Variance as a Catalyst: Efficient and Transferable Semantic Erasure Adversarial Attack for Customized Diffusion Models

Jiachen Yang, Yusong Wang, Yanmei Fang, Yunshu Dai, Fangjun Huang
2/14/2026

No official code found. Best implementation: youyuanyi/variance-as-Catalyst (6 stars, last commit 2025-07-08).

Semantic Scholar

Code Implementations(1)

youyuanyi/variance-as-CatalystCommunity70%

Variance as a Catalyst: Efficient and Transferable Semantic Erasure Adversarial Attack for Customized Diffusion Models

62Shell, PythonJan 29, 20251 years ago

Similar Papers

InstantSplamp: Fast and Generalizable Stenography Framework for Generative Gaussian Splatting

95% match1 repo252026

Position: Future Research and Challenges Remain Towards AI for Software Engineering

95% match2026

Propensity Matters: Measuring and Enhancing Balancing for Recommendation

95% match2026

Linearly Controllable GAN: Unsupervised Feature Categorization and Decomposition for Image Generation and Manipulation

95% match1 repo32026

SiReRAG: Indexing Similar and Related Information for Multihop Reasoning

94% match2026

A Simple Approach for Visual Room Rearrangement: 3D Mapping and Semantic Search

94% match2026

Cite this paper

@article{yang2026variance,
  title  = {Variance as a Catalyst: Efficient and Transferable Semantic Erasure Adversarial Attack for Customized Diffusion Models},
  author = {Jiachen Yang and Yusong Wang and Yanmei Fang and Yunshu Dai and Fangjun Huang},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:23f50dd5217bd7a743a96c3ae7503ccebdd1a4c7},
  journal = {ICML 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/23f50dd5217bd7a743a96c3ae7503ccebdd1a4c7.svg)](https://codeofpaper.com/paper/23f50dd5217bd7a743a96c3ae7503ccebdd1a4c7)

Discussion