Pith. sign in

REVIEW 1 major objections 5 minor 100 references

Identifying Prompted Artist Names from Generated Images

T0 review · 1 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper introduces a 1.95M-image benchmark and shows that no current vision model can reliably identify which artist name was used in the prompt of a generated image.

desk verdict Large, transparent benchmark for prompted-artist identification; method comparison is solid for three generators, but the Midjourney subset uses a different artist split than the stated one, so the held-out generalization story needs a fix. read the letter →

arxiv 2507.18633 v1 pith:KWEQWDKN submitted 2025-07-24 cs.CV

Grace Su , Sheng-Yu Wang , Aaron Hertzmann , Eli Shechtman , Jun-Yan Zhu , Richard Zhang This is my paper · ORCID
classification cs.CV
keywords promptedartistidentificationtext-to-imagegenerationstyleattributiongeneratedimagedetectionbenchmarkgeneralizationprototypicalnetworksnames
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper's central claim is that prompted-artist identification—predicting, from the image alone, which artist's name appeared in the text prompt that generated it—is a well-defined task that current vision methods cannot yet solve reliably. To back this claim, the authors build the first large-scale benchmark for the task: 1.95 million generated images, 110 frequently prompted artist names, spanning simple and real-user complex prompts, four text-to-image generators, and prompts with one, two, or three artists. Their experiments show consistent generalization patterns: classifiers and few-shot prototypical networks trained on artist-prompted images do best on seen artists and complex prompts, while style descriptors trained on real artwork transfer better to simple and held-out prompts; multi-artist prompts are the hardest, and no method approaches saturation. The benchmark matters because online platforms ban artist-named generations, yet without access to the original prompt there is currently no reliable way to detect them.

What carries the argument

The load-bearing object is the structured benchmark dataset itself. For each of 110 frequently prompted artist names, images are generated by inserting the name into the same content prompts—simple prompts written for the task and complex prompts scraped from real users—across SDXL, SD1.5, PixArt-Σ, and Midjourney, with 100 seen artists, 10 held-out artists, and separate multi-artist subsets with two and three names. The 'same content prompt, different artist name' design isolates the effect of the artist name from the effect of the content, and the held-out artists and prompts force methods to generalize rather than memorize. The evaluation then compares retrieval-based baselines (CLIP, DINOv2, contrastive style descriptors, data attribution features) with fine-tuned classifiers and prototypical networks, using CLIP embedding similarity to quantify how strongly an artist name influences the output.

What would settle it

Regenerate a held-out set of the benchmark's complex SDXL prompts twice—once with the artist name and once with the identical prompt and seed but no artist name—and have a strong binary detector or human raters pick which image was artist-prompted. If accuracy is at chance on complex prompts, many test items are visually unanswerable, and the measured headroom would overstate the limits of vision methods.

Watch

Extended reading notes

Core claim

The paper establishes that recognizing the artist an image generator was told to imitate is a distinct problem from recognizing artistic style in real paintings, and that the gap between the two is measurable. On the released benchmark, methods trained on real artwork (contrastive style descriptors) generalize well to simple prompts and held-out artists, where the artist's style is visually apparent, but trained classifiers and prototypical networks—which learn from generated, artist-prompted images—surpass them on complex prompts and seen artists. Across all settings, the best methods stay below 91% accuracy, performance drops consistently as prompts become more complex and on images from PixArt compared with SDXL and SD1.5, and prompts with multiple artists remain the hardest case. The authors also show that adding training images from one generator does not improve performance on an unseen generator, and that each added artist name in a prompt has a smaller visual effect than the previous one.

Load-bearing premise

The benchmark assumes that an artist name placed in a prompt leaves a detectable visual trace in the generated image for the four generators studied; when prompts are complex or the generator is PixArt, the paper's own measurements show that trace weakens, and it excluded SD3.5 and FLUX because they showed almost no trace.

Editorial extensions

If this is right

  • Prompted-artist identification cannot be equated with style recognition of real artwork; the two tasks have different generalization curves, so moderation tools need training on generated, artist-prompted images.
  • Deployed detection will be most reliable on simple prompts and familiar artists, and least reliable on multi-artist prompts and generators like PixArt where the artist's influence is weak.
  • No current method generalizes across text-to-image generators: adding training data from one generator does not help on another, so cross-generator robustness must be handled explicitly.
  • The benchmark's headroom is large—best methods are far below saturation on every setting—so artist-name detection is an open problem rather than a solved one.
  • Multi-artist prompts are the clearest failure mode; because each additional name dilutes the visual trace, a practical system would need to predict a set of names rather than a single label.

Reading between the lines

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

  • If the measured patterns hold for future generators that do respond to artist names, moderation could be framed as open-set retrieval against a maintained artist database, with held-out artist accuracy as the primary deployment metric rather than closed-set accuracy.
  • The persistent gap between real-art style descriptors and generated-image classifiers suggests a generator-aware style encoder—one that models how each text-to-image model attenuates artist influence—could transfer better across generators than either current family.
  • A testable extension of the paper's CLIP similarity analysis is a per-artist 'promptability' score (how much an artist name changes the generated image); if such scores predict classification difficulty, they could be used to choose which artists need extra reference data or targeted training.
  • Because the benchmark excludes SD3.5 and FLUX on the grounds that they ignore artist names, the paper implies a moderation system for those models would need to detect style imitation from prompt description rather than from artist names, a different and possibly harder signal.
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

1 major / 5 minor

Summary. The paper introduces a large-scale benchmark for identifying which artist name was invoked in a text-to-image prompt, given only the generated image. The dataset comprises roughly 1.95M images across 110 artist names and four generalization axes: seen versus held-out artists, simple versus complex prompts, multiple text-to-image generators (SDXL, SD1.5, PixArt-Σ, and a collected Midjourney subset), and prompts with two or three artist names. The authors evaluate retrieval-based baselines (CLIP, DINOv2, CSD, AbC) and trained classifiers (prototypical networks and a vanilla classifier), reporting accuracy and mAP@10 with bootstrapped confidence intervals. The central empirical claims are that supervised and few-shot methods generalize better on seen artists and complex prompts, style descriptors transfer better on simple prompts and held-out artists, multi-artist prompts are the most difficult, and no method approaches saturation.

