aiqbench analysis
This research addresses a critical safety gap in visual autoregressive models by introducing SACE, the first scale-aware framework designed to remove specific concepts without triggering semantic collapse or visual artifacts common in diffusion-based erasure methods. By rigorously validating its theoretical axioms and demonstrating minimal training overhead across diverse domains, the study offers a robust solution for aligning high-fidelity text-to-image synthesis with safety constraints.
SACE: Concept Erasure at the Semantic Singularity in Visual Autoregressive Models
arXiv:2606.15819v1 Announce Type: cross Abstract: The rapid progress of visual autoregressive (VAR) models has unlocked a transformative frontier for high-fidelity text-to-image synthesis, while heightening concerns over the safety alignment of generated content. Naive application of existing erasure techniques to VAR models causes catastrophic semantic collapse and visual artifacts, since they are predominantly designed for the homogeneous denoising steps of diffusion models. To address this foundational challenge, we first propose the Semantic Singularity Axiom, which posits that any target semantic concept embedded within a prompt is definitively locked at Scale-0. Then rigorously validate this axiom through our proposed Incremental Semantic Saliency Analysis (ISSA),which also enable the community to transparently inspect the coarse-to-fine semantic injection process. Guided by this insight, we introduce the first scale-aware concept erasure framework (SACE) for VAR models. By strictly confining interventions to the first scale, our approach couples an Entropy-Regularized Erasure Objective to prevent high-entropy sampling degeneration, alongside a restorative preservation loss to safely anchor the integrity of entangled benign priors. Extensive experiments demonstrate that our method achieves surgical concept erasure performance across various domains with minimal training overhead, timely and elegently resolute the critical safety vulnerabilities inherent in emerging VAR architectures. Code is available at: https://github.com/limerenceysy/SACE}{https://github.com/limerenceysy/SACE.
Abstract:The rapid progress of visual autoregressive (VAR) models has unlocked a transformative frontier for high-fidelity text-to-image synthesis, while heightening concerns over the safety alignment of generated content. Naive application of existing erasure techniques to VAR models causes catastrophic semantic collapse and visual artifacts, since they are predominantly designed for the homogeneous denoising steps of diffusion models. To address this foundational challenge, we first propose the Semantic Singularity Axiom, which posits that any target semantic concept embedded within a prompt is definitively locked at Scale-0. Then rigorously validate this axiom through our proposed Incremental Semantic Saliency Analysis (ISSA),which also enable the community to transparently inspect the coarse-to-fine semantic injection process. Guided by this insight, we introduce the first scale-aware concept erasure framework (SACE) for VAR models. By strictly confining interventions to the first scale, our approach couples an Entropy-Regularized Erasure Objective to prevent high-entropy sampling degeneration, alongside a restorative preservation loss to safely anchor the integrity of entangled benign priors. Extensive experiments demonstrate that our method achieves surgical concept erasure performance across various domains with minimal training overhead, timely and elegently resolute the critical safety vulnerabilities inherent in emerging VAR architectures. Code is available at: this https URL}{this https URL.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2606.15819 [cs.CV]
(or arXiv:2606.15819v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.15819 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Full text under the publisher's open licence · 299 words
Related coverage
CleanVideo: Adaptive Concept Erasure for Text-to-Video Diffusion Models
arXiv · Sep 18, 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
From Concept Erasure to Style Purification: Contrastive Eigenbases for Artist Style Protection
arXiv · Sep 30, 2026
[AINews] The Future of Latent Space
Latent Space · Sep 25, 2026