mlsanityData to 7 Oct 2026

LoReUn: Data Itself Implicitly Provides Cues to Improve Machine Unlearning

arXiv 2507.22499 ↗First posted 30 Jul 2025cs.LG cs.AIcited by 3

On the map: LLM safety and security → Vision-language and text-to-image safety

Abstract (arXiv)

Recent generative models face significant risks of producing harmful content, which has underscored the importance of machine unlearning (MU) as a critical technique for eliminating the influence of undesired data. However, existing MU methods typically assign the same weight to all data to be forgotten, which makes it difficult to effectively forget certain data that is harder to unlearn than others. In this paper, we empirically demonstrate that the loss of data itself can implicitly reflect its varying difficulty. Building on this insight, we introduce Loss-based Reweighting Unlearning (LoReUn), a simple yet effective plug-and-play strategy that dynamically reweights data during the unlearning process with minimal additional computational overhead. Our approach significantly reduces the gap between existing MU methods and exact unlearning in both image classification and generation tasks, effectively enhancing the prevention of harmful content generation in text-to-image diffusion models.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Vision-language and text-to-image safety, one of 12 directions in LLM safety and security. Builds on 52 of its 60 references that are in the map. First posted on arXiv 30 Jul 2025.

Builds on

52 of 60 references.

Closest ancestors · 3

Same direction, newest first

  1. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now18 Oct
  2. 2023Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models30 Mar
  3. 2022Red-Teaming the Stable Diffusion Safety Filter3 Oct

Other references · 49

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  5. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  7. 2019Machine Unlearningcited by 1,020
  8. 2021Membership Inference Attacks From First Principlescited by 839
  9. 2023Extracting Training Data from Diffusion Modelscited by 723
  10. 2018Learning to Reweight Examples for Robust Deep Learningcited by 670
  11. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 599
  12. 2023Erasing Concepts from Diffusion Modelscited by 530
37 more
  1. 2019Certified Data Removal from Machine Learning Modelscited by 512
  2. 2022Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Modelscited by 501
  3. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 447
  4. 2022Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Modelscited by 418
  5. 2023DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretrainingcited by 385
  6. 2023Unified Concept Editing in Diffusion Modelscited by 377
  7. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 376
  8. 2021Remember What You Want to Forget: Algorithms for Machine Unlearningcited by 360
  9. 2018Not All Samples Are Created Equal: Deep Learning with Importance Samplingcited by 323
  10. 2023Ablating Concepts in Text-to-Image Diffusion Modelscited by 312
  11. 2020Amnesiac Machine Learningcited by 305
  12. 2020Descent-to-Delete: Gradient-Based Methods for Machine Unlearningcited by 303
  13. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearningcited by 274
  14. 2020Approximate Data Deletion from Machine Learning Modelscited by 250
  15. 2021Machine Unlearning of Features and Labelscited by 223
  16. 2023Model Sparsity Can Simplify Machine Unlearningcited by 206
  17. 2023Glaze: Protecting Artists from Style Mimicry by Text-to-Image Modelscited by 200
  18. 2020Geometry-aware Instance-reweighted Adversarial Trainingcited by 198
  19. 2020fastai: A Layered API for Deep Learningcited by 183
  20. 2023Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Modelscited by 175
  21. 2023Adversarial Example Does Good: Preventing Painting Imitation from Diffusion Models via Adversarial Examplescited by 163
  22. 2023Raising the Cost of Malicious AI-Powered Image Editingcited by 160
  23. 2023Boundary Unlearningcited by 132
  24. 2021Machine Unlearning via Algorithmic Stabilitycited by 128
  25. 2015Online Batch Selection for Faster Training of Neural Networkscited by 121
  26. 2023On Provable Copyright Protection for Generative Modelscited by 97
  27. 2019Accelerating Deep Learning by Focusing on the Biggest Loserscited by 94
  28. 2020Rethinking Importance Weighting for Deep Learning under Distribution Shiftcited by 92
  29. 2023Mist: Towards Improved Adversarial Examples for Diffusion Modelscited by 91
  30. 2024What makes unlearning hard and what to do about itcited by 89
  31. 2023DoGE: Domain Reweighting with Generalization Estimationcited by 87
  32. 2021Probabilistic Margins for Instance Reweighting in Adversarial Trainingcited by 57
  33. 2024Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearningcited by 44
  34. 2024To Each (Textual Sequence) Its Own: Improving Memorized-Data Unlearning in Large Language Modelscited by 42
  35. 2021Reweighting Augmented Samples by Minimizing the Maximal Expected Losscited by 21
  36. 2025Dynamic Loss-Based Sample Reweighting for Improved Large Language Model Pretrainingcited by 16
  37. 2020Are Adversarial Examples Created Equal? A Learnable Weighted Minimax Risk for Robustness under Non-uniform Attackscited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2026Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarkscited by 6
  2. 2025RapidUn: Influence-Driven Parameter Reweighting for Efficient Large Language Model Unlearningcited by 0
  3. 2026When Retain Constraints Conflict: Mitigating Forget-Retain Interference in Tabular Datacited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Projected Gradient Unlearning for Text-to-Image Diffusion Models: Defending Against Concept Revival Attacks
  2. 2024Boosting Alignment for Post-Unlearning Text-to-Image Generative Models
  3. 2025UNO: Unlearning via Orthogonalization in Generative models
  4. 2026Compensation-free Machine Unlearning in Text-to-Image Diffusion Models by Eliminating the Mutual Information
  5. 2026PECKER: A Precisely Efficient Critical Knowledge Erasure Recipe For Machine Unlearning in Diffusion Models
  6. 2026Unlearning the Unpromptable: Prompt-free Instance Unlearning in Diffusion Models
  7. 2025Continual Unlearning for Text-to-Image Diffusion Models: A Regularization Perspective
  8. 2024Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models