Significance. If the benchmark is valid, it is a valuable public testbed for the responsible moderation of text-to-image content and for studying the relationship between prompted artist names and generated image style. The paper's strengths are its controlled prompt construction, the breadth of evaluated method families, transparent dataset release, bootstrapped uncertainty estimates, and several informative ablations (training data composition, prototype sources, and k-NN behavior). The problem is timely and the reported headroom is plausibly of practical interest. The main weaknesses are internal inconsistencies in the Midjourney evaluation split and the lack of a direct test of whether the remaining benchmark items are visually answerable, both of which affect the interpretation of the cross-model and 'headroom' conclusions.

major comments (1)
  1. [Section 3.3, Section 3.5, Section 5.1] The benchmark's validity rests on the assumption that an artist name explicitly placed in a prompt reliably and detectably alters the generated image. The authors themselves exclude SD3.5 and FLUX because this assumption fails, and Section 3.5 quantifies that PixArt images and complex prompts substantially dilute the artist's influence. Yet the benchmark retains PixArt and complex-prompt test items without establishing that these items are answerable in principle. As a result, the 'substantial headroom' conclusion in Section 5.1 may conflate genuinely difficult vision problems with test items where the image simply does not contain enough artist-specific signal to identify the prompted name. Please add an answerability analysis, such as human accuracy on a representative sample or an oracle-style upper bound per condition, and discuss whether the observed accuracy gaps reflect method limitations or unanswerable items.
minor comments (5)
  1. [Section 5.1, Table 8] The statement that 'none exceed 91% accuracy' is contradicted by Table 8 in the supplement, where CSD with artist-average retrieval reaches 92.0% on held-out artists with simple prompts; please qualify the claim to the main evaluation setting or adjust the text.
  2. [Section 3.2, Section 7.2] The main text says simple prompts use '500 different contents sampled from ChatGPT', while Section 7.2 states that 100 subjects were curated and lists a ChatGPT request for 100 subjects; these numbers should be reconciled.
  3. [Section 5.3] The text refers to 'our observation in Table 3.5'; this should be Section 3.5, since 3.5 is not a table.
  4. [Figure 6] The caption of Figure 6 describes the axes as 100-way and 10-way for all models, but Table 13 reports Midjourney as 34-way and 96-way; the figure caption and/or the figure itself should be corrected to reflect the actual label spaces.
  5. [Supplement, related work paragraph] There is a typo in 'by Leottaet al.'; it should read 'Leotta et al.'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: benchmark accuracies are direct evaluations on ground-truth prompt labels; the noted Midjourney split inconsistency is a correctness issue, not circularity.

full rationale

The paper's reported quantities are direct classification accuracies on a dataset whose labels are the ground-truth prompt strings used to generate each image; no parameter is fitted to a subset of data and then renamed as a prediction, and no claim is defined in terms of the quantity it is supposed to establish. Held-out artist evaluation is a genuine few-shot and retrieval test: Section 7.1 states that the held-out artists were cross-referenced against CSD's training captions ('We then cross-reference our filtered artists with the captions in CSD's dataset to ensure our held-out artists' names do not appear in them'), so CSD's held-out numbers are not contaminated by test-class training. The prototypical network obtains held-out prototypes from reference images at inference time, which is standard few-shot evaluation rather than circular fitting, and the vanilla classifier is explicitly not applied to held-out artists. The only self-citation found is Wang et al. [80] for blur/JPEG augmentation in Section 8; it is a training-detail citation that is not load-bearing for any benchmark conclusion, so it does not constitute circularity. Two internal consistency issues are outside this pass and do not change the circularity score: Section 3.1 specifies 100 seen/10 held-out artists, while Table 5(c) and Table 13 report the Midjourney subset as 35 seen/95 held-out artists (34-way/96-way), so Midjourney's seen-versus-held-out numbers are not on the same label space as the other generators; and Section 5.1's 'none exceed 91% accuracy' conflicts with Table 8's 92.0% CSD held-out simple-prompt result. These are correctness concerns, not evidence that any result reduces to its own inputs.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The benchmark depends on the four generators rendering artist styles in a detectable way, on accurate prompt-derived labels, and on the released metadata. The only fitted quantities are standard training hyperparameters and curation thresholds; no new entities are introduced.

free parameters (3)
  • Prototypical network training hyperparameters = lr=1e-6, temperature=0.07, batch=512, 1 epoch
    Best configuration selected by sweeping learning rates (1e-7 to 1e-5) and epochs (1-5) on the benchmark training set (Supplement Section 8). These values affect the reported prototypical network accuracies but are standard training choices.
  • Test-time reference prompt selection = first 5 prompts of each held-out test prompt set
    Held-out retrieval and prototype classes use images generated from the first 5 test prompts as references (Section 7.2). The paper does not analyze sensitivity to this choice.
  • Real image curation thresholds = aesthetic score > 4 or > 6, SSCD similarity < 0.8
    Ad hoc filtering of LAION-5B for the real artist reference images (Section 7.1). These thresholds determine the style references used by CSD and prototypical networks.
assumptions (4)
  • domain assumption Artist names in prompts leave a visually detectable trace in generated images for the four studied generators.
    The task is impossible without this. Section 3.3 shows SD3.5 and FLUX violate it and are excluded; Section 3.5 shows the trace is weaker for complex prompts and PixArt.
  • domain assumption JourneyDB prompt-image pairs and Llama 3 8B artist-name extraction provide accurate ground-truth artist labels.
    Complex prompts are filtered for artist names using Llama 3 (Section 7.2). Extraction errors would add label noise to the complex-prompt subsets.
  • domain assumption CLIP image similarity is a valid proxy for style alignment between generated and real artwork.
    Used in Section 3.5 to characterize the difficulty of the benchmark. If CLIP fails to capture style, the difficulty analysis could be misleading.
  • standard math Bootstrapped resampling gives valid confidence intervals for classification accuracy.
    Standard statistical procedure used for all reported error bars (Supplement Section 10.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Identifying Prompted Artist Names from Generated Images." pith.science (2026). https://pith.science/paper/KWEQWDKN

