mlsanityData to 7 Oct 2026

Evaluating Data Attribution for Text-to-Image Models

arXiv 2306.09345 ↗First posted 15 Jun 2023cs.CV cs.LGcited by 45

Not on the map. Why?

Cited by 20 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

While large text-to-image models are able to synthesize "novel" images, these images are necessarily a reflection of the training data. The problem of data attribution in such models -- which of the images in the training set are most responsible for the appearance of a given generated image -- is a difficult yet important one. As an initial step toward this problem, we evaluate attribution through "customization" methods, which tune an existing large-scale model toward a given exemplar object or style. Our key insight is that this allows us to efficiently create synthetic images that are computationally influenced by the exemplar by construction. With our new dataset of such exemplar-influenced images, we are able to evaluate various data attribution algorithms and different possible feature spaces. Furthermore, by training on our dataset, we can tune standard models, such as DINO, CLIP, and ViT, toward the attribution problem. Even though the procedure is tuned towards small exemplar sets, we show generalization to larger sets. Finally, by taking into account the inherent uncertainty of the problem, we can assign soft attribution scores over a set of training images.

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). Builds on 62 of its 78 references that are in the map. First posted on arXiv 15 Jun 2023.

Builds on

62 of 78 references.

References · 62

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  5. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  6. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  7. 2020Denoising Diffusion Implicit Modelscited by 9,655
  8. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  9. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  10. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  11. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  12. 2017Mask R-CNNcited by 7,259
38 more
  1. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  2. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  3. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  5. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  6. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,310
  7. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  8. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  9. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  10. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  11. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  12. 2022InstructPix2Pix: Learning to Follow Image Editing Instructionscited by 2,591
  13. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  14. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,231
  15. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  16. 2020Extracting Training Data from Large Language Modelscited by 2,144
  17. 2016Density estimation using Real NVPcited by 2,029
  18. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,903
  19. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  20. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,610
  21. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  22. 2019Contrastive Multiview Codingcited by 1,297
  23. 2022Imagic: Text-Based Real Image Editing with Diffusion Modelscited by 1,172
  24. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,099
  25. 2022Plug-and-Play Diffusion Features for Text-Driven Image-to-Image Translationcited by 897
  26. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  27. 2016COCO-Stuff: Thing and Stuff Classes in Contextcited by 858
  28. 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 806
  29. 2023Extracting Training Data from Diffusion Modelscited by 723
  30. 2019Data Shapley: Equitable Valuation of Data for Machine Learningcited by 617
  31. 2020Estimating Training Data Influence by Tracing Gradient Descentcited by 565
  32. 2023Zero-shot Image-to-Image Translationcited by 503
  33. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 492
  34. 2022Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Modelscited by 418
  35. 2022DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Modelscited by 376
  36. 2021Membership Inference Attacks on Machine Learning: A Surveycited by 306
  37. 2023Your Diffusion Model is Secretly a Zero-Shot Classifiercited by 301
  38. 2019Towards Efficient Data Valuation Based on the Shapley Valuecited by 279

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 20; 2025: 11; 2026: 10.

45 papers.

Sort:

By area:

  1. 2024Measuring Style Similarity in Diffusion Modelscited by 106
  2. 2023Intriguing Properties of Data Attribution on Diffusion Modelscited by 47
  3. 2023The Journey, Not the Destination: How Data Guides Diffusion Modelscited by 42
  4. 2025A Survey on Mechanistic Interpretability for Multi-Modal Foundation Modelscited by 34
  5. 2023Stable Diffusion Exposed: Gender Bias from Prompt to Imagecited by 25
  6. 2024Data Attribution for Text-to-Image Models by Unlearning Synthesized Imagescited by 22
  7. 2024Diffusion Attribution Score: Evaluating Training Data Influence in Diffusion Modelscited by 21
  8. 2024DiffSim: Taming Diffusion Models for Evaluating Visual Similaritycited by 20
  9. 2024The Mirrored Influence Hypothesis: Efficient Data Influence Estimation by Harnessing Forward Passescited by 16
  10. 2024ProMark: Proactive Diffusion Watermarking for Causal Attributioncited by 16
  11. 2024DiffusionPID: Interpreting Diffusion via Partial Information Decompositioncited by 16
  12. 2024Customizing Text-to-Image Diffusion with Object Viewpoint Controlcited by 13
33 more
  1. 2024A Survey of Defenses Against AI-Generated Visual Media: Detection,Disruption, and Authenticationcited by 12
  2. 2024Replication in Visual Diffusion Models: A Survey and Outlookcited by 11
  3. 2025A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learningcited by 11
  4. 2024U Can't Gen This? A Survey of Intellectual Property Protection Methods for Data in Generative AIcited by 7
  5. 2024A LoRA is Worth a Thousand Picturescited by 7
  6. 2025Fast Data Attribution for Text-to-Image Modelscited by 7
  7. 2024Enhancing Training Data Attribution for Large Language Models with Fitting Error Considerationcited by 6
  8. 2025SCFlow: Implicitly Learning Style and Content Disentanglement with Flow Modelscited by 5
  9. 2024An Efficient Framework for Crediting Data Contributors of Diffusion Modelscited by 4
  10. 2026Motion Attribution for Video Generationcited by 4
  11. 2023CopyScope: Model-level Copyright Infringement Quantification in the Diffusion Workflowcited by 3
  12. 2024IntroStyle: Training-Free Introspective Style Attribution using Diffusion Featurescited by 3
  13. 2025Content ARCs: Decentralized Content Rights in the Age of Generative AIcited by 3
  14. 2025Concept-TRAK: Understanding how diffusion models learn concepts through concept-level attributioncited by 3
  15. 2025AMCR: A Framework for Assessing and Mitigating Copyright Risks in Generative Modelscited by 3
  16. 2026StyleDecoupler: Generalizable Artistic Style Disentanglementcited by 3
  17. 2024Model Lakescited by 2
  18. 2024AI Royalties -- an IP Framework to Compensate Artists & IP Holders for AI-Generated Contentcited by 2
  19. 2024FIDAVL: Fake Image Detection and Attribution using Vision-Language Modelcited by 2
  20. 2025The Iconicity of the Generated Imagecited by 2
  21. 2024Enhancing High-Energy Particle Physics Collision Analysis through Graph Data Attribution Techniquescited by 1
  22. 2025Identifying Prompted Artist Names from Generated Imagescited by 1
  23. 2026Unsupervised Synthetic Image Attribution: Alignment and Disentanglementcited by 1
  24. 2024Opt-In Art: Learning Art Styles Only from Few Examplescited by 0
  25. 2025Concept Reachability in Diffusion Models: Beyond Dataset Constraintscited by 0
  26. 2025The Persistence of Cultural Memory: Investigating Multimodal Iconicity in Diffusion Modelscited by 0
  27. 2026A Vision for Multisensory Intelligence: Sensing, Science, and Synergycited by 0
  28. 2026GUDA: Counterfactual Group-wise Training Data Attribution for Diffusion Models via Unlearningcited by 0
  29. 2026TokenTrace: Multi-Concept Attribution through Watermarked Token Recoverycited by 0
  30. 2026Contrastive-Augmented Flow Matching for Style-Content Disentanglementcited by 0
  31. 2026FrED: External Data Influence Estimation via Domain Knowledge Graph Groundingcited by 0
  32. 2026Distilling Diffusion Score Discrepancy for Efficient Training Data Attributioncited by 0
  33. 2026TrueMuse: A Benchmark for Data Attribution in Text-to-Music Modelscited by 0