mlsanityData to 7 Oct 2026

Scissorhands: Scrub Data Influence via Connection Sensitivity in Networks

arXiv 2401.06187 ↗First posted 11 Jan 2024cs.LG cs.CVcited by 50

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

Machine unlearning has become a pivotal task to erase the influence of data from a trained model. It adheres to recent data regulation standards and enhances the privacy and security of machine learning applications. In this work, we present a new machine unlearning approach Scissorhands. Initially, Scissorhands identifies the most pertinent parameters in the given model relative to the forgetting data via connection sensitivity. By reinitializing the most influential top-k percent of these parameters, a trimmed model for erasing the influence of the forgetting data is obtained. Subsequently, Scissorhands fine-tunes the trimmed model with a gradient projection-based approach, seeking parameters that preserve information on the remaining data while discarding information related to the forgetting data. Our experimental results, conducted across image classification and image generation tasks, demonstrate that Scissorhands, showcases competitive performance when compared to existing methods. Source code is available at https://github.com/JingWu321/Scissorhands.

Key facts. Cited by 50 papers in the map (data through 7 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 54 of its 75 references that are in the map. First posted on arXiv 11 Jan 2024.

Builds on

54 of 75 references.

Closest ancestors · 13

Same direction, newest first

  1. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation19 Oct
  2. 2023Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening15 Aug
  3. 2023Boundary Unlearning21 Mar
  4. 2022Deep Regression Unlearning15 Oct
  5. 2022A Survey of Machine Unlearning6 Sep
  6. 2022Evaluating Machine Unlearning via Epistemic Uncertainty23 Aug
  7. 2022Deep Unlearning via Randomized Conditionally Independent Hessians15 Apr
  8. 2022Towards Adversarial Evaluations for Inexact Machine Unlearning17 Jan
  9. 2022Zero-Shot Machine Unlearning14 Jan
  10. 2021Fast Yet Effective Machine Unlearning17 Nov
  11. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearning27 Sep
  12. 2021Machine Unlearning of Features and Labels26 Aug
1 more
  1. 2021Remember What You Want to Forget: Algorithms for Machine Unlearning4 Mar

Other references · 41

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  5. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  6. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  7. 2017Gradient Episodic Memory for Continual Learningcited by 1,871
  8. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  9. 2019Similarity of Neural Network Representations Revisitedcited by 1,695
  10. 2019Machine Unlearningcited by 1,020
  11. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
  12. 2018SNIP: Single-shot Network Pruning based on Connection Sensitivitycited by 802
29 more
  1. 2019MaskGAN: Towards Diverse and Interactive Facial Image Manipulationcited by 624
  2. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 599
  3. 2023Erasing Concepts from Diffusion Modelscited by 530
  4. 2019Certified Data Removal from Machine Learning Modelscited by 512
  5. 2022Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Modelscited by 501
  6. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 447
  7. 2023Unified Concept Editing in Diffusion Modelscited by 377
  8. 2020Descent-to-Delete: Gradient-Based Methods for Machine Unlearningcited by 303
  9. 2023Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Modelscited by 297
  10. 2022Red-Teaming the Stable Diffusion Safety Filtercited by 257
  11. 2020Approximate Data Deletion from Machine Learning Modelscited by 250
  12. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  13. 2023Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Modelscited by 175
  14. 2020Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observationscited by 171
  15. 2015On Wasserstein Two Sample Testing and Related Families of Nonparametric Testscited by 165
  16. 2020Mixed-Privacy Forgetting in Deep Networkscited by 158
  17. 2022Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?cited by 150
  18. 2020DeltaGrad: Rapid retraining of machine learning modelscited by 143
  19. 2021Graph Unlearningcited by 129
  20. 2022The Right to be Forgotten in Federated Learning: An Efficient Realization with Rapid Retrainingcited by 109
  21. 2021Federated Unlearning via Class-Discriminative Pruningcited by 108
  22. 2019A Signal Propagation Perspective for Pruning Neural Networks at Initializationcited by 96
  23. 2022Federated Unlearning: How to Efficiently Erase a Client in FL?cited by 90
  24. 2020Machine Unlearning: Linear Filtration for Logit-based Classifierscited by 89
  25. 2023TrojDiff: Trojan Attacks on Diffusion Models with Diverse Targetscited by 88
  26. 2022Federated Unlearning with Knowledge Distillationcited by 70
  27. 2023FACET: Fairness in Computer Vision Evaluation Benchmarkcited by 58
  28. 2023GNNDelete: A General Strategy for Unlearning in Graph Neural Networkscited by 43
  29. 2021SSSE: Efficiently Erasing Samples from Trained Machine Learning Modelscited by 28

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 7; 2025: 23; 2026: 20.

50 papers.

