Copertina

Erased, But Not Forgotten: Erased Rectified Flow Transformers Still Remain Unsafe Under Concept Attack

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Jiang, Nanxiang, Fan, Zhaoxin, Kang, Enhan, Gao, Daiheng, Zhou, Yun, Chang, Yanxia, Zhu, Zheng, Jin, Yeying, Wu, Wenjun
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

EraseAnything++: Enabling Concept Erasure in Rectified Flow Transformers Leveraging Multi-Object Optimization
di: Fan, Zhaoxin, et al.
Pubblicazione: (2026)

Z-Erase: Enabling Concept Erasure in Single-Stream Diffusion Transformers
di: Jiang, Nanxiang, et al.
Pubblicazione: (2026)

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

Revoking Amnesia: RL-based Trajectory Optimization to Resurrect Erased Concepts in Diffusion Models
di: Gao, Daiheng, et al.
Pubblicazione: (2025)

Erased but Not Forgotten: How Backdoors Compromise Concept Erasure
di: Braun, Tobias, et al.
Pubblicazione: (2025)

SAGE: Exploring the Boundaries of Unsafe Concept Domain with Semantic-Augment Erasing
di: Zhu, Hongguang, et al.
Pubblicazione: (2025)

Pruning for Robust Concept Erasing in Diffusion Models
di: Yang, Tianyun, et al.
Pubblicazione: (2024)

EAR: Erasing Concepts from Unified Autoregressive Models
di: Fan, Haipeng, et al.
Pubblicazione: (2025)

Erasing Concepts, Steering Generations: A Comprehensive Survey of Concept Suppression
di: Xie, Yiwei, et al.
Pubblicazione: (2025)

ESC: Erasing Space Concept for Knowledge Deletion
di: Lee, Tae-Young, et al.
Pubblicazione: (2025)

When Are Concepts Erased From Diffusion Models?
di: Lu, Kevin, et al.
Pubblicazione: (2025)

You Only Erase Once: Erasing Anything without Bringing Unexpected Content
di: Zhu, Yixing, et al.
Pubblicazione: (2026)

Erase then Rectify: A Training-Free Parameter Editing Approach for Cost-Effective Graph Unlearning
di: Yang, Zhe-Rui, et al.
Pubblicazione: (2024)

EffectErase: Joint Video Object Removal and Insertion for High-Quality Effect Erasing
di: Fu, Yang, et al.
Pubblicazione: (2026)

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

Do Concept Replacement Techniques Really Erase Unacceptable Concepts?
di: Das, Anudeep, et al.
Pubblicazione: (2025)

Rectified Flow For Structure Based Drug Design
di: Zhang, Daiheng, et al.
Pubblicazione: (2024)

Side Effects of Erasing Concepts from Diffusion Models
di: Saha, Shaswati, et al.
Pubblicazione: (2025)

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

Erasing Undesirable Concepts in Diffusion Models with Adversarial Preservation
di: Bui, Anh, et al.
Pubblicazione: (2024)

Concept Corrector: Erase concepts on the fly for text-to-image diffusion models
di: Meng, Zheling, et al.
Pubblicazione: (2025)

PROBE: Diagnosing Residual Concept Capacity in Erased Text-to-Video Diffusion Models
di: Xie, Yiwei, et al.
Pubblicazione: (2026)

EraseFlow: Learning Concept Erasure Policies via GFlowNet-Driven Alignment
di: Kusumba, Abhiram, et al.
Pubblicazione: (2025)

SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders
di: Cassano, Enrico, et al.
Pubblicazione: (2025)

EraseDraw: Learning to Draw Step-by-Step via Erasing Objects from Images
di: Canberk, Alper, et al.
Pubblicazione: (2024)

Bi-Erasing: A Bidirectional Framework for Concept Removal in Diffusion Models
di: Chen, Hao, et al.
Pubblicazione: (2025)

Beyond Fixed Anchors: Precisely Erasing Concepts with Sibling Exclusive Counterparts
di: Zhang, Tong, et al.
Pubblicazione: (2025)

Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Erasers
di: Huang, Chi-Pin, et al.
Pubblicazione: (2023)

One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications
di: Lyu, Mengyao, et al.
Pubblicazione: (2023)

Erase to Improve: Erasable Reinforcement Learning for Search-Augmented LLMs
di: Wang, Ziliang, et al.
Pubblicazione: (2025)

Scaling of Long-Range Loop-Erased Random Walks
di: Xiao, Tianning, et al.
Pubblicazione: (2026)

Prompt Sliders for Fine-Grained Control, Editing and Erasing of Concepts in Diffusion Models
di: Sridhar, Deepak, et al.
Pubblicazione: (2024)

T2VUnlearning: A Concept Erasing Method for Text-to-Video Diffusion Models
di: Ye, Xiaoyu, et al.
Pubblicazione: (2025)

Erasing Concepts from Text-to-Image Diffusion Models with Few-shot Unlearning
di: Fuchi, Masane, et al.
Pubblicazione: (2024)

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

Erasing Thousands of Concepts: Towards Scalable and Practical Concept Erasure for Text-to-Image Diffusion Models
di: Seo, Hoigi, et al.
Pubblicazione: (2026)

Slowing Learning by Erasing Simple Features
di: Quirke, Lucia, et al.
Pubblicazione: (2025)

Erasing Undesirable Influence in Diffusion Models
di: Wu, Jing, et al.
Pubblicazione: (2024)

Érase una vez unos daticos...
di: Sergio A. Rojas
Pubblicazione: (2017)

Step-by-Step Reasoning Attack: Revealing 'Erased' Knowledge in Large Language Models
di: Sinha, Yash, et al.
Pubblicazione: (2025)