hu, Y . Shen, P . Wallis, Z. Allen-Zhu, Y . Li, S. Wang, L. Wang, and W. Chen, “Lora: Low-rank adaptation of large language mo dels,” arXiv preprint arXiv:2106.09685 , 2021

    2021 arXiv

  51. [59]

    Qlora: Efficient finetuning of quantized llms,

    T. Dettmers, A. Pagnoni, A. Holtzman, and L. Zettlemoye r, “Qlora: Efficient finetuning of quantized llms,” Advances in Neural Information Processing Systems, vol. 36, 2024

    2024

  52. [60]

    Diffstyler: Diffusion-based localized image s tyle transfer,

    S. Li, “Diffstyler: Diffusion-based localized image s tyle transfer,” arXiv preprint arXiv:2403.18461, 2024

    2024

  53. [61]

    Lcm-lora: A universal stable-diffusion acceleration module,

    S. Luo, Y . Tan, S. Patil, D. Gu, P . von Platen, A. Passos, L . Huang, J. Li, and H. Zhao, “Lcm-lora: A universal stable-diffusion acceleration module,” arXiv preprint arXiv:2311.05556 , 2023

    2023 arXiv

  54. [62]

    T2i- adapter: Learning adapters to dig out more controllable abi lity for text- to-image diffusion models,

    C. Mou, X. Wang, L. Xie, Y . Wu, J. Zhang, Z. Qi, and Y . Shan, “T2i- adapter: Learning adapters to dig out more controllable abi lity for text- to-image diffusion models,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 38, pp. 4296–4304, 2024

    2024

  55. [63]

    Ip-adapter: Text compatible image prompt adapter for text-to-image diffusi on models,

    H. Y e, J. Zhang, S. Liu, X. Han, and W. Y ang, “Ip-adapter: Text compatible image prompt adapter for text-to-image diffusi on models,” arXiv preprint arXiv:2308.06721 , 2023

    2023 arXiv

  56. [64]

    Optimizing prompts f or text-to-image generation,

    Y . Hao, Z. Chi, L. Dong, and F. Wei, “Optimizing prompts f or text-to-image generation,” Advances in Neural Information Processing Systems, vol. 36, 2024

    2024

  57. [65]

    Sparsity-and hybrid ity-inspired visual parameter-efficient fine-tuning for medical diagnos is,

    M. Liu, L. Xu, S. Liu, and J. Zhang, “Sparsity-and hybrid ity-inspired visual parameter-efficient fine-tuning for medical diagnos is,” in In- ternational Conference on Medical Image Computing and Comp uter- Assisted Intervention, pp. 627–637, Springer, 2024

    2024

  58. [66]

    Zero: M emory optimizations toward training trillion parameter models,

    S. Rajbhandari, J. Rasley, O. Ruwase, and Y . He, “Zero: M emory optimizations toward training trillion parameter models, ” in SC20: In- ternational Conference for High Performance Computing, Ne tworking, Storage and Analysis , pp. 1–16, IEEE, 2020

    2020

  59. [67]

    Efficient large-scale language model training on gpu clus ters using megatron-lm,

    D. Narayanan, M. Shoeybi, J. Casper, P . LeGresley, M. Pa twary, V . Korthikanti, D. V ainbrand, P . Kashinkunti, J. Bernauer,B. Catanzaro, et al. , “Efficient large-scale language model training on gpu clus ters using megatron-lm,” in Proceedings of the International Conferenc...

    2021

  60. [68]

    xdit: an infer ence engine for diffusion transformers (dits) with massive para llelism,

    J. Fang, J. Pan, X. Sun, A. Li, and J. Wang, “xdit: an infer ence engine for diffusion transformers (dits) with massive para llelism,” arXiv preprint arXiv:2411.01738, 2024

    2024 arXiv

  61. [69]

    Deep generative model for imag e inpainting with local binary pattern learning and spatial a ttention,

    H. Wu, J. Zhou, and Y . Li, “Deep generative model for imag e inpainting with local binary pattern learning and spatial a ttention,” IEEE Transactions on Multimedia , vol. 24, pp. 4016–4027, 2021

    2021

  62. [70]

    Gen erative image inpainting with contextual attention,

    J. Y u, Z. Lin, J. Y ang, X. Shen, X. Lu, and T. S. Huang, “Gen erative image inpainting with contextual attention,” in Proceedings of the IEEE conference on computer vision and pattern recognition , pp. 5505–5514, 2018

    2018

  63. [71]