@misc{pith2026250718633,
  author       = {Pith},
  title        = {Pith review of: Identifying Prompted Artist Names from Generated Images},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KWEQWDKN}},
  note         = {Machine review of arXiv:2507.18633}
}
read the original abstract

A common and controversial use of text-to-image models is to generate pictures by explicitly naming artists, such as "in the style of Greg Rutkowski". We introduce a benchmark for prompted-artist recognition: predicting which artist names were invoked in the prompt from the image alone. The dataset contains 1.95M images covering 110 artists and spans four generalization settings: held-out artists, increasing prompt complexity, multiple-artist prompts, and different text-to-image models. We evaluate feature similarity baselines, contrastive style descriptors, data attribution methods, supervised classifiers, and few-shot prototypical networks. Generalization patterns vary: supervised and few-shot models excel on seen artists and complex prompts, whereas style descriptors transfer better when the artist's style is pronounced; multi-artist prompts remain the most challenging. Our benchmark reveals substantial headroom and provides a public testbed to advance the responsible moderation of text-to-image models. We release the dataset and benchmark to foster further research: https://graceduansu.github.io/IdentifyingPromptedArtists/

Figures

Figures reproduced from arXiv: 2507.18633 by the authors.

Figure 1
Figure 1. Prompted Artist Identification Benchmark. We introduce the first large-scale benchmark for identifying prompted artist names from generated images. The benchmark covers four axes of generalization that match realistic use cases: (1) Artists: we collect artists commonly used in prompts and simulate open-set artist classification by testing on artists not seen during training. (2) Prompt complexity: users describe ima… view at source ↗
Figure 2
Figure 2. Prompted Artist Identification Dataset. We construct a structured dataset of 1.95M images to benchmark different methods on predicting prompted artist names from generated images. To help disentangle the effect of prompting given an artist name, we query a text-to-image model given the same content prompt (rows), but insert different artist names (columns). Our dataset consists of images generated by SDXL [59], SD1.… view at source ↗
Figure 3
Figure 3. Dataset Statistics. The prompted artist identification benchmark employs a structured dataset of 1.95M images to evaluate vision methods across four axes of generalization. We collect 110 of the most frequently prompted artists, split into 100 seen and 10 held-out artists (1 st chart). Next, we collect 1,000 complex prompts and 500 simple prompts in which artist names are inserted (2 nd chart). For seen artists, we … view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Statistics on number of artists per prompt. To under￾stand how frequently text-to-image model users prompt for multi￾ple artists, we take a random sample of 10,000 prompts collected from Midjourney users in JourneyDB [78] and visualize the dis￾tribution of the number o…
Figure 5
Figure 5. Figure 5: Multi-artist prompted images. In the example images shown, we observe that the effect of adding each artist’s name in the prompt diminishes as more artists are added. Each row shows a set of images generated with the same prompt and generation seed, and the number of a…
Figure 6
Figure 6. Figure 6: Single-artist prediction results. We compare the prompted artist classification accuracy of various visual representation meth￾ods. We test within our seen artists set (100-way classification, x-axis) and on held-out artists (10-way classification, y-axis). We also tes…
Figure 7
Figure 7. Figure 7: Generalization to unseen generators. For the best-performing methods, prototypical networks and CSD, we evaluate perfor￾mance on unseen text-to-image models by progressively increasing the training set of images. Cells below the green diagonals evaluate generalization …
Figure 8
Figure 8. Figure 8: Multi-artist prediction results. We evaluate visual rep￾resentation methods on the multi-artist classification task, where the input image is prompted with multiple artists’ names. We test on SDXL-generated images with 2 artists and 3 artists in the prompt, and report …
Figure 10
Figure 10. Figure 10: Ablation on number of training prompts. We plot the prototypical network classification accuracy on seen artists (left) and held-out artists (right) as a function of the number of prompts in the training set. Classification accuracy on seen artists improves steadily w…
Figure 11
Figure 11. Figure 11: Artist statistics. We plot the frequency of each artist’s name in our LAION training set, with artists labeled public do￾main (blue) or non-public domain (orange), as defined by 95 years post-mortem. Out of our set of 100 seen artists, 45 are non-public domain, and th…
Figure 12
Figure 12. Figure 12: FLUX and SD3.5 examples. For FLUX.1-dev [40] and SD3.5 [22], inserting artist names into the prompt frequently has little to no influence on the generated image’s style compared to using the same prompt without any artist name. Most of the resulting images are photore…
Figure 13
Figure 13. Figure 13: Effect of k on accuracy for k-NN classification. We perform k-NN evaluation on multiple k values and find that for each retrieval-based method, the performance only changes a small amount as k increases. Thus, the performance ranking of the methods remains unchanged a…
Figure 14
Figure 14. Figure 14: Bootstrapping procedure convergence. We plot stan￾dard error against the number of bootstrap iterations for the pro￾totypical network’s artist classification accuracy on SDXL images and simple prompts. We use 2000 bootstrap iterations in our main benchmark evaluation …

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

100 extracted references · 67 canonical work pages

  1. [1]

    Updated artist name guidelines, 2023

    Adobe. Updated artist name guidelines, 2023. Accessed: 2024-11-13. 1

    2023

  2. [2]

    Stable safety, 2025

    Stability AI. Stable safety, 2025. Accessed on April 16,

    2025

  3. [3]

