Your Unlearning Gives You Away: Identifying Erased Concepts in Diffusion Models

arXiv:2610.05601 · cs.LG, cs.AI · Submitted 2026-10-04 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-10-04

Updated: 2026-10-04

Code: https://github.com/fastai/imagenette

Terminology

• Diffusion Model — 1× in this paper · explained in 9 episodes
• Diffusion Models — 1× in this paper · explained in 6 episodes

Sources

• Erasing Undesirable Concepts in Diffusion Models with Adversarial Preservation
• Fantastic Targets for Concept Erasure in Diffusion Models and Where To Find Them
• ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning
• Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts
• SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
• Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking
• Forget Many, Forget Right: Scalable and Precise Concept Unlearning in Diffusion Models
• SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
• A Probabilistic Fluctuation based Membership Inference Attack for Diffusion Models
• Recovering the Pre-Fine-Tuning Weights of Generative Models
• Z-Erase: Enabling Concept Erasure in Single-Stream Diffusion Transformers
• One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Framework
• AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models
• SPEED: Scalable, Precise, and Efficient Concept Erasure for Diffusion Models
• Unlearning Concepts from Text-to-Video Diffusion Models
• Circumventing Concept Erasure Methods For Text-to-Image Generative Models
• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis
• Wan: Open and Advanced Large-Scale Video Generative Models
• Membership Inference Attacks Against Text-to-image Generation Models
• TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Models

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks