Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models
Open accessThe DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Abstract
Cutting-edge diffusion models produce images with high quality and customizability, enabling them to be used for commercial art and graphic design purposes. But do diffusion models create unique works of art, or are they replicating content directly from their training sets? In this work, we study image retrieval frameworks that enable us to compare generated images with training samples and detect when content has been replicated. Applying our frameworks to diffusion models trained on multiple datasets including Oxford flowers, Celeb-A, ImageNet, and LAION, we discuss how factors such as training set size impact rates of content replication. We also identify cases where diffusion models, including the popular Stable Diffusion model, blatantly copy from their training data. Project page: https://somepago.github.io/diffrep.html
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00986 · Cited by 35,534
Rethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138
High-Resolution Image Synthesis with Latent Diffusion Models
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01042 · Cited by 20,302
Learning Transferable Visual Models From Natural Language Supervision
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al. (2021) · arXiv · doi:10.48550/arxiv.2103.00020 · Cited by 15,293
Denoising Diffusion Probabilistic Models
Ho, J., Jain, A., & Abbeel, P. (2020) · arXiv · doi:10.48550/arxiv.2006.11239 · Cited by 8,712
Masked Autoencoders Are Scalable Vision Learners
He, K., Chen, X., Xie, S., Li, Y., Dollar, P., & Girshick, R. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01553 · Cited by 8,489
The Pascal Visual Object Classes Challenge: A Retrospective
Everingham, M., Eslami, S. M. A., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-014-0733-5 · Cited by 7,139
Representation Learning with Contrastive Predictive Coding
Oord, A. V. D., Li, Y., & Vinyals, O. (2018) · arXiv · doi:10.48550/arxiv.1807.03748 · Cited by 6,242 · cited 1×
“SSCD [47] is a self-supervised copy detection method trained in the style of SimCLR [13] using InfoNCE loss [43], entropy regularization on latent space representations, and many strong augmentations.”
MethodEmerging Properties in Self-Supervised Vision Transformers
Caron, M., Touvron, H., Misra, I., Jegou, H., Mairal, J., Bojanowski, P., & Joulin, A. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00951 · Cited by 5,638 · cited 5×
“0 DINO [12], ViT-B/16, split-product 32.”
BackgroundMethodInfluentialCutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
Yun, S., Han, D., Oh, S. J., Chun, S., Choe, J., & Yoo, Y. (2019) · arXiv · doi:10.48550/arxiv.1905.04899 · Cited by 5,439
CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features
Yun, S., Han, D., Chun, S., Oh, S. J., Yoo, Y., & Choe, J. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00612 · Cited by 5,400
Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning
Grill, J.-B., Strub, F., Altché, F., Tallec, C., Richemond, P. H., Buchatskaya, E., Doersch, C., Pires, B. A., Guo, Z. D., Azar, M. G., et al. (2020) · arXiv · doi:10.48550/arxiv.2006.07733 · Cited by 3,846 · cited 1×
“A natural basis for image retrieval methods are self-supervised models that inherently learn strong feature descriptors, matching similar images to similar representations [11,13,15,29,31].”
MethodAutomated Flower Classification over a Large Number of Classes
Nilsback, M.-E., & Zisserman, A. (2008) · 2008 Sixth Indian Conference on Computer Vision, Graphics & Image Processing · doi:10.1109/icvgip.2008.47 · Cited by 3,442 · cited 4×
“For Oxford Flowers [41], we train models on 100, 1083 (top 5 classes), and 8189 (complete dataset) images.”
MethodInfluentialDeep Clustering for Unsupervised Learning of Visual Features
Caron, M., Bojanowski, P., Joulin, A., & Douze, M. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01264-9_9 · Cited by 2,674
Hierarchical Text-Conditional Image Generation with CLIP Latents
Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., & Chen, M. (2022) · arXiv · doi:10.48550/arxiv.2204.06125 · Cited by 2,511 · cited 1×
“These mega-datasets have the power to forge commercial models like DALL·E [54] and Stable Diffusion [56], but also bring with them a number of legal and ethical risks [7].”
BackgroundRePaint: Inpainting using Denoising Diffusion Probabilistic Models
Lugmayr, A., Danelljan, M., Romero, A., Yu, F., Timofte, R., & Van Gool, L. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01117 · Cited by 1,778 · cited 2×
“Our IN-Dif-Patch dataset is created by masking 80% of the image except for a random square patch and then outpainting the rest of the pixels using the method proposed by Lugmayr et al [38].”
MethodAn Empirical Study of Training Self-Supervised Vision Transformers
Chen, X., Xie, S., & He, K. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00950 · Cited by 1,648 · cited 3×
“To see why this is a problem, consider an example in which generated image I gen contains a car and a tree directly stolen from two unrelated images I car and I tree , respectively. d Then we would like d ( gen , I car ) and ( gen , I tree ) to be very small indicating replication.”
BackgroundMethodA Closer Look at Memorization in Deep Networks
Arpit, D., Jastrzębski, S., Ballas, N., Krueger, D., Bengio, E., Kanwal, M. S., Maharaj, T., Fischer, A., Courville, A., Bengio, Y., et al. (2017) · arXiv · doi:10.48550/arxiv.1706.05394 · Cited by 1,307 · cited 1×
“To ML theorists, memorization is synonymous with overfitting [2, 21, 23].”
