- concept erasure
- concept debiasing
- Backdoor attack
- Backdoor defense
- Adversarial attack
- Inference attack
- Copyright
- A tool for defense
Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models
Ablating Concepts in Text-to-Image Diffusion Models
Rickrolling the Artist: Injecting Backdoors into Text Encoders for Text-to-Image Synthesis
Erasing Concepts from Diffusion Models
Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models
Inst-Inpaint: Instructing to Remove Objects with Diffusion Models
Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Models
Towards Safe Self-Distillation of Internet-Scale Text-to-Image Diffusion Models
Unified Concept Editing in Diffusion Models
Implicit Concept Removal of Diffusion Models
To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now
Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Erasers
All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Models
One-dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications
EraseDiff: Erasing Data Influence in Diffusion Models
Separable Multi-Concept Erasure from Diffusion Models
SalUn Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?
Get What You Want, Not What You Don't: Image Content Suppression for Text-to-Image Diffusion Models
Localizing and Editing Knowledge in Text-to-Image Generative Models
UnlearnCanvas: A Stylized Image Dataset to Benchmark Machine Unlearning for Diffusion Models
Universal Prompt Optimizer for Safe Text-to-Image Generation
Circumventing Concept Erasure Methods For Text-to-Image Generative Models
MACE: Mass Concept Erasure in Diffusion Models
Position: Towards Implicit Prompt For Text-To-Image Models
Editing Massive Concepts in Text-to-Image Diffusion Models
Removing Undesirable Concepts in Text-to-Image Generative Models with Learnable Prompts
On Mechanistic Knowledge Localization in Text-to-Image Generative Models
Concept Arithmetics for Circumventing Concept Inhibition in Diffusion Models
ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning
R.A.C.E. : Robust Adversarial Concept Erasure for Secure Text-to-Image Diffusion Model
Pruning for Robust Concept Erasing in Diffusion Models
Erasing Concepts from Text-to-Image Diffusion Models with Few-shot Unlearning
Text Guided Image Editing with Automatic Concept Locating and Forgetting
Robust Concept Erasure Using Task Vectors
Unlearning Concepts in Diffusion Model via Concept Domain Correction and Concept Preserving Gradient
Data Attribution for Text-to-Image Models by Unlearning Synthesized Images
CLIPAway: Harmonizing Focused Embeddings for Removing Objects via Diffusion Models
Instructing Text-to-Image Generation Models on Fairness
Debiasing Pretrained Generative Models by Uniformly Sampling Semantic Attributes
De-stereotyping Text-to-image Models through Prompt Tuning
Stable Bias: Evaluating Societal Representations in Diffusion Models
Debiasing Vision-Language Models via Biased Prompts
Discovering and Mitigating Biases in CLIP-based Image Editing
Unified Concept Editing in Diffusion Models
Finetuning Text-to-Image Diffusion Models for Fairness
Fair Text-to-Image Diffusion via Fair Mapping
Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generation
Debiasing Text-to-Image Diffusion Models
Balancing Act: Distribution-Guided Debiasing in Diffusion Models
Training Unbiased Diffusion Models From Biased Dataset
Severity Controlled Text-to-Image Generative Model Bias Manipulation
OpenBias: Open-set Bias Detection in Text-to-Image Generative Models
SCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation
FairRAG: Fair Human Generation via Fair Retrieval Augmentation
EquiPrompt: Debiasing Diffusion Models via Iterative Bootstrapping in Chain of Thoughts
How to Backdoor Diffusion Models?
TrojDiff: Trojan Attacks on Diffusion Models with Diverse Targets
VillanDiffusion: A Unified Backdoor Attack Framework for Diffusion Models
Rickrolling the Artist: Injecting Backdoors into Text Encoders for Text-to-Image Synthesis
Text-to-Image Diffusion Models can be Easily Backdoored through Multimodal Data Poisoning
BAGM: A Backdoor Attack for Manipulating Text-to-Image Generative Models
From Trojan Horses to Castle Walls: Unveiling Bilateral Backdoor Effects in Diffusion Models
The Stronger the Diffusion Model, the Easier the Backdoor: Data Poisoning to Induce Copyright Breaches Without Adjusting Finetuning Pipeline
Personalization as a Shortcut for Few-Shot Backdoor Attack against Text-to-Image Diffusion Models
Nightshade: Prompt-Specific Poisoning Attacks on Text-to-Image Generative Models
Invisible Backdoor Attacks on Diffusion Models
A Recipe for Watermarking Diffusion Models
DisDet: Exploring Detectability of Backdoor Attack on Diffusion Models
How to remove backdoors in diffusion models?
Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shift
Ufid: A unified framework for input-level backdoor detection on diffusion models
FreezeAsGuard: Mitigating Illegal Adaptation of Diffusion Models via Selective Tensor Freezing
TERD: A Unified Framework for Safeguarding Diffusion Models Against Backdoors
A Recipe for Watermarking Diffusion Models
The Stable Signature: Rooting Watermarks in Latent Diffusion Models
Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and Robust
DIAGNOSIS: Detecting Unauthorized Data Usages in Text-to-image Diffusion Models
“RingID: Rethinking Tree-Ring Watermarking for Enhanced Multi-Key Identification
Gaussian Shading: Provable Performance-Lossless Image Watermarking for Diffusion Models
“Lazy” Layers to Make Fine-Tuned Diffusion Models More Traceable
A Training-Free Plug-and-Play Watermark Framework for Stable Diffusion
ModelLock: Locking Your Model With a Spell
Disguised Copyright Infringement of Latent Diffusion Models
WMAdapter: Adding WaterMark Control to Latent Diffusion Models
AquaLoRA: Toward White-box Protection for Customized Stable Diffusion Models via Watermark LoRA
Steganalysis on Digital Watermarking: Is Your Defense Truly Impervious?
Black-box Backdoor Defense via Zero-shot Image Purification
DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models
I will update periodically...