paper-with-me

홈 › Papers

To Erase, or Not to Erase: Robust Training-Free Concept Erasure with Preservation aware Adaptive Ranked Subspace Expansion

2026-07-26 · Shaswati Saha, Rajasekhar Anguluri, Manas Gaur arxiv

Concept erasure techniques (CETs) edit text-to-image diffusion models to erase undesired targets such as NSFW content or copyrighted styles, while preserving model utility on benign concepts. Current CETs face a trade-off between erasure robustness and utility: stronger edits erase the target more reliably but degrade utility on non-target concepts, and vice versa. This stems from how existing methods define what to erase and what to preserve. Many CETs rely on static concept banks specified manually, generated by LLMs, or selected by CLIP image-text similarity. Such banks do not model how prompts steer the model during denoising, leaving it vulnerable to triggers that reintroduce the target while suppressing nearby benign concepts. We present Preservation-aware Adaptive Ranked Subspace Expansion (PARSE), a training-free framework for robust concept erasure in latent diffusion models. Given a target, PARSE queries the diffusion model with classifier-free guidance to dynamically discover target-inducing erase concepts and nearby retain concepts in the model vocabulary. It then edits the cross-attention value space with a preservation-aware projection that removes target directions while leaving retain directions intact. For triggers beyond this vocabulary-indexed space, PARSE iteratively searches for re-emergence triggers by textual inversion and adaptively expands the erased subspace only when a new trigger direction does not conflict with retain semantics. We also introduce the Balanced Erasure Utility Score (BEUS), which combines robustness (ASR under multiple attacks) and utility preservation (FID) via bounded monotone transforms and harmonic mean aggregation. Experiments on NSFW, artistic style, and object erasure, with a large-scale robustness-utility analysis over many CET baselines, show that PARSE erases multiple concepts robustly without sacrificing post-edit utility.

📄 PDF Abstract BibTeX arXiv:2607.23492

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

VideoEraser: Concept Erasure in Text-to-Video Diffusion Models

2025-08-21 · Naen Xu, Jinghuai Zhang, Changjiang Li, Zhi Chen 외 arxiv

The rapid growth of text-to-video (T2V) diffusion models has raised concerns about privacy, copyright, and safety due to their potential misuse in generating harmful or misleading content. These models are often trained …

Z-Erase: Enabling Concept Erasure in Single-Stream Diffusion Transformers

2026-03-26 · Nanxiang Jiang, Zhaoxin Fan, Baisen Wang, Daiheng Gao 외 arxiv

Concept erasure serves as a vital safety mechanism for removing unwanted concepts from text-to-image (T2I) models. While extensively studied in U-Net and dual-stream architectures (e.g., Flux), this task remains under-ex…

Image Generation

AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models

2026-02-06 · Fengpeng Li, Kemou Li, Qizhou Wang, Bo Han 외 arxiv

Concept erasure helps stop diffusion models (DMs) from generating harmful content; but current methods face robustness retention trade off. Robustness means the model fine-tuned by concept erasure methods resists reactiv…

ActErase: A Training-Free Paradigm for Precise Concept Erasure via Activation Redirection

2026-01-01 · Yi Sun, Xinhao Zhong, Hongyan Li, Yimin Zhou 외 arxiv

Recent advances in text-to-image diffusion models have demonstrated remarkable generation capabilities, yet they raise significant concerns regarding safety, copyright, and ethical implications. Existing concept erasure …

Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference

2026-10-01 · Yongliang Wu, Haori Lu, Jinqi Luo, Wei Cao 외 arxiv

Concept erasure removes copyright-protected, privacy-sensitive, or otherwise undesirable concepts from pretrained text-to-image diffusion models to support content governance and compliance. As erasure requests arrive ov…