Cover Image

Closing the Safety Gap: Surgical Concept Erasure in Visual Autoregressive Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhong, Xinhao, Zhou, Yimin, Zhang, Zhiqi, Li, Junhao, Sun, Yi, Chen, Bin, Xia, Shu-Tao, Wang, Xuan, Xu, Ke
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

FlowErase-RL: Rethinking Concept Erasure as Reward Optimization in Flow Matching Models
by: Sun, Yi, et al.
Published: (2026)

Differential Vector Erasure: Unified Training-Free Concept Erasure for Flow Matching Models
by: Zhang, Zhiqi, et al.
Published: (2026)

ActErase: A Training-Free Paradigm for Precise Concept Erasure via Activation Redirection
by: Sun, Yi, et al.
Published: (2026)

CPC-VAR:Continual Personalized and Compositional Generation in Visual Autoregressive Models
by: Li, Junhao, et al.
Published: (2026)

Going Beyond Feature Similarity: Effective Dataset Distillation based on Class-Aware Conditional Mutual Information
by: Zhong, Xinhao, et al.
Published: (2024)

FastVAR: Linear Visual Autoregressive Modeling via Cached Token Pruning
by: Guo, Hang, et al.
Published: (2025)

Customize Your Visual Autoregressive Recipe with Set Autoregressive Modeling
by: Liu, Wenze, et al.
Published: (2024)

Hierarchical Features Matter: A Deep Exploration of Progressive Parameterization Method for Dataset Distillation
by: Zhong, Xinhao, et al.
Published: (2024)

ACE: Anti-Editing Concept Erasure in Text-to-Image Models
by: Wang, Zihao, et al.
Published: (2025)

ACE: Attentional Concept Erasure in Diffusion Models
by: Carter, Finn
Published: (2025)

FADE: Adversarial Concept Erasure in Flow Models
by: Fu, Zixuan, et al.
Published: (2025)

Robust Concept Erasure Using Task Vectors
by: Pham, Minh, et al.
Published: (2024)

Temporal Saliency-Guided Distillation: A Scalable Framework for Distilling Video Datasets
by: Gu, Xulin, et al.
Published: (2025)

GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models
by: Han, Ning, et al.
Published: (2025)

M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Models
by: Weng, Ju-Hsuan, et al.
Published: (2025)

A Single Neuron Works: Precise Concept Erasure in Text-to-Image Diffusion Models
by: He, Qinqin, et al.
Published: (2025)

EraseAnything: Enabling Concept Erasure in Rectified Flow Transformers
by: Gao, Daiheng, et al.
Published: (2024)

Minimalist Concept Erasure in Generative Models
by: Zhang, Yang, et al.
Published: (2025)

Enhancing Gradient Inversion Attacks in Federated Learning via Hierarchical Feature Optimization
by: Fang, Hao, et al.
Published: (2026)

Erased or Dormant? Rethinking Concept Erasure Through Reversibility
by: Liu, Ping, et al.
Published: (2025)

TRACE: Trajectory-Constrained Concept Erasure in Diffusion Models
by: Carter, Finn
Published: (2025)

Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models
by: Zhang, Yimeng, et al.
Published: (2024)

RealEra: Semantic-level Concept Erasure via Neighbor-Concept Mining
by: Liu, Yufan, et al.
Published: (2024)

VVS: Accelerating Speculative Decoding for Visual Autoregressive Generation via Partial Verification Skipping
by: Dong, Haotian, et al.
Published: (2025)

Semantic Surgery: Zero-Shot Concept Erasure in Diffusion Models
by: Xiong, Lexiang, et al.
Published: (2025)

VideoEraser: Concept Erasure in Text-to-Video Diffusion Models
by: Xu, Naen, et al.
Published: (2025)

Efficient Conditional Generation on Scale-based Visual Autoregressive Models
by: Liu, Jiaqi, et al.
Published: (2025)

Mass Concept Erasure in Diffusion Models with Concept Hierarchy
by: Tu, Jiahang, et al.
Published: (2026)

Prototype-Guided Concept Erasure in Diffusion Models
by: Cai, Yuze, et al.
Published: (2026)

SPEED: Scalable, Precise, and Efficient Concept Erasure for Diffusion Models
by: Li, Ouxiang, et al.
Published: (2025)

Reliable and Efficient Concept Erasure of Text-to-Image Diffusion Models
by: Gong, Chao, et al.
Published: (2024)

Visual Self-Refinement for Autoregressive Models
by: Wang, Jiamian, et al.
Published: (2025)

Dynamic Mixture-of-Experts for Visual Autoregressive Model
by: Vincenti, Jort, et al.
Published: (2025)

Erasing More Than Intended? How Concept Erasure Degrades the Generation of Non-Target Concepts
by: Amara, Ibtihel, et al.
Published: (2025)

Where Concept Erasure Should Occur: Concept-Layer Alignment in Text-to-Video Diffusion Models
by: Xie, Yiwei, et al.
Published: (2026)

Adversarial Error Correction for Visual Autoregressive Generation
by: Bi, Ligong, et al.
Published: (2026)

Rethinking Robust Adversarial Concept Erasure in Diffusion Models
by: Yin, Qinghong, et al.
Published: (2025)

Dynamic Eraser for Guided Concept Erasure in Diffusion Models
by: Gong, Qinghui
Published: (2026)

Separable Multi-Concept Erasure from Diffusion Models
by: Zhao, Mengnan, et al.
Published: (2024)

Unified Autoregressive Visual Generation and Understanding with Continuous Tokens
by: Fan, Lijie, et al.
Published: (2025)