Diffusion Model Unlearning

Follow this topic Ask about this topic

Diffusion model unlearning studies how to suppress unwanted content—such as an identity, style, object, or unsafe imagery—in image and video generators without broadly damaging their abilities. Methods include fine-tuning and distillation, localized edits to weights or attention, and training-free interventions that steer activations, embeddings, or latents during generation. Evaluation increasingly tests not only whether prompts produce the target, but also adversarial recovery, related-concept retention, compositional quality, and persistence across frames or sequential edits. A central tension remains: concepts share representations, so stronger or more robust erasure can harm neighboring concepts, while apparent suppression may leave visual knowledge recoverable through other pathways.

Latest papers

77 papers19 papers in the last four weeks, up 138% on the four weeks before. 0.1% of all new papers.

Concept Acupuncturist: An Efficient and Precise Concept Erasure Method for Diffusion Models via Sparse Autoencoders
Oct 23, 2026Yiran Xiahou, Siliang Qin, Kai Chen +2Diffusion Model Unlearning Concept Erasure
Institute of Information Engineering, Chinese Academy of Sciences, China · Department of Computer Science, Metropolitan College, Boston University, USA · Institute of Data Science, NationalUniversity of SingaporeSingapore, Singapore
Push-Pull Attentional Anchoring for Diffusion Concept Erasure
Oct 20, 2026Nattanat Chatthee, Tagon Sompong, Ekapol Chuangsuwanich +1Concept Erasure Diffusion Model Unlearning
VISTEC, Thailand · Chulalongkorn University, Thailand
Enhanced Masked Diffusion Unlearning
Oct 18, 2026Tomer Garber, Tom TirerMachine Unlearning Diffusion Model Unlearning
Faculty of Engineering, Bar-Ilan University, Ramat Gan, Israel · Red Hat
BEM: Learning Bregman Divergences for Concept Erasure in Flow Matching Models
Oct 17, 2026Mohamed-Hicham Leghettas, Markus PüschelDiffusion Model Unlearning Flow Matching
Department of Computer Science, ETH Zurich, Switzerland
Compositional Concepts Erasing: Disentangling Object-Style via Domain Manifold Unlearning
Oct 10, 2026Yang Xiao, Kaiyuan Deng, Gen Li +4Diffusion Model Unlearning Machine Unlearning
University of Tulsa · The University of Arizona · Clemson University +1
Enabling Preference-driven Unlearning in Few-step Distilled Text-to-Image Diffusion Models
Oct 7, 2026Gaurav Patel, Jun Fang, Greg Ver Steeg +2Preference Optimization for Diffusion Models Diffusion Model Unlearning
Purdue University · Amazon AGI
Compositional Concept Erasure in Text-to-Image Diffusion Models via Hierarchically Grounded Semantic Surgery
Oct 5, 2026Chen Dai, Ganyu Zou, Nathan Self +5Diffusion Model Unlearning Concept Erasure
Department of Computer Science Virginia Tech Alexandria, Virginia, USA · Verisign, Inc. Reston, Virginia, USA
Your Unlearning Gives You Away: Identifying Erased Concepts in Diffusion Models
Oct 4, 2026Kaiyuan Deng, Yuchen Li, Yang Xiao +3Diffusion Model Unlearning Concept Erasure
The University of Arizona · Peking University · The University of Tulsa +1
No Concept Escapes the Audit: Auditing-Aware Unlearning for Verifiable Concept Erasure in Diffusion Models
Oct 4, 2026Kaiyuan Deng, Yuchen Li, Gen Li +5Diffusion Model Unlearning T2I Diffusion Models
The University of Arizona · Peking University · Clemson University +2
VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models
Oct 4, 2026Qianlong Xiang, Miao Zhang, Kun Wang +3Diffusion Model Unlearning T2I Diffusion Models
Harbin Institute of Technology (Shenzhen) · City University of Hong Kong · Shenzhen Loop Area Institute +2
FADE: Frame-Aware Diffusion-Transformer-based Multi-Concept Erasure for Video Unlearning
Oct 2, 2026Yuchen Li, Kaiyuan Deng, Chaoran Feng +2Diffusion Model Unlearning Machine Unlearning
Peking University · University of Arizona
Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference
Oct 1, 2026Yongliang Wu, Haori Lu, Jinqi Luo +3Diffusion Model Unlearning Image Editing
University of Illinois Urbana-Champaign · University of Pennsylvania · National University of Singapore
RASteer: Retain-Aware Activation Steering for Concept Erasure in Diffusion Models
Oct 1, 2026Yongliang Wu, Haori Lu, Yulun Wu +3Diffusion Model Unlearning Activation Steering
University of Illinois Urbana-Champaign · University of Pennsylvania · National University of Singapore
FOMO: Forget the Concept, Don't Miss Out on the Scene in Selective Video Unlearning
Sep 30, 2026Łukasz Rudnik, Agnieszka Polowczyk, Alicja Polowczyk +1Video Diffusion Models Diffusion Model Unlearning
Jagiellonian University · IDEAS Research Institute