    The alt-right manipulated my comic

    Sarah Andersen. The alt-right manipulated my comic. then a.i. claimed it. The New York Times, 2022. 1

    2022

  4. [4]

    Hrs-bench: Holistic, reliable and scalable benchmark for text-to-image models

    Eslam Mohamed Bakr, Pengzhan Sun, Xiaoqian Shen, Faizan Farooq Khan, Li Erran Li, and Mohamed Elhoseiny. Hrs-bench: Holistic, reliable and scalable benchmark for text-to-image models. In IEEE Conference on Computer Vi- sion and Pattern Recognition (CVPR), pages 20041–20053,

  5. [5]

    Leaked: the names of more than 16,000 non- consenting artists allegedly used to train midjourney’s ai,

    Theo Belci. Leaked: the names of more than 16,000 non- consenting artists allegedly used to train midjourney’s ai,

  6. [6]

    Gallerygpt: Analyzing paintings with large multimodal mod- els

    Yi Bin, Wenhao Shi, Yujuan Ding, Zhiqiang Hu, Zheng Wang, Yang Yang, See-Kiong Ng, and Heng Tao Shen. Gallerygpt: Analyzing paintings with large multimodal mod- els. In Proceedings of the 32nd ACM International Confer- ence on Multimedia, pages 7734–7743, 2024. 10

    2024

  7. [7]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. In IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR), 2021. 6, 8

    2021

  8. [8]

    Schwartz

    Stephen Casper, Daniel Filan, Gabriel Nicholas, Miles Brundage, and Matthew L. Schwartz. Measuring the success of diffusion models at imitating human artists. In Workshop on Generative AI and Law (GenLaw), International Confer- ence on Machine Learning (ICML), 2023. 3

    2023

Show all 100 references
  1. [9]

    Fake-Inversion: Learning to detect images from un- seen models by inverting stable diffusion

    George Cazenavette, Avneesh Sud, Thomas Leung, and Ben Usman. Fake-Inversion: Learning to detect images from un- seen models by inverting stable diffusion. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024. 3

    2024

  2. [10]

    What makes fake images detectable? understanding prop- erties that generalize

    Lucy Chai, David Bau, Ser-Nam Lim, and Phillip Isola. What makes fake images detectable? understanding prop- erties that generalize. In European Conference on Computer Vision (ECCV), 2020. 3

    2020

  3. [11]

    https://chat.openai.com/chat, 2024

    ChatGPT. https://chat.openai.com/chat, 2024. 4, 15

    2024

  4. [12]

    Pixart- σ: Weak-to-strong training of dif- fusion transformer for 4k text-to-image generation

    Junsong Chen, Chongjian Ge, Enze Xie, Yue Wu, Lewei Yao, Xiaozhe Ren, Zhongdao Wang, Ping Luo, Huchuan Lu, and Zhenguo Li. Pixart- σ: Weak-to-strong training of dif- fusion transformer for 4k text-to-image generation. In Euro- pean Conference on Computer Vision (ECCV) , 2024. ...

    2024

  5. [13]

    Twigma: A dataset of ai- generated images with metadata from twitter

    Yiqun Chen and James Y Zou. Twigma: A dataset of ai- generated images with metadata from twitter. Conference on Neural Information Processing Systems (NeurIPS) , 36: 37748–37760, 2023. 3

    2023

  6. [14]

    Tailored visions: Enhancing text-to-image generation with personalized prompt rewriting

    Zijie Chen, Lichao Zhang, Fangsheng Weng, Lili Pan, and Zhenzhong Lan. Tailored visions: Enhancing text-to-image generation with personalized prompt rewriting. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 7727–7736, 2024. 3

    2024

  7. [15]

    What is your data worth to gpt? llm- scale data valuation with influence functions

    Sang Keun Choe et al. What is your data worth to gpt? llm- scale data valuation with influence functions. arXiv preprint arXiv:2405.13954, 2024. 3

    2024 arXiv

  8. [16]

    On the detection of synthetic images generated by diffusion mod- els

    Riccardo Corvi, Davide Cozzolino, Giada Zingarini, Gio- vanni Poggi, Koki Nagano, and Luisa Verdoliva. On the detection of synthetic images generated by diffusion mod- els. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023. 3

    2023

  9. [17]

    Raising the bar of ai-generated image detection with clip

    Davide Cozzolino, Giovanni Poggi, Riccardo Corvi, Matthias Nießner, and Luisa Verdoliva. Raising the bar of ai-generated image detection with clip. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024. 3

    2024

  10. [18]

    Imagenet: A large-scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In IEEE Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 248–255. Ieee, 2009. 6

    2009

  11. [19]

    The deepfake detection challenge (dfdc) dataset

    Brian Dolhansky, Joanna Bitton, Ben Pflaum, Jikuo Lu, Russ Howes, Menglin Wang, and Cristian Canton Ferrer. The deepfake detection challenge (dfdc) dataset. arXiv preprint arXiv:2006.07397, 2020. 3

    2006 arXiv

  12. [20]

    The llama 3 herd of models

    Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Ab- hishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. The llama 3 herd of models. arXiv preprint arXiv:2407.21783 ,

    arXiv

  13. [21]

    Online detection of ai-generated images

    David C Epstein, Ishan Jain, Oliver Wang, and Richard Zhang. Online detection of ai-generated images. In IEEE International Conference on Computer Vision (ICCV) Work- shop, 2023. 3, 7

    2023

  14. [22]

    Scaling recti- fied flow transformers for high-resolution image synthesis

    Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas M ¨uller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, et al. Scaling recti- fied flow transformers for high-resolution image synthesis. In Forty-first international conference on mach...

