Compositional Concept Erasure in Text-to-Image Diffusion Models via Hierarchically Grounded Semantic Surgery

arXiv:2610.07337 · cs.CV · Submitted 2026-10-05 · Read on arXiv

cs.CV

Submitted: 2026-10-05

Updated: 2026-10-05

Comments: Accepted at BMVC 2026

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Diffusion Model — 1× in this paper · explained in 8 episodes
• Diffusion Models — 1× in this paper · explained in 6 episodes

Sources

• Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models
• SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
• Reliable and Efficient Concept Erasure of Text-to-Image Diffusion Models
• Classifier-Free Diffusion Guidance
• Denoising Diffusion Probabilistic Models
• Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Erasers
• TraSCE: Trajectory Steering for Concept Erasure
• Localized Concept Erasure for Text-to-Image Diffusion Models Using Training-Free Gated Low-Rank Adaptation
• Get What You Want, Not What You Don't: Image Content Suppression for Text-to-Image Diffusion Models
• One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications
• Diffusion Adaptive Text Embedding for Text-to-Image Diffusion Models
• Side Effects of Erasing Concepts from Diffusion Models
• Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?
• Scissorhands: Scrub Data Influence via Connection Sensitivity in Networks
• Erasing Undesirable Influence in Diffusion Models
• Semantic Surgery: Zero-Shot Concept Erasure in Diffusion Models
• MMA-Diffusion: MultiModal Attack on Diffusion Models
• SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generation
• Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models