Cover Image

Safety Alignment Backfires: Preventing the Re-emergence of Suppressed Concepts in Fine-tuned Text-to-Image Diffusion Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Kim, Sanghyun, Choi, Moonseok, Shin, Jinwoo, Lee, Juho
Format:	Preprint
Published:	2024
Subjects:	Artificial Intelligence Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Safeguard Text-to-Image Diffusion Models with Human Feedback Inversion
by: Kim, Sanghyun, et al.
Published: (2024)

StarFT: Robust Fine-tuning of Zero-shot Models via Spuriosity Alignment
by: Kim, Younghyun, et al.
Published: (2025)

DiffExp: Efficient Exploration in Reward Fine-tuning for Text-to-Image Diffusion Models
by: Chae, Daewon, et al.
Published: (2025)

A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models
by: Moon, Taehong, et al.
Published: (2024)

Adversarial Robustification via Text-to-Image Diffusion Models
by: Choi, Daewon, et al.
Published: (2024)

Direct Consistency Optimization for Robust Customization of Text-to-Image Diffusion Models
by: Lee, Kyungmin, et al.
Published: (2024)

Improving Diffusion Models for Authentic Virtual Try-on in the Wild
by: Choi, Yisol, et al.
Published: (2024)

TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias
by: Jo, Sanghyun, et al.
Published: (2024)

DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing
by: Choi, June Suk, et al.
Published: (2024)

Confidence-aware Denoised Fine-tuning of Off-the-shelf Models for Certified Robustness
by: Jang, Suhyeok, et al.
Published: (2024)

Personalized Safety Alignment for Text-to-Image Diffusion Models
by: Lei, Yu, et al.
Published: (2025)

Improving Motion in Image-to-Video Models via Adaptive Low-Pass Guidance
by: Choi, June Suk, et al.
Published: (2025)

TRACE: Your Diffusion Model is Secretly an Instance Edge Detector
by: Jo, Sanghyun, et al.
Published: (2025)

Restoration-Aligned Generative Flow Models for Blind Motion Deblurring
by: Kim, Insoo, et al.
Published: (2026)

PCBEAR: Pose Concept Bottleneck for Explainable Action Recognition
by: Lee, Jongseo, et al.
Published: (2025)

Efficient Fine-Tuning and Concept Suppression for Pruned Diffusion Models
by: Shirkavand, Reza, et al.
Published: (2024)

Visual Concept-driven Image Generation with Text-to-Image Diffusion Model
by: Rahman, Tanzila, et al.
Published: (2024)

Growth Inhibitors for Suppressing Inappropriate Image Concepts in Diffusion Models
by: Chen, Die, et al.
Published: (2024)

DreamFlow: High-Quality Text-to-3D Generation by Approximating Probability Flow
by: Lee, Kyungmin, et al.
Published: (2024)

Dual-Stream Diffusion for World-Model Augmented Vision-Language-Action Model
by: Won, John, et al.
Published: (2025)

Object-Driven One-Shot Fine-tuning of Text-to-Image Diffusion with Prototypical Embedding
by: Lu, Jianxiang, et al.
Published: (2024)

Blending Concepts with Text-to-Image Diffusion Models
by: Olearo, Lorenzo, et al.
Published: (2025)

Prompt-Based Safety Guidance Is Ineffective for Unlearned Text-to-Image Diffusion Models
by: Shin, Jiwoo, et al.
Published: (2025)

DreamBlend: Advancing Personalized Fine-tuning of Text-to-Image Diffusion Models
by: Ram, Shwetha, et al.
Published: (2024)

Disentangled Concepts Speak Louder Than Words: Explainable Video Action Recognition
by: Lee, Jongseo, et al.
Published: (2025)

Curriculum Fine-tuning of Vision Foundation Model for Medical Image Classification Under Label Noise
by: Yu, Yeonguk, et al.
Published: (2024)

Where Concept Erasure Should Occur: Concept-Layer Alignment in Text-to-Video Diffusion Models
by: Xie, Yiwei, et al.
Published: (2026)

Localized Concept Erasure in Text-to-Image Diffusion Models via High-Level Representation Misdirection
by: Lee, Uichan, et al.
Published: (2026)

Controllable Human Image Generation with Personalized Multi-Garments
by: Choi, Yisol, et al.
Published: (2024)

Lipsum-FT: Robust Fine-Tuning of Zero-Shot Models Using Random Text Guidance
by: Nam, Giung, et al.
Published: (2024)

SafeCtrl: Region-Based Safety Control for Text-to-Image Diffusion via Detect-Then-Suppress
by: Zhang, Lingyun, et al.
Published: (2025)

SafeCtrl: Region-Aware Safety Control for Text-to-Image Diffusion via Detect-Then-Suppress
by: Zhang, Lingyun, et al.
Published: (2026)

FT-Shield: A Watermark Against Unauthorized Fine-tuning in Text-to-Image Diffusion Models
by: Cui, Yingqian, et al.
Published: (2023)

Semantic Guidance Tuning for Text-To-Image Diffusion Models
by: Kang, Hyun, et al.
Published: (2023)

Large-Scale Text-to-Image Model with Inpainting is a Zero-Shot Subject-Driven Image Generator
by: Shin, Chaehun, et al.
Published: (2024)

A Comprehensive Survey on Concept Erasure in Text-to-Image Diffusion Models
by: Kim, Changhoon, et al.
Published: (2025)

Model Already Knows the Best Noise: Bayesian Active Noise Selection via Attention in Video Diffusion Model
by: Kim, Kwanyoung, et al.
Published: (2025)

Translation of Text Embedding via Delta Vector to Suppress Strongly Entangled Content in Text-to-Image Diffusion Models
by: Koh, Eunseo, et al.
Published: (2025)

Latent Space Super-Resolution for Higher-Resolution Image Generation with Diffusion Models
by: Jeong, Jinho, et al.
Published: (2025)

Subject-Diffusion:Open Domain Personalized Text-to-Image Generation without Test-time Fine-tuning
by: Ma, Jian, et al.
Published: (2023)