mlsanityData to 7 Oct 2026

Fine-Grained Erasure in Text-to-Image Diffusion-based Foundation Models

arXiv 2503.19783 ↗First posted 25 Mar 2025cs.CVcited by 22

On the map: Diffusion models → Robustness, privacy and adversarial aspects

also in 1 more
Abstract (arXiv)

Existing unlearning algorithms in text-to-image generative models often fail to preserve the knowledge of semantically related concepts when removing specific target concepts: a challenge known as adjacency. To address this, we propose FADE (Fine grained Attenuation for Diffusion Erasure), introducing adjacency aware unlearning in diffusion models. FADE comprises two components: (1) the Concept Neighborhood, which identifies an adjacency set of related concepts, and (2) Mesh Modules, employing a structured combination of Expungement, Adjacency, and Guidance loss components. These enable precise erasure of target concepts while preserving fidelity across related and unrelated concepts. Evaluated on datasets like Stanford Dogs, Oxford Flowers, CUB, I2P, Imagenette, and ImageNet1k, FADE effectively removes target concepts with minimal impact on correlated concepts, achieving atleast a 12% improvement in retention performance over state-of-the-art methods.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 22 of its 33 references that are in the map. First posted on arXiv 25 Mar 2025.

Builds on

22 of 33 references.

Closest ancestors · 8

Same direction, newest first

  1. 2023One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications26 Dec
  2. 2023Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Erasers29 Nov
  3. 2023Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?16 Oct
  4. 2023Unified Concept Editing in Diffusion Models25 Aug
  5. 2023Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models30 Mar
  6. 2023Ablating Concepts in Text-to-Image Diffusion Models23 Mar
  7. 2023Erasing Concepts from Diffusion Models13 Mar
  8. 2022Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models9 Nov

Other references · 14

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  4. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  5. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  6. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  7. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  8. 2020Are we done with ImageNet?cited by 329
  9. 2020fastai: A Layered API for Deep Learningcited by 183
  10. 2020Editable Neural Networkscited by 154
  11. 2022When does dough become a bagel? Analyzing the remaining mistakes on ImageNetcited by 60
  12. 2023Adapt then Unlearn: Exploring Parameter Space Semantics for Unlearning in Generative Adversarial Networkscited by 11
2 more
  1. 2025Continual Unlearning for Foundational Text-to-Image Models without Generalization Erosioncited by 11
  2. 2023Automated Classification of Model Errors on ImageNetcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 8; 2026: 14.

22 papers.

Sort:

By area:

  1. 2025Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Conceptscited by 61
  2. 2025Erasing Concepts, Steering Generations: A Comprehensive Survey of Concept Suppressioncited by 10
  3. 2025SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoderscited by 8
  4. 2026Neighbor-Aware Localized Concept Erasure in Text-to-Image Diffusion Modelscited by 4
  5. 2025ACE: Attentional Concept Erasure in Diffusion Modelscited by 3
  6. 2026OrthoEraser: Coupled-Neuron Orthogonal Projection for Concept Erasurecited by 3
  7. 2026To Erase, or Not to Erase: Robust Training-Free Concept Erasure with Preservation aware Adaptive Ranked Subspace Expansioncited by 3
  8. 2025Quantum-Inspired Audio Unlearning: Towards Privacy-Preserving Voice Biometricscited by 2
  9. 2025A Unified Framework for Diffusion Model Unlearning with f-Divergencecited by 2
  10. 2025GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Modelscited by 2
  11. 2025M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Modelscited by 2
  12. 2026FlowFixer: Towards Detail-Preserving Subject-Driven Generationcited by 1
10 more
  1. 2026Unlearning the Unpromptable: Prompt-free Instance Unlearning in Diffusion Modelscited by 1
  2. 2026Relationship-Aware Safety Unlearning for Multimodal LLMscited by 0
  3. 2026Graph Propagated Projection Unlearning: A Unified Framework for Vision and Audio Discriminative Modelscited by 0
  4. 2026Empty SPACE: Cross-Attention Sparsity for Concept Erasure in Diffusion Modelscited by 0
  5. 2026BARRIER: Bounded Activation Regions for Robust Information Erasurecited by 0
  6. 2026Forgetting Has Neighbors: Localized Collateral Forgetting in Machine Unlearningcited by 0
  7. 2026AutoAnchor: Stable Diffusion Unlearning Using Cross-Attention as a Manifold Surrogatecited by 0
  8. 2026Stress Testing Concept Erasure with Large Language Model Agentscited by 0
  9. 2026MapRoute++: Surrogate-Guided Semantic Routing for Visual Concept Unlearningcited by 0
  10. 2026TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Erasure or Erosion? Evaluating Compositional Degradation in Unlearned Text-To-Image Diffusion Models
  2. 2025Continual Unlearning for Text-to-Image Diffusion Models: A Regularization Perspective
  3. 2026Erasing Thousands of Concepts: Towards Scalable and Practical Concept Erasure for Text-to-Image Diffusion Models
  4. 2024Erasing Concepts from Text-to-Image Diffusion Models with Few-shot Unlearning
  5. 2026CoreUnlearn: Rethinking Concept Unlearning through Disentangled Component-Level Erasure in Text-guided Diffusion Models
  6. 2026Forget Many, Forget Right: Scalable and Precise Concept Unlearning in Diffusion Models
  7. 2026TILDE: TILt-based Distributional Erasure for Concept Unlearning
  8. 2025Localized Concept Erasure for Text-to-Image Diffusion Models Using Training-Free Gated Low-Rank Adaptation