  15. [23]

    The journey, not the desti- nation: How data guides diffusion models

    Kristian Georgiev, Joshua Vendrow, Hadi Salman, Sung Min Park, and Aleksander Madry. The journey, not the desti- nation: How data guides diffusion models. arXiv preprint arXiv:2312.06205, 2023. 3

    2023 arXiv

  16. [24]

    Ai generation faqs, 2024

    Getty Images. Ai generation faqs, 2024. Accessed: 2024- 11-13. 1

    2024

  17. [25]

    Studying large lan- guage model generalization with influence functions

    Roger Grosse, Juhan Bae, Cem Anil, Nelson Elhage, Alex Tamkin, Amirhossein Tajdini, Benoit Steiner, Dustin Li, Esin Durmus, Ethan Perez, et al. Studying large lan- guage model generalization with influence functions. arXiv preprint arXiv:2308.03296, 2023. 3 11

    2023 arXiv

  18. [26]

    This artist is dominating ai-generated art

    Melissa Heikkil ¨a. This artist is dominating ai-generated art. and he’s not happy about it. MIT Technology Review, 2022. 1

    2022

  19. [27]

    Wildfake: A large- scale challenging dataset for ai-generated images detection

    Yan Hong, Jianfu Zhang, Jianming Feng, Haoxing Chen, Jun Lan, Huijia Zhu, and Weiqiang Wang. Wildfake: A large- scale challenging dataset for ai-generated images detection. arXiv preprint arXiv:2402.11843, 2024. 3

    2024 arXiv

  20. [28]

    T2i-compbench++: An enhanced and comprehensive benchmark for compositional text-to-image generation

    Kaiyi Huang, Chengqi Duan, Kaiyue Sun, Enze Xie, Zhen- guo Li, and Xihui Liu. T2i-compbench++: An enhanced and comprehensive benchmark for compositional text-to-image generation. IEEE Transactions on Pattern Analysis and Ma- chine Intelligence, 2025. 3

    2025

  21. [29]

    Unlearning traces the in- fluential training data of language models

    Masaru Isonuma and Ivan Titov. Unlearning traces the in- fluential training data of language models. arXiv preprint arXiv:2401.15241, 2024. 3

    2024 arXiv

  22. [30]

    Genai arena: An open evaluation platform for generative models

    Dongfu Jiang, Max Ku, Tianle Li, Yuansheng Ni, Shizhuo Sun, Rongqi Fan, and Wenhu Chen. Genai arena: An open evaluation platform for generative models. In Conference on Neural Information Processing Systems (NeurIPS), 2024. 4

    2024

  23. [31]

    Deeperforensics-1.0: A large-scale dataset for real-world face forgery detection

    Liming Jiang, Ren Li, Wayne Wu, Chen Qian, and Chen Change Loy. Deeperforensics-1.0: A large-scale dataset for real-world face forgery detection. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 3

    2020

  24. [32]

    Recognizing image style

    Sergey Karayev, Matthew Trentacoste, Helen Han, Aseem Agarwala, Trevor Darrell, Aaron Hertzmann, and Holger Winnemoeller. Recognizing image style. arXiv preprint arXiv:1311.3715, 2013. 3

    2013 arXiv

  25. [33]

    Fakeavceleb: A novel audio-video multimodal deep- fake dataset

    Hasam Khalid, Shahroz Tariq, Minha Kim, and Simon S Woo. Fakeavceleb: A novel audio-video multimodal deep- fake dataset. In Conference on Neural Information Process- ing Systems (NeurIPS), 2021. 3

    2021

  26. [34]

    Supervised contrastive learning

    Prannay Khosla, Piotr Teterwak, Chen Wang, Aaron Sarna, Yonglong Tian, Phillip Isola, Aaron Maschinot, Ce Liu, and Dilip Krishnan. Supervised contrastive learning. Conference on Neural Information Processing Systems (NeurIPS), 2020. 6, 19, 20

    2020

  27. [35]

    Pick-a-pic: An open dataset of user preferences for text-to-image genera- tion

    Yuval Kirstain, Adam Polyak, Uriel Singer, Shahbuland Matiana, Joe Penna, and Omer Levy. Pick-a-pic: An open dataset of user preferences for text-to-image genera- tion. Conference on Neural Information Processing Systems (NeurIPS), 36:36652–36663, 2023. 3

    2023

  28. [36]

    The mirrored influence hypothesis: Effi- cient data influence estimation by harnessing forward passes

    Myeongseob Ko, Feiyang Kang, Weiyan Shi, Ming Jin, Zhou Yu, and Ruoxi Jia. The mirrored influence hypothesis: Effi- cient data influence estimation by harnessing forward passes. In IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR), 2024. 3

    2024

  29. [37]

    Understanding black-box predictions via influence functions

    Pang Wei Koh and Percy Liang. Understanding black-box predictions via influence functions. In International Confer- ence on Machine Learning (ICML), 2017. 3

    2017

  30. [38]

    Introstyle: Training-free introspective style attribution using diffusion features

    Anand Kumar, Jiteng Mu, and Nuno Vasconcelos. Introstyle: Training-free introspective style attribution using diffusion features. arXiv preprint arXiv:2412.14432, 2024. 3

    2024 arXiv

  31. [39]

    Multi-concept customization of text-to-image diffusion

    Nupur Kumari, Bingliang Zhang, Richard Zhang, Eli Shechtman, and Jun-Yan Zhu. Multi-concept customization of text-to-image diffusion. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. 10

    2023

  32. [40]

    Black Forest Labs. Flux. https://github.com/ black-forest-labs/flux, 2024. 4, 16, 17

    2024

  33. [41]

    Safety review for laion 5b, 2023

    LAION.ai. Safety review for laion 5b, 2023. Accessed on November 21, 2024. 15

    2023

  34. [42]

    Holis- tic evaluation of text-to-image models

    Tony Lee, Michihiro Yasunaga, Chenlin Meng, Yifan Mai, Joon Sung Park, Agrim Gupta, Yunzhi Zhang, Deepak Narayanan, Hannah Teufel, Marco Bellagente, et al. Holis- tic evaluation of text-to-image models. Conference on Neu- ral Information Processing Systems (NeurIPS) , 36:69981...

