DSS: Dynamic Semantic Steering for Robust Concept Erasure in Diffusion Models

arXiv:2604.16483 · cs.CV, cs.AI · Submitted 2026-04-13 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-04-13

Updated: 2026-09-12

Comments: 29 pages,29 figures

Code: https://github.com/CompVis/stable-diffusion

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

Terminology

• Diffusion Model — 51× in this paper · explained in 8 episodes
• Diffusion Models — 50× in this paper · explained in 6 episodes
• Embeddings — 20× in this paper · explained in 2 episodes
• Core — 16× in this paper · explained in 2 episodes
• Decoupling — 7× in this paper · explained in 2 episodes
• Loss Function — 7× in this paper · explained in 2 episodes
• Semantic Drift — 6× in this paper · explained in 3 episodes
• Alignment — 5× in this paper · explained in 1 episode
• Steering — 5× in this paper · explained in 2 episodes
• Verification — 5× in this paper · explained in 2 episodes
• Dynamic Semantic Steering — 4× in this paper
• Robustness — 4× in this paper · explained in 5 episodes
• Sensitive Semantic Guided Correction — 4× in this paper
• AUC — 3× in this paper · explained in 2 episodes
• Ablation — 3× in this paper · explained in 1 episode
• Adaptation — 3× in this paper · explained in 1 episode
• Fidelity — 3× in this paper · explained in 2 episodes
• Generalization — 3× in this paper · explained in 13 episodes
• Adversarial Attacks — 2× in this paper · explained in 1 episode
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Pruning — 2× in this paper · explained in 4 episodes
• Training-Free Framework — 2× in this paper · explained in 2 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Accountability — 1× in this paper · explained in 2 episodes
• Automatic Sensitive Semantic Boundary Modeling — 1× in this paper
• Benchmark — 1× in this paper · explained in 2 episodes
• Clustering — 1× in this paper · explained in 2 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Gaussian Kernel Density Estimation — 1× in this paper
• Interpretability — 1× in this paper · explained in 15 episodes
• Low-Rank Adaptation (LoRA) — 1× in this paper · explained in 9 episodes
• Machine Unlearning — 1× in this paper · explained in 2 episodes
• Nash Equilibrium — 1× in this paper · explained in 2 episodes
• Rectified Flow — 1× in this paper · explained in 3 episodes
• Semantic Alignment — 1× in this paper · explained in 3 episodes
• Sparsity — 1× in this paper · explained in 4 episodes

Sources

• GPT-4 Technical Report
• Multimodal datasets: misogyny, pornography, and malignant stereotypes
• CURE: Concept Unlearning via Orthogonal Representation Editing in Diffusion Models
• SafeGen: Mitigating Sexually Explicit Content Generation in Text-to-Image Models
• Dark Miner: Defend against undesirable generation for text-to-image diffusion models
• LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs
• PromptGuard: Soft Prompt-Guided Unsafe Content Moderation for Text-to-Image 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