    Grig: Few-shot generative residual image inpaint ing,

    W. Lu, X. Jiang, X. Jin, Y .-L. Y ang, M. Gong, T. Wang, K. Sh i, and H. Zhao, “Grig: Few-shot generative residual image inpaint ing,” arXiv preprint arXiv:2304.12035, 2023

    2023 arXiv

  64. [72]

    Latentpaint : Image inpainting in latent space with diffusion models,

    C. Corneanu, R. Gadde, and A. M. Martinez, “Latentpaint : Image inpainting in latent space with diffusion models,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 4334–4343, 2024

    2024

  65. [73]

    Repaint: Inpainting using denoising diffusio n probabilis- tic models,

    A. Lugmayr, M. Danelljan, A. Romero, F. Y u, R. Timofte, a nd L. V an Gool, “Repaint: Inpainting using denoising diffusio n probabilis- tic models,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 11461–11471, 2022

    2022

  66. [74]

    Gradpaint: Gradi ent-guided inpainting with diffusion models,

    A. Grechka, G. Couairon, and M. Cord, “Gradpaint: Gradi ent-guided inpainting with diffusion models,” Computer Vision and Image Under- standing, vol. 240, p. 103928, 2024

    2024

  67. [75]

    Inout: Diverse image outpainting via gan inversio n,

    Y .-C. Cheng, C. H. Lin, H.-Y . Lee, J. Ren, S. Tulyakov, an d M.- H. Y ang, “Inout: Diverse image outpainting via gan inversio n,” in Proceedings of the IEEE/CVF Conference on Computer Vision a nd Pattern Recognition, pp. 11431–11440, 2022

    2022

  68. [76]

    Painting outside as inside: Edge guided image outpainting via bidirec- tional rearrangement with progressive step learning,

    K. Kim, Y . Y un, K.-W. Kang, K. Kong, S. Lee, and S.-J. Kang , “Painting outside as inside: Edge guided image outpainting via bidirec- tional rearrangement with progressive step learning,” in Proceedings of the IEEE/CVF winter conference on applications of computer vision, pp...

    2021

  69. [77]

    Towards reliable image outpainting: Learning structure-aware multimodal fusion with depth guidance,

    L. Zhang, C. Lin, K. Liao, and Y . Zhao, “Towards reliable image outpainting: Learning structure-aware multimodal fusion with depth guidance,” in ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pp. 1–5, IEEE, 2023

    2023

  70. [78]

    Em ergent correspondence from image diffusion,

    L. Tang, M. Jia, Q. Wang, C. P . Phoo, and B. Hariharan, “Em ergent correspondence from image diffusion,” Advances in Neural Information Processing Systems, vol. 36, pp. 1363–1389, 2023

    2023

  71. [79]

    Zero-1-to-3: Zero-shot one image to 3d object ,

    R. Liu, R. Wu, B. V an Hoorick, P . Tokmakov, S. Zakharov, a nd C. V ondrick, “Zero-1-to-3: Zero-shot one image to 3d object ,” in Proceedings of the IEEE/CVF international conference on co mputer vision, pp. 9298–9309, 2023

    2023

  72. [80]

    Wonder3d: Single image to 3d using cross-domain diffusion,

    X. Long, Y .-C. Guo, C. Lin, Y . Liu, Z. Dou, L. Liu, Y . Ma, S. -H. Zhang, M. Habermann, C. Theobalt, et al. , “Wonder3d: Single image to 3d using cross-domain diffusion,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 9970– 9980, 2024

    2024

  73. [81]

    Make-it-3d: High-fidelity 3d creation from a single image w ith diffu- sion prior,

    J. Tang, T. Wang, B. Zhang, T. Zhang, R. Yi, L. Ma, and D. Ch en, “Make-it-3d: High-fidelity 3d creation from a single image w ith diffu- sion prior,” in Proceedings of the IEEE/CVF international conference on computer vision , pp. 22819–22829, 2023

    2023

  74. [82]

    Adding conditional c ontrol to text-to-image diffusion models,

    L. Zhang, A. Rao, and M. Agrawala, “Adding conditional c ontrol to text-to-image diffusion models,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , pp. 3836–3847, 2023

    2023

  75. [83]

    Controlnet- xs: Designing an efficient and effective architecture for controlling tex t-to-image diffusion models,

    D. Zavadski, J.-F. Feiden, and C. Rother, “Controlnet- xs: Designing an efficient and effective architecture for controlling tex t-to-image diffusion models,” arXiv preprint arXiv:2312.06573 , 2023