    2023

  35. [43]

    Not with my name! inferring artists’ names of input strings employed by diffusion models

    Roberto Leotta, Oliver Giudice, Luca Guarnera, and Sebas- tiano Battiato. Not with my name! inferring artists’ names of input strings employed by diffusion models. In Interna- tional Conference on Image Analysis and Processing, pages 364–375. Springer, 2023. 3

    2023

  36. [44]

    Celeb-df: A large-scale challenging dataset for deep- fake forensics

    Yuezun Li, Xin Yang, Pu Sun, Honggang Qi, and Siwei Lyu. Celeb-df: A large-scale challenging dataset for deep- fake forensics. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 3

    2020

  37. [45]

    K-sort arena: Ef- ficient and reliable benchmarking for generative models via k-wise human preferences

    Zhikai Li, Xuewen Liu, Dongrong Joe Fu, Jianquan Li, Qingyi Gu, Kurt Keutzer, and Zhen Dong. K-sort arena: Ef- ficient and reliable benchmarking for generative models via k-wise human preferences. In IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR), 2025. 4

    2025

  38. [46]

    Rethinking open source generative ai: open-washing and the eu ai act

    Andreas Liesenfeld and Mark Dingemanse. Rethinking open source generative ai: open-washing and the eu ai act. InPro- ceedings of the 2024 ACM Conference on Fairness, Account- ability, and Transparency, pages 1774–1787, 2024. 4

    2024

  39. [47]

    A lora is worth a thousand pictures

    Chenxi Liu, Towaki Takikawa, and Alec Jacobson. A lora is worth a thousand pictures. arXiv preprint arXiv:2412.12048,

    arXiv

  40. [48]

    Deepart: Learning joint representations of visual arts

    Hui Mao, Ming Cheung, and James She. Deepart: Learning joint representations of visual arts. InProceedings of the 25th ACM international conference on Multimedia , pages 1183– 1191, 2017. 3

    2017

  41. [49]

    Detection of gan-generated fake im- ages over social networks

    Francesco Marra, Diego Gragnaniello, Davide Cozzolino, and Luisa Verdoliva. Detection of gan-generated fake im- ages over social networks. In IEEE Conference on Multi- media Information Processing and Retrieval (MIPR) , 2018. 3

    2018

  42. [50]

    Do gans leave artificial fingerprints? In 2019 IEEE Conference on Multimedia Information Process- ing and Retrieval (MIPR), 2019

    Francesco Marra, Diego Gragnaniello, Luisa Verdoliva, and Giovanni Poggi. Do gans leave artificial fingerprints? In 2019 IEEE Conference on Multimedia Information Process- ing and Retrieval (MIPR), 2019. 3

    2019

  43. [51]

    Artists rage against machines that mimic their work

    Louise Matsakis. Artists rage against machines that mimic their work. Wired, 2023. 1

    2023

  44. [52]

    https://midjourney.com , 2024

    Midjourney. https://midjourney.com , 2024. 2, 4, 10, 15, 16, 17

    2024

  45. [53]

    Rethinking copyright infringements in the era of text-to-image generative models

    Mazda Moayeri, Samyadeep Basu, Sriram Balasubramanian, Priyatham Kattakinda, Atoosa Chegini, Robert Brauneis, and Soheil Feizi. Rethinking copyright infringements in the era of text-to-image generative models. In International Conference on Learning Representations (ICLR), 2025. 3

    2025

  46. [54]

    Detecting gan generated fake images using co-occurrence matrices

    Lakshmanan Nataraj, Tajuddin Manhar Mohammed, BS Manjunath, Shivkumar Chandrasekaran, Arjuna Flenner, Jawadul H Bappy, and Amit K Roy-Chowdhury. Detecting gan generated fake images using co-occurrence matrices. In Electronic Imaging, 2019. 3 12

    2019

  47. [55]

    Towards uni- versal fake image detectors that generalize across generative models

    Utkarsh Ojha, Yuheng Li, and Yong Jae Lee. Towards uni- versal fake image detectors that generalize across generative models. In IEEE Conference on Computer Vision and Pat- tern Recognition (CVPR), 2023. 3

    2023

  48. [56]

    Dinov2: Learning robust visual features without super- vision

    Maxime Oquab, Timoth ´ee Darcet, Th´eo Moutakanni, Huy V V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel HAZIZA, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without super- vision. Transactions on Machine Learning Research,...

    2024

  49. [57]

    Community forensics: Using thousands of generators to train fake image detectors

    Jeongsoo Park and Andrew Owens. Community forensics: Using thousands of generators to train fake image detectors. In IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR), 2025. 3

    2025

  50. [58]

    A self-supervised descriptor for image copy detection

    Ed Pizzi, Sreya Dutta Roy, Sugosh Nagavara Ravindra, Priya Goyal, and Matthijs Douze. A self-supervised descriptor for image copy detection. In IEEE Conference on Computer Vi- sion and Pattern Recognition (CVPR), pages 14532–14542,

  51. [59]

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

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis. In International Con- ference on Learning Representations (ICLR), 2024. 2, 4, 15, 16, 17

    2024

  52. [60]

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

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

    2021

  53. [61]

    Hierarchical text-conditional image gen- eration with clip latents

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. Hierarchical text-conditional image gen- eration with clip latents. arXiv preprint arXiv:2204.06125,

    arXiv

  54. [62]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022. 2, 4, 15, 16, 17

    2022

  55. [63]

    FaceForen- sics++: Learning to detect manipulated facial images

    Andreas R ¨ossler, Davide Cozzolino, Luisa Verdoliva, Chris- tian Riess, Justus Thies, and Matthias Nießner. FaceForen- sics++: Learning to detect manipulated facial images. In IEEE International Conference on Computer Vision (ICCV),

  56. [64]

    Dreambooth: Fine tuning text-to-image diffusion models for subject-driven generation