BackgroundLAION-5B: An open large-scale dataset for training next generation image-text models
Schuhmann, C., Beaumont, R., Vencu, R., Gordon, C., Wightman, R., Cherti, M., Coombes, T., Katta, A., Mullis, C., Wortsman, M., et al. (2022) · arXiv · doi:10.48550/arxiv.2210.08402 · Cited by 1,022
Deep Image Retrieval: Learning Global Representations for Image Search
Gordo, A., Almazán, J., Revaud, J., & Larlus, D. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46466-4_15 · Cited by 787 · cited 1×
“High-performance descriptors can be fine-tuned specifically for retrieval after unsupervised training [49, 50] using structure-from-motion (SfM) or contrastive objectives [14, 27].”
BackgroundHigh-Resolution Image Synthesis with Latent Diffusion Models
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2021) · arXiv · doi:10.48550/arxiv.2112.10752 · Cited by 737
Extracting Training Data from Large Language Models
Carlini, N., Tramer, F., Wallace, E., Jagielski, M., Herbert-Voss, A., Lee, K., Roberts, A., Brown, T., Song, D., Erlingsson, U., et al. (2020) · arXiv · doi:10.48550/arxiv.2012.07805 · Cited by 727 · cited 2×
“It is well known that generative language models risk replication from their training set [9, 10] and the amount of replicated data is broadly proportional to the size of the model, amount of duplication of the data point in the training set, and the amount of prompting.”
BackgroundDreaming to Distill: Data-Free Knowledge Transfer via DeepInversion
Yin, H., Molchanov, P., Alvarez, J. M., Li, Z., Mallya, A., Hoiem, D., Jha, N. K., & Kautz, J. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00874 · Cited by 540
Does learning require memorization? a short tale about a long tail
Feldman, V. (2020) · Proceedings of the 52nd Annual ACM SIGACT Symposium on Theory of Computing · doi:10.1145/3357713.3384290 · Cited by 287
[Paper] Visual Instance Retrieval with Deep Convolutional Networks
Razavian, A. S., Sullivan, J., Carlsson, S., & Maki, A. (2016) · ITE Transactions on Media Technology and Applications · doi:10.3169/mta.4.251 · Cited by 276
Quantifying Memorization Across Neural Language Models
Carlini, N., Ippolito, D., Jagielski, M., Lee, K., Tramer, F., & Zhang, C. (2022) · arXiv · doi:10.48550/arxiv.2202.07646 · Cited by 204 · cited 2×
“It is well known that generative language models risk replication from their training set [9, 10] and the amount of replicated data is broadly proportional to the size of the model, amount of duplication of the data point in the training set, and the amount of prompting.”
BackgroundMultimodal datasets: misogyny, pornography, and malignant stereotypes
Birhane, A., Prabhu, V. U., & Kahembwe, E. (2021) · arXiv · doi:10.48550/arxiv.2110.01963 · Cited by 160
Training Vision Transformers for Image Retrieval
El-Nouby, A., Neverova, N., Laptev, I., & Jégou, H. (2021) · arXiv · doi:10.48550/arxiv.2102.05644 · Cited by 127 · cited 1×
“Recent approaches adopt strong vision transformers as architectural backbones for retrieval [6, 19, 27, 35, 61].”
BackgroundA Self-Supervised Descriptor for Image Copy Detection
Pizzi, E., Roy, S. D., Ravindra, S. N., Goyal, P., & Douze, M. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01413 · Cited by 101 · cited 3×
“SSCD [47] is a self-supervised copy detection method trained in the style of SimCLR [13] using InfoNCE loss [43], entropy regularization on latent space representations, and many strong augmentations.”
MethodMultiGrain: a unified image embedding for classes and instances
Berman, M., Jégou, H., Vedaldi, A., Kokkinos, I., & Douze, M. (2019) · arXiv · doi:10.48550/arxiv.1902.05509 · Cited by 69
Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models
Tirumala, K., Markosyan, A., Zettlemoyer, L., & Aghajanyan, A. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-2773 · Cited by 67
Membership Inference Attacks against GANs by Leveraging Over-representation Regions
Hu, H., & Pang, J. (2021) · Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3460120.3485338 · Cited by 16
Cited by
AI Art and its Impact on Artists
Jiang, H. H., Brown, L., Cheng, J., Khan, M., Gupta, A., Workman, D., Hanna, A., Flowers, J., & Gebru, T. (2023) · Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3600211.3604681 · Cited by 265
Typology of Risks of Generative Text-to-Image Models
Bird, C., Ungless, E., & Kasirzadeh, A. (2023) · Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3600211.3604722 · Cited by 121
Scalable Extraction of Training Data from (Production) Language Models
Nasr, M., Carlini, N., Hayase, J., Jagielski, M., Cooper, A. F., Ippolito, D., Choquette-Choo, C. A., Wallace, E., Tramèr, F., & Lee, K. (2023) · arXiv · doi:10.48550/arxiv.2311.17035 · Cited by 96
Foundation Models and Fair Use
Henderson, P., Li, X., Jurafsky, D., Hashimoto, T., Lemley, M. A., & Liang, P. (2023) · doi:10.2139/ssrn.4404340 · Cited by 57
Diffusion-Based Visual Art Creation: A Survey and New Perspectives
Wang, B., Chen, Q., & Wang, Z. (2025) · ACM Computing Surveys · doi:10.1145/3728459 · Cited by 40
Metadata
- Authors
- Somepalli, G., Singla, V., Goldblum, M., Geiping, J., & Goldstein, T.
- Venue
- 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2023-06
- Pages
- 6048-6058
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4386076098
- Cited by
- 194
- References
- 33
- FWCI
- 22.06
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.