paper-with-me

홈 › Papers

EraseAnything: Enabling Concept Erasure in Rectified Flow Transformers

2024-12-29 · Daiheng Gao, Shilin Lu, Shaw Walters, Wenbo Zhou, Jiaming Chu, Jie Zhang, Bang Zhang, Mengxi Jia, Jian Zhao, Zhaoxin Fan, Weiming Zhang

Removing unwanted concepts from large-scale text-to-image (T2I) diffusion models while maintaining their overall generative quality remains an open challenge. This difficulty is especially pronounced in emerging paradigms, such as Stable Diffusion (SD) v3 and Flux, which incorporate flow matching and transformer-based architectures. These advancements limit the transferability of existing concept-erasure techniques that were originally designed for the previous T2I paradigm (e.g., SD v1.4). In this work, we introduce EraseAnything, the first method specifically developed to address concept erasure within the latest flow-based T2I framework. We formulate concept erasure as a bi-level optimization problem, employing LoRA-based parameter tuning and an attention map regularizer to selectively suppress undesirable activations. Furthermore, we propose a self-contrastive learning strategy to ensure that removing unwanted concepts does not inadvertently harm performance on unrelated ones. Experimental results demonstrate that EraseAnything successfully fills the research gap left by earlier methods in this new T2I paradigm, achieving state-of-the-art performance across a wide range of concept erasure tasks.

📄 PDF Abstract BibTeX arXiv:2412.20413

Code (1)

tomguluson92/eraseanything 공식 구현 jax

Tasks

Contrastive Learning

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음
Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

EraseAnything++: Enabling Concept Erasure in Rectified Flow Transformers Leveraging Multi-Object Optimization

2026-03-01 · Zhaoxin Fan, Nanxiang Jiang, Daiheng Gao, Shiji Zhou 외 arxiv

Removing undesired concepts from large-scale text-to-image (T2I) and text-to-video (T2V) diffusion models while preserving overall generative quality remains a major challenge, particularly as modern models such as Stabl…

Video Generation

TRACE: Trajectory-Constrained Concept Erasure in Diffusion Models

2025-05-29 · Finn Carter

Text-to-image diffusion models have shown unprecedented generative capability, but their ability to produce undesirable concepts (e.g.~pornographic content, sensitive identities, copyrighted styles) poses serious concern…

DenoisingFairness

GEM: Geometric Erasure by Contrastive Velocity Matching in Rectified Flows

2026-05-29 · Jonas Henry Grebe, Tobias Braun, Anna Rohrbach, Marcus Rohrbach arxiv

While the rapid adoption of multimodal generative models offers immense potential, it has also increased the risks of harmful content synthesis, deepfakes, and copyright infringements. To address these challenges, concep…

Erased, But Not Forgotten: Erased Rectified Flow Transformers Still Remain Unsafe Under Concept Attack

2025-10-01 · Nanxiang Jiang, Zhaoxin Fan, Enhan Kang, Daiheng Gao 외 arxiv

Recent advances in text-to-image (T2I) diffusion models have enabled impressive generative capabilities, but they also raise significant safety concerns due to the potential to produce harmful or undesirable content. Whi…

Robust Concept Erasure in Diffusion Models: A Theoretical Perspective on Security and Robustness

2025-09-15 · Zixuan Fu, Yan Ren, Finn Carter, Chenyue Wen 외 arxiv

Diffusion models have achieved unprecedented success in image generation but pose increasing risks in terms of privacy, fairness, and security. A growing demand exists to \emph{erase} sensitive or harmful concepts (e.g.,…

Image Generation