    Nataniel Ruiz, Yuanzhen Li, Varun Jampani, Yael Pritch, Michael Rubinstein, and Kfir Aberman. Dreambooth: Fine tuning text-to-image diffusion models for subject-driven generation. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 22500–22510, 2023. 10

    2023

  57. [65]

    Aladin: All layer adaptive instance normalization for fine- grained style similarity

    Dan Ruta, Saeid Motiian, Baldo Faieta, Zhe Lin, Hailin Jin, Alex Filipkowski, Andrew Gilbert, and John Collomosse. Aladin: All layer adaptive instance normalization for fine- grained style similarity. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. 3

    2021

  58. [66]

    Deep transfer learning for art classification problems

    Matthia Sabatelli, Mike Kestemont, Walter Daelemans, and Pierre Geurts. Deep transfer learning for art classification problems. In Proceedings Of The European conference on computer vision (ECCV) workshops, 2018. 3

    2018

  59. [67]

    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, et al. Photorealistic text-to-image diffusion models with deep language understanding. Conference on Neural Informati...

    2022

  60. [68]

    Large-scale classifica- tion of fine-art paintings: Learning the right metric on the right feature

    Babak Saleh and Ahmed Elgammal. Large-scale classifica- tion of fine-art paintings: Learning the right metric on the right feature. arXiv preprint arXiv:1505.00855, 2015. 3

    2015 arXiv

  61. [69]

    https : / / huggingface

    Gustavo Santana. https : / / huggingface . co / datasets / Gustavosta / Stable - Diffusion - Prompts, 2024. 3, 15

    2024

  62. [70]

    Laion-5b: An open large-scale dataset for training next generation image-text models

    Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Worts- man, et al. Laion-5b: An open large-scale dataset for training next generation image-text models. Conference on Neural Inf...

    2022

  63. [71]

    De-fake: Detection and attribution of fake images generated by text- to-image generation models

    Zeyang Sha, Zheng Li, Ning Yu, and Yang Zhang. De-fake: Detection and attribution of fake images generated by text- to-image generation models. In Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communica- tions Security, 2023. 3

    2023

  64. [72]

    Art- brain: An explainable end-to-end toolkit for classification and attribution of ai-generated art and style

    Ravidu Suien Rammuni Silva, Ahmad Lotfi, Isibor Kennedy Ihianle, Golnaz Shahtahmassebi, and Jordan J Bird. Art- brain: An explainable end-to-end toolkit for classification and attribution of ai-generated art and style. arXiv preprint arXiv:2412.01512, 2024. 3

    2024 arXiv

  65. [73]

    Prototypical networks for few-shot learning

    Jake Snell, Kevin Swersky, and Richard Zemel. Prototypical networks for few-shot learning. Conference on Neural Infor- mation Processing Systems (NeurIPS) , 2017. 6, 8, 10, 16, 18, 19, 20, 21, 22

    2017

  66. [74]

    Artists rights 101

    Artists Rights Society. Artists rights 101. Artists Rights So- ciety, 2024. 16

    2024

  67. [75]

    Diffusion art or digital forgery? investigating data replication in diffusion models

    Gowthami Somepalli, Vasu Singla, Micah Goldblum, Jonas Geiping, and Tom Goldstein. Diffusion art or digital forgery? investigating data replication in diffusion models. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. 8

    2023

  68. [76]

    Investigating style similarity in diffusion models

    Gowthami Somepalli, Anubhav Gupta, Kamal Gupta, Shra- may Palta, Micah Goldblum, Jonas Geiping, Abhinav Shri- vastava, and Tom Goldstein. Investigating style similarity in diffusion models. In European Conference on Computer Vision (ECCV), 2024. 3, 6, 10, 15, 18, 21, 22

    2024

  69. [77]

    Omniart: a large- scale artistic benchmark

    Gjorgji Strezoski and Marcel Worring. Omniart: a large- scale artistic benchmark. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 14 (4):1–21, 2018. 3

    2018

  70. [78]

    Journeydb: A benchmark for generative im- age understanding

    Keqiang Sun, Junting Pan, Yuying Ge, Hao Li, Haodong Duan, Xiaoshi Wu, Renrui Zhang, Aojun Zhou, Zipeng Qin, Yi Wang, et al. Journeydb: A benchmark for generative im- age understanding. Conference on Neural Information Pro- cessing Systems (NeurIPS), 2024. 2, 3, 4, 5, 15, 16

    2024

  71. [79]

    Artwhisperer: A dataset for characterizing human-ai interactions in artistic creations

    Kailas V odrahalli and James Zou. Artwhisperer: A dataset for characterizing human-ai interactions in artistic creations. 13 In International Conference on Machine Learning , pages 49627–49654. PMLR, 2024. 3

    2024

  72. [80]

    Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, and Alexei A. Efros. Cnn-generated images are sur- prisingly easy to spot. . . for now. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 3, 16

    2020

  73. [81]

    Evaluating data attribution for text-to-image mod- els

    Sheng-Yu Wang, Alexei A Efros, Jun-Yan Zhu, and Richard Zhang. Evaluating data attribution for text-to-image mod- els. In IEEE International Conference on Computer Vision (ICCV), 2023. 3, 6, 8, 21, 22

    2023

  74. [82]

    Data attribution for text-to-image models by unlearning synthesized images

    Sheng-Yu Wang, Aaron Hertzmann, Alexei A Efros, Jun-Yan Zhu, and Richard Zhang. Data attribution for text-to-image models by unlearning synthesized images. In Conference on Neural Information Processing Systems (NeurIPS), 2024. 3

    2024

  75. [83]

    Dire for diffusion-generated image detection

    Zhendong Wang, Jianmin Bao, Wengang Zhou, Weilun Wang, Hezhen Hu, Hong Chen, and Houqiang Li. Dire for diffusion-generated image detection. In IEEE International Conference on Computer Vision (ICCV), 2023. 3

    2023

  76. [84]

    Diffu- siondb: A large-scale prompt gallery dataset for text-to- image generative models