Weeding Out Bad Seeds: Initial-Noise-Robust Unlearning for Text-to-Image Diffusion Models
Sep 29, 2026Arian Komaei Koma, Seyed Amir Kasaei, Aida Aryafar +6Diffusion Model Unlearning Machine Unlearning
Sharif University of Technology · Hong Kong University of Science and Technology
Motion Concept Unlearning in Video Diffusion Models
Sep 29, 2026Ping Liu, Chi ZhangVideo Diffusion Models Diffusion Transformer
Computer Science and Engineering University of Nevada, Reno Reno, NV, USA · Department of Mathematics National University of Singapore Singapore
Data Unlearning via Inverse Distillation
Sep 28, 2026Aleksei Leonov, Nikita Kornilov, Zhenhe Zhang +3Machine Unlearning Diffusion Model Unlearning
AI Foundation lab, Moscow, Russia · MIRAI, Moscow, Russia · Applied AI Institute, Moscow, Russia +2
eval-unlearn: Benchmarking unlearning in Text-to-Image Diffusion Models
Sep 28, 2026Mansi, Nikhil Raghavan, Zixia Huang +4Diffusion Model Unlearning T2I Generation Evaluation
Imperial College London
You Can't Have It Both Ways: Concept Entanglement Limits Diffusion Model Unlearning
Sep 28, 2026Yian Wang, Ali Ebrahimpour-Boroojeny, Hari Sundaram +1Diffusion Model Unlearning Machine Unlearning
University of Illinois Urbana-Champaign Urbana, IL, USA
Concept Score Relearning: A Unified Cross-Architecture Attack on Concept Erasure
Sep 27, 2026Hong Xi Tae, Jiaming Zhang, Wenwen He +2Diffusion Model Unlearning Concept Erasure
College of Computing and Data Science, Nanyang Technological University, Singapore
Continual Data Unlearning in Diffusion Models via Transition-based Regularization
Sep 26, 2026Sunbeom Jeong, Sehwan Kim, Sangwoo Hong +1Diffusion Model Unlearning Machine Unlearning
Seoul National University · Konkuk University
HyperErase: Scale-Calibrated Hypernetwork for Multi-Concept Erasure in Text-to-Image Models
Sep 25, 2026Yi Sun, Xinhao Zhong, Zhiqi Zhang +3Safe T2I Generation Diffusion Model Unlearning
Harbin Institute of Technology, Shenzhen · Jilin University · Tsinghua Shenzhen International Graduate School, Tsinghua University +1
Mitigating Sequential Reappearance in Diffusion Data-Point Unlearning
Sep 21, 2026Donghyun Kim, Taehyuk Lee, Jinyeong Kim +4Memorization in Generative Models Diffusion Model Unlearning
Department of Computer Science and Engineering, Konkuk University Trustworthy Machine Learning Lab Seoul, Republic of Korea
GRACE: Adaptive Concept Erasure with Geometry-Guided Retention in Diffusion Models
Sep 11, 2026Qinghui Gong, Yihuai Liang, Yuanlun Xie +3Diffusion Model Fine-Tuning Diffusion Model Unlearning
School of Information Science and Technology, Southwest Jiaotong University, Chengdu 611756, China · School of Electronic Information and Electrical Engineering, Chengdu University, Chengdu, China · Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian, China +1
Certifying Concept Unlearning in Text-to-Image Diffusion Models
Sep 10, 2026Mansi, Luca Marzari, Francesco LeofanteDiffusion Model Unlearning Concept Erasure
Imperial College London · TU Wien
EraseSAE: Surgical Concept Erasure in Text-to-Video Diffusion Models via Sparse Autoencoders
Sep 3, 2026Xinghao Wang, Dong Li, Wei Yu +5Video Diffusion Models Sparse Autoencoders
University of Science and Technology of China, Hefei, China · HiDream.ai Inc. · Anhui Province Key Laboratory of Digital Security, China
Gaussian Core LoRA: Distribution-Aware Dynamic Adaptation for Broad Concept Erasure
Sep 1, 2026Qinghui Gong, Xunlei Chen, Yu-Xuan Zhang +2Diffusion Model Fine-Tuning Low-Rank Adaptation
Southwest Jiaotong University · University of Electronic Science and Technology of China
TINA+: Probing Residual Visual Knowledge in Unlearned Diffusion Models via Diffusion-Consistent Text-Free Inversion
Aug 18, 2026Qianlong Xiang, Miao Zhang, Kun Wang +3T2I Diffusion Models Diffusion Model Unlearning
School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen 518055, China · City University of Hong Kong, Hong Kong SAR, China · Shenzhen Loop Area Institute, Shenzhen 518045, China +4
MapRoute++: Surrogate-Guided Semantic Routing for Visual Concept Unlearning
Aug 13, 2026Ashok Urlana, L. D. M. S. Sai Teja, Vivek Hruday Kavuri +1Diffusion Model Unlearning Machine Unlearning
1IIIT Hyderabad · 2TCS Research, India · 3NIT Silchar
Semantic Steering for Controllable Generation: Tuning-Free Concept Erasure in Multimodal Diffusion Transformers
Aug 13, 2026Qiao Li, Xiaomeng Fu, Yuanshu Zhao +3Diffusion Model Unlearning Activation Steering
Institute of Information Engineering, Chinese Academy of Sciences Beijing, China · School of Cyber Security, University of Chinese Academy of Sciences Beijing, China