    2023 arXiv

  76. [84]

    Uni-controlnet: All-in-one control to text-to-ima ge diffusion models,

    S. Zhao, D. Chen, Y .-C. Chen, J. Bao, S. Hao, L. Y uan, and K .-Y . K. Wong, “Uni-controlnet: All-in-one control to text-to-ima ge diffusion models,” Advances in Neural Information Processing Systems , vol. 36, 2024

    2024

  77. [85]

    Styledrop: Text-to-image generation in any style,

    K. Sohn, N. Ruiz, K. Lee, D. C. Chin, I. Blok, H. Chang, J. B arber, L. Jiang, G. Entis, Y . Li, et al. , “Styledrop: Text-to-image generation in any style,” arXiv preprint arXiv:2306.00983 , 2023

    2023 arXiv

  78. [86]

    Any-to-any style transfer: M aking picasso and da vinci collaborate,

    S. Liu, J. Y e, and X. Wang, “Any-to-any style transfer: M aking picasso and da vinci collaborate,” arXiv preprint arXiv:2304.09728 , 2023

    2023 arXiv

  79. [87]

    Clipstyler: Image style transfer w ith a single text condition,

    G. Kwon and J. C. Y e, “Clipstyler: Image style transfer w ith a single text condition,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 18062–18071, 2022

    2022

  80. [88]

    Styleshot: A snapshot on any style,

    J. Gao, Y . Liu, Y . Sun, Y . Tang, Y . Zeng, K. Chen, and C. Zha o, “Styleshot: A snapshot on any style,” arXiv preprint arXiv:2407.01414, 2024

    2024 arXiv

  81. [89]

    Metrop olis theorem and its applications in single image detail enhance ment,

    H. Jiang, M. Asad, J. Liu, H. Zhang, and D. Cheng, “Metrop olis theorem and its applications in single image detail enhance ment,” arXiv preprint arXiv:2302.09762, 2023

    2023 arXiv

  82. [90]

    Multi-scale image decomposition using a l ocal statistical edge model,

    K.-M. Wong, “Multi-scale image decomposition using a l ocal statistical edge model,” in 2021 IEEE 7th International Conference on Virtual Reality (ICVR) , pp. 10–18, IEEE, 2021

    2021

  83. [91]

    Crnet: A detail-preserving network for unified i mage restoration and enhancement task,

    K. Y ang, T. Hu, K. Dai, G. Chen, Y . Cao, W. Dong, P . Wu, Y . Zh ang, and Q. Y an, “Crnet: A detail-preserving network for unified i mage restoration and enhancement task,” arXiv preprint arXiv:2404.14132 , 2024

    2024 arXiv

  84. [92]

    Ecaformer: Low-light image enhancement using cross attention,

    Y . Ruan, H. Ma, W. Li, and X. Wang, “Ecaformer: Low-light image enhancement using cross attention,” arXiv preprint arXiv:2406.13281 , 2024

    2024 arXiv

  85. [93]

    Gans trained by a two time-scale update rule converge to a lo cal nash equilibrium,

    M. Heusel, H. Ramsauer, T. Unterthiner, B. Nessler, and S. Hochreiter, “Gans trained by a two time-scale update rule converge to a lo cal nash equilibrium,” Advances in neural information processing systems , vol. 30, 2017

    2017

  86. [94]

    On aliased resizing and surprising subtleties in gan evaluation,

    G. Parmar, R. Zhang, and J.-Y . Zhu, “On aliased resizing and surprising subtleties in gan evaluation,” in CVPR, 2022

    2022

  87. [95]

    Demysti- fying mmd gans,

    M. Bi´ nkowski, D. J. Sutherland, M. Arbel, and A. Gretto n, “Demysti- fying mmd gans,” arXiv preprint arXiv:1801.01401 , 2018

    2018 arXiv

  88. [96]

    The unreasonable effectiveness of deep features as a perceptua l metric,

    R. Zhang, P . Isola, A. A. Efros, E. Shechtman, and O. Wang , “The unreasonable effectiveness of deep features as a perceptua l metric,” in Proceedings of the IEEE conference on computer vision and pa ttern recognition, pp. 586–595, 2018

    2018

  89. [97]

    Imagereward: Learning and evaluating human preferences f or text- to-image generation,

    J. Xu, X. Liu, Y . Wu, Y . Tong, Q. Li, M. Ding, J. Tang, and Y . Dong, “Imagereward: Learning and evaluating human preferences f or text- to-image generation,” Advances in Neural Information Processing Systems, vol. 36, 2024