    Zijie J Wang, Evan Montoya, David Munechika, Haoyang Yang, Benjamin Hoover, and Duen Horng Chau. Diffu- siondb: A large-scale prompt gallery dataset for text-to- image generative models. In The Annual Meeting of the As- sociation for Computational Linguistics (ACL), 2023. 3

    2023

  77. [85]

    Not only generative art: Stable diffusion for content-style disentangle- ment in art analysis

    Yankun Wu, Yuta Nakashima, and Noa Garcia. Not only generative art: Stable diffusion for content-style disentangle- ment in art analysis. In Proceedings of the 2023 ACM Inter- national conference on multimedia retrieval, pages 199–208,

    2023

  78. [86]

    Imagere- ward: Learning and evaluating human preferences for text- to-image generation

    Jiazheng Xu, Xiao Liu, Yuchen Wu, Yuxuan Tong, Qinkai Li, Ming Ding, Jie Tang, and Yuxiao Dong. Imagere- ward: Learning and evaluating human preferences for text- to-image generation. Conference on Neural Information Pro- cessing Systems (NeurIPS), 36:15903–15935, 2023. 3

    2023

  79. [87]

    Ip- adapter: Text compatible image prompt adapter for text-to- image diffusion models

    Hu Ye, Jun Zhang, Sibo Liu, Xiao Han, and Wei Yang. Ip- adapter: Text compatible image prompt adapter for text-to- image diffusion models. arXiv preprint arXiv:2308.06721,

    arXiv

  80. [88]

    Scaling autoregressive models for content-rich text-to-image generation

    Jiahui Yu, Yuanzhong Xu, Jing Yu Koh, Thang Luong, Gun- jan Baid, Zirui Wang, Vijay Vasudevan, Alexander Ku, Yin- fei Yang, Burcu Karagol Ayan, et al. Scaling autoregressive models for content-rich text-to-image generation. Transac- tions on Machine Learning Research, 2022. 3

    2022

  81. [89]

    Attributing fake images to gans: Learning and analyzing gan fingerprints

    Ning Yu, Larry S Davis, and Mario Fritz. Attributing fake images to gans: Learning and analyzing gan fingerprints. In IEEE International Conference on Computer Vision (ICCV),

  82. [90]

    Detect- ing and simulating artifacts in gan fake images

    Xu Zhang, Svebor Karaman, and Shih-Fu Chang. Detect- ing and simulating artifacts in gan fake images. In IEEE In- ternational Workshop on Information Forensics and Security (WIFS), 2019. 3

    2019

  83. [91]

    Evolvedirector: Approaching advanced text-to-image generation with large vision-language models

    Rui Zhao, Hangjie Yuan, Yujie Wei, Shiwei Zhang, Yuchao Gu, Lingmin Ran, Xiang Wang, Jay Zhangjie Wu, David Junhao Zhang, Yingya Zhang, et al. Evolvedirector: Approaching advanced text-to-image generation with large vision-language models. Conference on Neural Information Proc...

    2024

  84. [92]

    Stylebreeder: Exploring and democratizing artistic styles through text-to- image models

    Matthew Zheng, Enis Simsar, Hidir Yesiltepe, Federico Tombari, Joel Simon, and Pinar Yanardag. Stylebreeder: Exploring and democratizing artistic styles through text-to- image models. In Conference on Neural Information Pro- cessing Systems (NeurIPS), 2024. 3

    2024

  85. [93]

    Intriguing properties of data attribution on diffusion models

    Xiaosen Zheng, Tianyu Pang, Chao Du, Jing Jiang, and Min Lin. Intriguing properties of data attribution on diffusion models. In International Conference on Learning Represen- tations (ICLR), 2024. 3

    2024

  86. [94]

    Genimage: A million-scale benchmark for de- tecting ai-generated image

    Mingjian Zhu, Hanting Chen, Qiangyu Yan, Xudong Huang, Guanyu Lin, Wei Li, Zhijun Tu, Hailin Hu, Jie Hu, and Yunhe Wang. Genimage: A million-scale benchmark for de- tecting ai-generated image. In Conference on Neural Infor- mation Processing Systems (NeurIPS), 2023. 3

    2023

  87. [95]

    Wilddeepfake: A challenging real-world dataset for deepfake detection

    Bojia Zi, Minghao Chang, Jingjing Chen, Xingjun Ma, and Yu-Gang Jiang. Wilddeepfake: A challenging real-world dataset for deepfake detection. In Proceedings of the 28th ACM International Conference on Multimedia, 2020. 3 14 Supplementary Material We include additional dataset ...

    2020

  88. [97]

    A picture of <content> in the style of <artist>

    Dataset Curation Details 7.1. Artist Curation As mentioned in Section 3 of the main text, to collect a list of artists commonly used in text-to-image models, we manually de-duplicate and filter our set of seen artists from an initial list of 400 artists frequently prompted by ...

  89. [98]

    Training Details Prototypical Network. We train the prototypical net- work [73] for prompted artist identification by fine-tuning the pre-trained CLIP [60] ViT-L/14 image encoder end-to- end on our prompted artist identification dataset. We ap- ply blur and JPEG training data ...

  90. [99]

    All experiments were conducted with SDXL images from our prompted artist identification dataset

    Experiments and Ablations We analyze and validate the design choices for each of the visual representation methods we evaluate on our bench- mark in the main paper. All experiments were conducted with SDXL images from our prompted artist identification dataset. All complex pro...

  91. [100]

    Bootstrapping Procedure To estimate the statistical significance of each evaluation on our benchmark, we use a bootstrapping procedure

    Evaluation Details 10.1. Bootstrapping Procedure To estimate the statistical significance of each evaluation on our benchmark, we use a bootstrapping procedure. We bootstrap each model’s predictions by resampling the evalu- ation images, with replacement, by artist name, promp...

  92. [2024]

    Accessed: November 20, 2024. 15

    2024

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.