Oct 23, 2026J/K move · Enter open · S save

Concept Acupuncturist: An Efficient and Precise Concept Erasure Method for Diffusion Models via Sparse Autoencoders

Yiran Xiahou, Siliang Qin, Kai Chen, Shengzhi Zhang+1

Institute of Information Engineering, Chinese Academy of Sciences, China · Department of Computer Science, Metropolitan College, Boston University, USA · Institute of Data Science, NationalUniversity of SingaporeSingapore, Singapore

Diffusion Model Unlearning Concept Erasure Diffusion Model Fine-Tuning

Open paper AskPDF arXiv

Diffusion-based image generation models produce high-quality images but may generate harmful or policy-violating content, motivating machine unlearning (MU) to selectively remove target concepts while preserving general capabilities. However, existing MU approaches often struggle with the distributed and entangled nature of concept representations, leading to inefficient optimization and imprecise edits when applied through large-scale fine-tuning over full diffusion trajectories. To bridge this gap, we propose Concept Acupuncturist (ConceptACP), a novel MU framework that integrates interpretability-guided analysis with inference-free fine-tuning for efficient and precise concept removal. By establishing mappings between semantic concepts and cross-attention blocks in the U-Net denoiser and localizing the most concept-relevant block for targeted optimization, ConceptACP avoids full diffusion trajectory optimization, thereby improving fine-tuning efficiency. To mitigate concept entanglement, we propose a feature-level correction mechanism based on Sparse Autoencoders (SAEs), where SAE-derived sparse, semantically meaningful feature activations are manipulated to construct targeted correction signals, which are then projected back to the output space of the corresponding block and used as supervision for precise concept removal. Consequently, ConceptACP eliminates the need for external inference-time modules while achieving efficient and fine-grained unlearning. Instantiated on SDXL-Turbo, ConceptACP improves the forgetting success rate by 36.34 percentage points in the standard one-step generation setting on average over the best-performing baseline, while preserving image quality and prompt fidelity, and generalizes effectively across other Stable Diffusion backbones. Our code is available at https://github.com/Eran-Xia/ConceptACP.