analiza aiqbench
This research addresses the critical challenge of unauthorized style mimicry in text-to-image diffusion models by shifting from aggressive concept erasure to a "style purification" approach. The proposed CAPE framework uses contrastive eigenbases to suppress specific artistic attributes like brushstrokes and textures during inference while preserving semantic content, offering a training-free solution for copyright protection.
From Concept Erasure to Style Purification: Contrastive Eigenbases for Artist Style Protection
arXiv:2602.08059v2 Announce Type: replace-cross Abstract: Text-to-image diffusion models can reproduce specific artists visual styles at extremely low cost, raising copyright and deployment safety concerns about unauthorized style mimicry. Existing model-side protection methods generally follow ordinary concept erasure, emphasizing aggressive deletion or redirection of target styles. However, our causal intervention analysis shows that the central issue is not insufficient erasure strength, but a mismatch between artist styles and this paradigm: unlike ordinary object concepts, artist styles do not form compact, localized editable semantic units. Consequently, sparse editing and fixed retain lists struggle to suppress target styles while preserving generation utility. We therefore reformulate artist style protection as style purification, suppressing target style expression during inference while preserving the requested content and visual structure. We propose CAPE (Contrastive Artist Style Purification with Eigenbases), a training-free framework against artist style mimicry. CAPE constructs contrastive triplets around the target request and formulates style direction estimation as a generalized eigenvalue problem, capturing style-related directions that remain stable across content variations and are less affected by shared content. During inference, CAPE employs the Adaptive Suppression Controller to assign suppression strengths to different tokens based on Q, K, and V responses, and performs target style suppression on the K and V paths of self-attention. Experimental results show that CAPE effectively weakens target artist characteristics, including brushstrokes, textures, and local color processing, while better preserving major semantic entities, scene composition, and visual structures.
Abstract:Text-to-image diffusion models can reproduce specific artists visual styles at extremely low cost, raising copyright and deployment safety concerns about unauthorized style mimicry. Existing model-side protection methods generally follow ordinary concept erasure, emphasizing aggressive deletion or redirection of target styles. However, our causal intervention analysis shows that the central issue is not insufficient erasure strength, but a mismatch between artist styles and this paradigm: unlike ordinary object concepts, artist styles do not form compact, localized editable semantic units. Consequently, sparse editing and fixed retain lists struggle to suppress target styles while preserving generation utility. We therefore reformulate artist style protection as style purification, suppressing target style expression during inference while preserving the requested content and visual structure. We propose CAPE (Contrastive Artist Style Purification with Eigenbases), a training-free framework against artist style mimicry. CAPE constructs contrastive triplets around the target request and formulates style direction estimation as a generalized eigenvalue problem, capturing style-related directions that remain stable across content variations and are less affected by shared content. During inference, CAPE employs the Adaptive Suppression Controller to assign suppression strengths to different tokens based on Q, K, and V responses, and performs target style suppression on the K and V paths of self-attention. Experimental results show that CAPE effectively weakens target artist characteristics, including brushstrokes, textures, and local color processing, while better preserving major semantic entities, scene composition, and visual structures.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2602.08059 [cs.CV]
(or arXiv:2602.08059v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2602.08059 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 328 cuvinte
Subiecte conexe
To Erase, or Not to Erase: Robust Training-Free Concept Erasure with Preservation aware Adaptive Ranked Subspace Expansion
arXiv · 28 iul. 2026
No Concept Escapes the Audit: Auditing-Aware Unlearning for Verifiable Concept Erasure in Diffusion Models
arXiv · 7 oct. 2026
Concept Unlearning via Cross-Attention Activation Projection for Diffusion Models
arXiv · 1 oct. 2026
EraseSAE: Surgical Concept Erasure in Text-to-Video Diffusion Models via Sparse Autoencoders
arXiv · 5 sept. 2026