Sort:

By area:

  1. 2024Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Modelscited by 164
  2. 2025SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoderscited by 75
  3. 2024ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruningcited by 51
  4. 2024UnlearnCanvas: Stylized Image Dataset for Enhanced Machine Unlearning Evaluation in Diffusion Modelscited by 47
  5. 2025Learning to Unlearn while Retaining: Combating Gradient Conflicts in Machine Unlearningcited by 19
  6. 2024MUNBa: Machine Unlearning via Nash Bargainingcited by 17
  7. 2024Targeted Unlearning with Single Layer Unlearning Gradientcited by 16
  8. 2024Unstable Unlearning: The Hidden Risk of Concept Resurgence in Diffusion Modelscited by 14
  9. 2025One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Frameworkcited by 14
  10. 2024Memories of Forgotten Conceptscited by 9
  11. 2025Rethinking Machine Unlearning in Image Generation Modelscited by 9
  12. 2025Sparse Autoencoder as a Zero-Shot Classifier for Concept Erasing in Text-to-Image Diffusion Modelscited by 8
38 more
  1. 2025SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoderscited by 8
  2. 2025CE-SDWV: Effective and Efficient Concept Erasure for Text-to-Image Diffusion Models via a Semantic-Driven Word Vocabularycited by 7
  3. 2025Go Beyond Your Means: Unlearning with Per-Sample Gradient Orthogonalizationcited by 7
  4. 2025UnGuide: Learning to Forget with LoRA-Guided Diffusion Modelscited by 7
  5. 2025Concept Unlearning by Modeling Key Steps of Diffusion Processcited by 5
  6. 2025Locality-Aware Continual Unlearning for Diffusion Modelscited by 5
  7. 2025Reference-Specific Unlearning Metrics Can Hide the Truth: A Reality Checkcited by 4
  8. 2025EMMA: Concept Erasure Benchmark with Comprehensive Semantic Metrics and Diverse Categoriescited by 4
  9. 2025Module-Aware Parameter-Efficient Machine Unlearning on Transformerscited by 3
  10. 2026UnHype: CLIP-Guided Hypernetworks for Dynamic LoRA Unlearningcited by 3
  11. 2026AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Modelscited by 3
  12. 2026RAZOR: Ratio-Aware Layer Editing for Targeted Unlearning in Vision Transformers and Diffusion Modelscited by 3
  13. 2025Rethinking the Vulnerability of Concept Erasure and a New Methodcited by 2
  14. 2025Memory Self-Regeneration: Uncovering Hidden Knowledge in Unlearned Modelscited by 2
  15. 2025SPQR: A Multi-Dimensional Benchmark for Safety Alignment under Benign Model Adaptationcited by 2
  16. 2026Localized Concept Erasure in Text-to-Image Diffusion Models via High-Level Representation Misdirectioncited by 2
  17. 2026TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Modelscited by 2
  18. 2025RUB: Evaluating Residual Knowledge in Unlearned Modelscited by 1
  19. 2025IMU: Influence-guided Machine Unlearningcited by 1
  20. 2025Rethinking Robust Adversarial Concept Erasure in Diffusion Modelscited by 1
  21. 2025Towards Irreversible Machine Unlearning for Diffusion Modelscited by 1
  22. 2026Critic-Guided Reinforcement Unlearning in Text-to-Image Diffusioncited by 1
  23. 2026Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion Modelscited by 1
  24. 2026Erasure or Erosion? Evaluating Compositional Degradation in Unlearned Text-To-Image Diffusion Modelscited by 1
  25. 2026Beyond Text Prompts: Precise Concept Erasure through Text-Image Collaborationcited by 1
  26. 2026Evaluation without Generation: Non-Generative Assessment of Harmful Model Specialization with Applications to CSAMcited by 1
  27. 2026Disentangled Sparse Representations for Concept-Separated Diffusion Unlearningcited by 1
  28. 2025Few-Shot Concept Unlearning with Low Rank Adaptationcited by 0
  29. 2025Bi-Erasing: A Bidirectional Framework for Concept Removal in Diffusion Modelscited by 0
  30. 2026Forgetting is Competition: Rethinking Unlearning as Representation Interference in Diffusion Modelscited by 0
  31. 2026Compensation-free Machine Unlearning in Text-to-Image Diffusion Models by Eliminating the Mutual Informationcited by 0
  32. 2026Empty SPACE: Cross-Attention Sparsity for Concept Erasure in Diffusion Modelscited by 0
  33. 2026Concept Removal for Frontier Image Generative Modelscited by 0
  34. 2026Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Modelscited by 0
  35. 2026DriftScope: Measuring The Hidden Effects of Diffusion Model Adaptationcited by 0
  36. 2026TILDE: TILt-based Distributional Erasure for Concept Unlearningcited by 0
  37. 2026TINA+: Probing Residual Visual Knowledge in Unlearned Diffusion Models via Diffusion-Consistent Text-Free Inversioncited by 0
  38. 2026You Can't Have It Both Ways: Concept Entanglement Limits Diffusion Model Unlearningcited by 0