mlsanityData to 14 Oct 2026

On Provable Copyright Protection for Generative Models

arXiv 2302.10870 ↗First posted 21 Feb 2023cs.LG stat.MLcited by 98

Not on the map. Why?

Cited by 28 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

There is a growing concern that learned conditional generative models may output samples that are substantially similar to some copyrighted data C that was in their training set. We give a formal definition of textitnear access-freeness (NAF) and prove bounds on the probability that a model satisfying this definition outputs a sample similar to C, even if C is included in its training set. Roughly speaking, a generative model p is textitk-NAF if for every potentially copyrighted data C, the output of p diverges by at most k-bits from the output of a model q that textitdid not access C at all. We also give generative model learning algorithms, which efficiently modify the original generative model learning algorithm in a black box manner, that output generative models with strong bounds on the probability of sampling protected content. Furthermore, we provide promising experiments for both language (transformers) and image (diffusion) generative models, showing minimal degradation in output quality while ensuring strong protections against sampling protected content.

Key facts. Cited by 98 papers in the map (data through 14 Oct 2026). Builds on 16 of its 40 references that are in the map. First posted on arXiv 21 Feb 2023.

Builds on

16 of 40 references.

References · 16

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  2. 2022Training Compute-Optimal Large Language Modelscited by 3,019
  3. 2020Extracting Training Data from Large Language Modelscited by 2,189
  4. 2022Quantifying Memorization Across Neural Language Modelscited by 763
  5. 2023Extracting Training Data from Diffusion Modelscited by 732
  6. 2021Deduplicating Training Data Makes Language Models Bettercited by 714
  7. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 412
  8. 2022Deduplicating Training Data Mitigates Privacy Risks in Language Modelscited by 314
  9. 2020Differentially Private Learning Needs Better Features (or Much More Data)cited by 235
  10. 2022Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Modelscited by 220
  11. 2021Deep Learning with Label Differential Privacycited by 117
  12. 2022Do Language Models Plagiarize?cited by 73
4 more
  1. 2019Do We Train on Test Data? Purging CIFAR of Near-Duplicatescited by 62
  2. 2018Privacy-preserving Predictioncited by 48
  3. 2020The Trade-Offs of Private Predictioncited by 22
  4. 2022Formalizing Human Ingenuity: A Quantitative Framework for Copyright Law's Substantial Similaritycited by 15

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 31; 2024: 39; 2025: 19; 2026: 9.

98 papers.

Sort:

By area:

  1. 2024TOFU: A Task of Fictitious Unlearning for LLMscited by 533
  2. 2023Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignmentcited by 401
  3. 2023Understanding and Mitigating Copying in Diffusion Modelscited by 209
  4. 2023Foundation Models and Fair Usecited by 116
  5. 2024DE-COP: Detecting Copyrighted Content in Language Models Training Datacited by 74
  6. 2023The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AIcited by 73
  7. 2023SILO Language Models: Isolating Legal Risk In a Nonparametric Datastorecited by 71
  8. 2023On the Generalization Properties of Diffusion Modelscited by 68
  9. 2024Erasing Undesirable Influence in Diffusion Modelscited by 57
  10. 2023Attention Scheme Inspired Softmax Regressioncited by 49
  11. 2023An Over-parameterized Exponential Regressioncited by 45
  12. 2023MetaCloak: Preventing Unauthorized Subject-driven Text-to-image Diffusion-based Synthesis via Meta-learningcited by 44
38 more
  1. 2024A Geometric Framework for Understanding Memorization in Generative Modelscited by 40
  2. 2024Copyright Protection in Generative AI: A Technical Perspectivecited by 38
  3. 2023SoK: Memorization in General-Purpose Large Language Modelscited by 37
  4. 2023Anticipatory Music Transformercited by 36
  5. 2024CopyBench: Measuring Literal and Non-Literal Reproduction of Copyright-Protected Text in Language Model Generationcited by 35
  6. 2024The Stronger the Diffusion Model, the Easier the Backdoor: Data Poisoning to Induce Copyright Breaches Without Adjusting Finetuning Pipelinecited by 33
  7. 2023DiffInfinite: Large Mask-Image Synthesis via Parallel Random Patch Diffusion in Histopathologycited by 29
  8. 2023How to Protect Copyright Data in Optimization of Large Language Models?cited by 26
  9. 2024Diffusion Soup: Model Merging for Text-to-Image Diffusion Modelscited by 26
  10. 2023Differentially Private Attention Computationcited by 25
  11. 2023MoPe: Model Perturbation-based Privacy Attacks on Language Modelscited by 25
  12. 2024Purifying Large Language Models by Ensembling a Small Language Modelcited by 25
  13. 2025Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Modelscited by 25
  14. 2023Can Copyright be Reduced to Privacy?cited by 24
  15. 2024CPR: Retrieval Augmented Generation for Copyright Protectioncited by 24
  16. 2024Fantastic Copyrighted Beasts and How (Not) to Generate Themcited by 24
  17. 2024Automatic Jailbreaking of the Text-to-Image Generative AI Systemscited by 23
  18. 2025Generalization through variance: how noise shapes inductive biases in diffusion modelscited by 23
  19. 2025On the Edge of Memorization in Diffusion Modelscited by 22
  20. 2023On Copyright Risks of Text-to-Image Diffusion Modelscited by 20
  21. 2024Causal Estimation of Memorisation Profilescited by 20
  22. 2023The Ethical Implications of Generative Audio Models: A Systematic Literature Reviewcited by 18
  23. 2023Beyond Classical Attention: Quantum Attention for Scalable Computationcited by 18
  24. 2024An Economic Solution to Copyright Challenges of Generative AIcited by 18
  25. 2024Can Watermarking Large Language Models Prevent Copyrighted Text Generation and Hide Training Data?cited by 18
  26. 2023Training Data Protection with Compositional Diffusion Modelscited by 17
  27. 2025Taking a Big Step: Large Learning Rates in Denoising Score Matching Prevent Memorizationcited by 17
  28. 2023Auditing and Generating Synthetic Data with Controllable Trust Trade-offscited by 13
  29. 2023What Should Data Science Education Do with Large Language Models?cited by 13
  30. 2023Unmasking Transformers: A Theoretical Approach to Data Recovery via Attention Weightscited by 12
  31. 2024Synthetic Face Datasets Generation via Latent Space Exploration from Brownian Identity Diffusioncited by 12
  32. 2024Adaptive Pre-training Data Detection for Large Language Models via Surprising Tokenscited by 11
  33. 2024Replication in Visual Diffusion Models: A Survey and Outlookcited by 11
  34. 2024Not All Similarities Are Created Equal: Leveraging Data-Driven Biases to Inform GenAI Copyright Disputescited by 10
  35. 2024Disguised Copyright Infringement of Latent Diffusion Modelscited by 10
  36. 2025The Landscape of Memorization in LLMs: Mechanisms, Measurement, and Mitigationcited by 10
  37. 2023Unpacking the Ethical Value Alignment in Big Modelscited by 9
  38. 2024An Auditing Test To Detect Behavioral Shift in Language Modelscited by 9

50 most cited of 98 All 98 (JSON)