mlsanityData to 7 Oct 2026

On the De-duplication of LAION-2B

arXiv 2303.12733 ↗First posted 17 Mar 2023cs.CV cs.AIcited by 47

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Generative models, such as DALL-E, Midjourney, and Stable Diffusion, have societal implications that extend beyond the field of computer science. These models require large image databases like LAION-2B, which contain two billion images. At this scale, manual inspection is difficult and automated analysis is challenging. In addition, recent studies show that duplicated images pose copyright problems for models trained on LAION2B, which hinders its usability. This paper proposes an algorithmic chain that runs with modest compute, that compresses CLIP features to enable efficient duplicate detection, even for vast image volumes. Our approach demonstrates that roughly 700 million images, or about 30%, of LAION-2B's images are likely duplicated. Our method also provides the histograms of duplication on this dataset, which we use to reveal more examples of verbatim copies by Stable Diffusion and further justify the approach. The current version of the de-duplicated set will be distributed online.

Key facts. Cited by 47 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 16 of its 25 references that are in the map. First posted on arXiv 17 Mar 2023.

Builds on

16 of 25 references.

Closest ancestors · 1

Same direction, newest first

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec

Other references · 15

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  3. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  4. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,848
  5. 2017Billion-scale similarity search with GPUscited by 2,477
  6. 2021LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairscited by 1,544
  7. 2016Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphscited by 853
  8. 2023Extracting Training Data from Diffusion Modelscited by 723
  9. 2022Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Modelscited by 418
  10. 2017Lossy Image Compression with Compressive Autoencoderscited by 289
  11. 2023StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesiscited by 212
  12. 2022Versatile Diffusion: Text, Images and Variations All in One Diffusion Modelcited by 199
3 more
  1. 2022A Self-Supervised Descriptor for Image Copy Detectioncited by 175
  2. 2021A Survey on Locality Sensitive Hashing Algorithms and their Applicationscited by 40
  3. 2021CE-Dedup: Cost-Effective Convolutional Neural Nets Training based on Image Deduplicationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 8; 2024: 22; 2025: 14; 2026: 3.

47 papers.

Sort:

By area:

  1. 2023DataComp: In search of the next generation of multimodal datasetscited by 635
  2. 2023OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documentscited by 316
  3. 2023Understanding and Mitigating Copying in Diffusion Modelscited by 205
  4. 2024Building and better understanding vision-language models: insights and future directionscited by 169
  5. 2024Unveiling Encoder-Free Vision-Language Modelscited by 90
  6. 2024WorldDreamer: Towards General World Models for Video Generation via Predicting Masked Tokenscited by 87
  7. 2023An Inverse Scaling Law for CLIP Trainingcited by 76
  8. 2023Battle of the Backbones: A Large-Scale Comparison of Pretrained Models across Computer Vision Taskscited by 72
  9. 2024DenseFusion-1M: Merging Vision Experts for Comprehensive Multimodal Perceptioncited by 72
  10. 2023A Reproducible Extraction of Training Images from Diffusion Modelscited by 60
  11. 2024ControlVAR: Exploring Controllable Visual Autoregressive Modelingcited by 58
  12. 2024Unveiling and Mitigating Memorization in Text-to-image Diffusion Models through Cross Attentioncited by 57
35 more
  1. 2024Copyright Protection in Generative AI: A Technical Perspectivecited by 35
  2. 2024The Evolution of Multimodal Model Architecturescited by 21
  3. 2023The Devil is in the Details: A Deep Dive into the Rabbit Hole of Data Filteringcited by 19
  4. 2024The Synergy between Data and Multi-Modal Large Language Models: A Survey from Co-Development Perspectivecited by 14
  5. 2025DASH: Detection and Assessment of Systematic Hallucinations of VLMscited by 14
  6. 2024Replication in Visual Diffusion Models: A Survey and Outlookcited by 11
  7. 2025Generalization of Diffusion Models Arises with a Balanced Representation Spacecited by 10
  8. 2024Slight Corruption in Pre-training Data Makes Better Diffusion Modelscited by 9
  9. 2024mOSCAR: A Large-scale Multilingual and Multimodal Document-level Corpuscited by 8
  10. 2024U Can't Gen This? A Survey of Intellectual Property Protection Methods for Data in Generative AIcited by 7
  11. 2024An Inversion-based Measure of Memorization for Diffusion Modelscited by 6
  12. 2024SCAN: Bootstrapping Contrastive Pre-training for Data Efficiencycited by 6
  13. 2025Neglected Risks: The Disturbing Reality of Children's Images in Datasets and the Urgent Call for Accountabilitycited by 6
  14. 2025How Diffusion Models Memorizecited by 6
  15. 2023Memory Triggers: Unveiling Memorization in Text-To-Image Generative Models through Word-Level Duplicationcited by 5
  16. 2024MemBench: Memorized Image Trigger Prompt Dataset for Diffusion Modelscited by 5
  17. 2024AsCAN: Asymmetric Convolution-Attention Networks for Efficient Recognition and Generationcited by 5
  18. 2024Image Copy Detection for Diffusion Modelscited by 4
  19. 2025CopyJudge: Automated Copyright Infringement Identification and Mitigation in Text-to-Image Diffusion Modelscited by 4
  20. 2024MemControl: Mitigating Memorization in Diffusion Models via Automated Parameter Selectioncited by 3
  21. 2024CPSample: Classifier Protected Sampling for Guarding Training Data During Diffusioncited by 3
  22. 2024Capacity Control is an Effective Memorization Mitigation Mechanism in Text-Conditional Diffusion Modelscited by 2
  23. 2025The Devil is in the Prompts: De-Identification Traces Enhance Memorization Risks in Synthetic Chest X-Ray Generationcited by 2
  24. 2025FOCUS: Unified Vision-Language Modeling for Interactive Editing Driven by Referential Segmentationcited by 2
  25. 2025HyperVL: An Efficient and Dynamic Multimodal Large Language Model for Edge Devicescited by 2
  26. 2024LoyalDiffusion: A Diffusion Model Guarding Against Data Replicationcited by 1
  27. 2025The Cake that is Intelligence and Who Gets to Bake it: An AI Analogy and its Implications for Participationcited by 1
  28. 2025On the de-duplication of the Lakh MIDI datasetcited by 1
  29. 2025FPAN: Mitigating Replication in Diffusion Models through the Fine-Grained Probabilistic Addition of Noise to Token Embeddingscited by 0
  30. 2025Finding DoRI: Discovery of Retained Images in Diffusion Modelscited by 0
  31. 2025The Persistence of Cultural Memory: Investigating Multimodal Iconicity in Diffusion Modelscited by 0
  32. 2025INQUIRE-Search: Interactive Discovery in Large-Scale Biodiversity Databasescited by 0
  33. 2026Dynamic Cluster Data Sampling for Efficient and Long-Tail-Aware Vision-Language Pre-trainingcited by 0
  34. 2026EVOLVE: Efficient Learned Volume Compression with Variable-Rate Encoding on a Cross-Domain Databasecited by 0
  35. 2026IP Protection in the Era of Visual Generative AI: A Surveycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Mitigate Replication and Copying in Diffusion Models with Generalized Caption and Dual Fusion Enhancement
  2. 2024Not Every Image is Worth a Thousand Words: Quantifying Originality in Stable Diffusion
  3. 2024Aligned Datasets Improve Detection of Latent Diffusion-Generated Images
  4. 2024Community Forensics: Using Thousands of Generators to Train Fake Image Detectors
  5. 2025Detecting AI-Generated Images via Diffusion Snap-Back Reconstruction: A Forensic Approach
  6. 2026A Difference-in-Difference Approach to Detecting AI-Generated Images
  7. 2022Generated Faces in the Wild: Quantitative Comparison of Stable Diffusion, Midjourney and DALL-E 2
  8. 2023Token Merging for Fast Stable Diffusion