    2024

  90. [98]

    Clip-agiqa: Boos ting the performance of ai-generated image quality assessment with clip,

    Z. Tang, Z. Wang, B. Peng, and J. Dong, “Clip-agiqa: Boos ting the performance of ai-generated image quality assessment with clip,” arXiv preprint arXiv:2408.15098, 2024

    2024 arXiv

  91. [99]

    Human evaluation of text-to-image models on a multi-task benchma rk,

    V . Petsiuk, A. E. Siemenn, S. Surbehera, Z. Chin, K. Tyse r, G. Hunter, A. Raghavan, Y . Hicke, B. A. Plummer, O. Kerret, et al. , “Human evaluation of text-to-image models on a multi-task benchma rk,” arXiv preprint arXiv:2211.12112, 2022

    2022 arXiv

  92. [100]

    Hum an-ai co- creation: evaluating the impact of large-scale text-to-im age generative models on the creative process,

    T. Turchi, S. Carta, L. Ambrosini, and A. Malizia, “Hum an-ai co- creation: evaluating the impact of large-scale text-to-im age generative models on the creative process,” in International Symposium on End User Development, pp. 35–51, Springer, 2023

    2023

  93. [101]

    Evaluating text-to-visual generation with im age-to-text generation,

    Z. Lin, D. Pathak, B. Li, J. Li, X. Xia, G. Neubig, P . Zhan g, and D. Ramanan, “Evaluating text-to-visual generation with im age-to-text generation,” in European Conference on Computer Vision, pp. 366–384, Springer, 2025

    2025

  94. [102]

    Llms and diffusion models i n ui/ux: Advancing human-computer interaction and design,

    L. Sun, M. Qin, and B. Peng, “Llms and diffusion models i n ui/ux: Advancing human-computer interaction and design,” OSF Preprints , Oct 2024

    2024

  95. [103]

    Clip- score: A reference-free evaluation metric for image captio ning,

    J. Hessel, A. Holtzman, M. Forbes, R. L. Bras, and Y . Cho i, “Clip- score: A reference-free evaluation metric for image captio ning,” arXiv preprint arXiv:2104.08718, 2021

    2021 arXiv

  96. [104]

    Enhanc- ing semantic fidelity in text-to-image synthesis: Attentio n regulation in diffusion models,

    Y . Zhang, T. T. Tzun, L. W. Hern, T. Sim, and K. Kawaguchi , “Enhanc- ing semantic fidelity in text-to-image synthesis: Attentio n regulation in diffusion models,” arXiv preprint arXiv:2403.06381 , 2024

    2024 arXiv

  97. [105]

    Hypernymy understan ding evalu- ation of text-to-image models via wordnet hierarchy,

    A. Baryshnikov and M. Ryabinin, “Hypernymy understan ding evalu- ation of text-to-image models via wordnet hierarchy,” arXiv preprint arXiv:2310.09247, 2023

    2023 arXiv

  98. [106]

    Memory-efficient fine-tuni ng for quantized diffusion model,

    H. Ryu, S. Lim, and H. Shim, “Memory-efficient fine-tuni ng for quantized diffusion model,” in European Conference on Computer Vision, pp. 356–372, Springer, 2025

    2025

  99. [107]

    Mlperf inference benchmark,

    V . J. Reddi, C. Cheng, D. Kanter, P . Mattson, G. Schmuel ling, C.-J. Wu, B. Anderson, M. Breughe, M. Charlebois, W. Chou, et al., “Mlperf inference benchmark,” in 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA) , pp. 446–459, IEEE, 2020

    2020

  100. [108]

    Carbon emission s and large neural network training,

    D. Patterson, J. Gonzalez, Q. Le, C. Liang, L.-M. Mungu ia, D. Rothchild, D. So, M. Texier, and J. Dean, “Carbon emission s and large neural network training,” arXiv preprint arXiv:2104.10350, 2021

    2021 arXiv

  101. [109]

    The c hallenges of image generation models in generating multi-component i mages,

    T. Y . Foong, S. Kotyan, P . Y . Mao, and D. V . V argas, “The c hallenges of image generation models in generating multi-component i mages,” arXiv preprint arXiv:2311.13620 , 2023

    2023 arXiv

  102. [110]

    Prefpa int: Aligning image inpainting diffusion model with human prefe rence,

    K. Liu, Z. Zhu, C. Li, H. Liu, H. Zeng, and J. Hou, “Prefpa int: Aligning image inpainting diffusion model with human prefe rence,” arXiv preprint arXiv:2410.21966 , 2024

    2024 arXiv

  103. [111]

    Playground v2. 5: Three insights towards enhancing aesthe tic quality in text-to-image generation,

    D. Li, A. Kamko, E. Akhgari, A. Sabet, L. Xu, and S. Doshi , “Playground v2. 5: Three insights towards enhancing aesthe tic quality in text-to-image generation,” arXiv preprint arXiv:2402.17245 , 2024

    2024 arXiv

  104. [112]

    Ai gciqa2023: A large-scale image quality assessment database for ai gene rated images: from the perspectives of quality, authenticity and correspon- dence,

    J. Wang, H. Duan, J. Liu, S. Chen, X. Min, and G. Zhai, “Ai gciqa2023: A large-scale image quality assessment database for ai gene rated images: from the perspectives of quality, authenticity and correspon- dence,” in CAAI International Conference on Artificial Intelligence , p...

    2023

  105. [113]

    Emu: Enhancing image generation models using photogenic needles in a haystack,

    X. Dai, J. Hou, C.-Y . Ma, S. Tsai, J. Wang, R. Wang, P . Zha ng, S. V andenhende, X. Wang, A. Dubey, et al. , “Emu: Enhancing image generation models using photogenic needles in a haystack,” arXiv preprint arXiv:2309.15807, 2023

    2023 arXiv

  106. [114]

    Promptmagician: Interactive prompt engineering for text- to-image creation,

    Y . Feng, X. Wang, K. K. Wong, S. Wang, Y . Lu, M. Zhu, B. Wan g, and W. Chen, “Promptmagician: Interactive prompt engineering for text- to-image creation,” IEEE Transactions on Visualization and Computer Graphics, 2023

    2023

  107. [115]

    Beau ti- fulprompt: Towards automatic prompt engineering for text- to-image synthesis,

    T. Cao, C. Wang, B. Liu, Z. Wu, J. Zhu, and J. Huang, “Beau ti- fulprompt: Towards automatic prompt engineering for text- to-image synthesis,” arXiv preprint arXiv:2311.06752 , 2023

    2023 arXiv

  108. [116]

    Neuroprompts: An a daptive framework to optimize prompts for text-to-image generatio n,

    S. Rosenman, V . Lal, and P . Howard, “Neuroprompts: An a daptive framework to optimize prompts for text-to-image generatio n,” arXiv preprint arXiv:2311.12229, 2023

    2023 arXiv

  109. [117]

    Securing large language models: Addressing bias, misinformation, a nd prompt attacks,

    B. Peng, K. Chen, M. Li, P . Feng, Z. Bi, J. Liu, and Q. Niu, “Securing large language models: Addressing bias, misinformation, a nd prompt attacks,” arXiv preprint arXiv:2409.08087 , 2024

    2024

  110. [118]

    Sneakypro mpt: Jail- breaking text-to-image generative models,

    Y . Y ang, B. Hui, H. Y uan, N. Gong, and Y . Cao, “Sneakypro mpt: Jail- breaking text-to-image generative models,” in 2024 IEEE symposium on security and privacy (SP) , pp. 897–912, IEEE, 2024

    2024

  111. [119]

    Guardt2i: Defend- ing text-to-image models from adversarial prompts,

    Y . Y ang, R. Gao, X. Y ang, J. Zhong, and Q. Xu, “Guardt2i: Defend- ing text-to-image models from adversarial prompts,” arXiv preprint arXiv:2403.01446, 2024

    2024 arXiv

  112. [120]

    Unsafebench: Benchmarking image safety classifiers on rea l-world and ai-generated images,

    Y . Qu, X. Shen, Y . Wu, M. Backes, S. Zannettou, and Y . Zha ng, “Unsafebench: Benchmarking image safety classifiers on rea l-world and ai-generated images,” arXiv preprint arXiv:2405.03486 , 2024

    2024 arXiv

  113. [121]

    Turbovit: Generating fast vi- sion transformers via generative architecture search,

    A. Wong, S. Abbasi, and S. Nair, “Turbovit: Generating fast vi- sion transformers via generative architecture search,” arXiv preprint arXiv:2308.11421, 2023

    2023 arXiv

  114. [122]

    Ultra-high-resolution image synthesis wit h pyramid diffusion model,

    J. Y ang, “Ultra-high-resolution image synthesis wit h pyramid diffusion model,” arXiv preprint arXiv:2403.12915 , 2024

    2024 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.