BARRIER: Bounded Activation Regions for Robust Information Erasure
cs.CV
Submitted: 2026-05-15
Updated: 2026-09-30
Code: https://github.com/OneAndZero24/BARRIER
Terminology
Sources
- SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
- Forget Many, Forget Right: Scalable and Precise Concept Unlearning in Diffusion Models
- SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
- Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
- CASteer: Cross-Attention Steering for Controllable Concept Erasure
- Concept Pinpoint Eraser for Text-to-image Diffusion Models via Residual Attention Gate
- The Illusion of Forgetting: Attack Unlearned Diffusion via Initial Latent Variable Optimization
- UnGuide: Learning to Forget with LoRA-Guided Diffusion Models
- LLM Unlearning via Neural Activation Redirection
- ActErase: A Training-Free Paradigm for Precise Concept Erasure via Activation Redirection
- Mass Concept Erasure in Diffusion Models with Concept Hierarchy
- Machine Unlearning of Features and Labels
- UnHype: CLIP-Guided Hypernetworks for Dynamic LoRA Unlearning
- Semantic Surgery: Zero-Shot Concept Erasure in Diffusion Models
- Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models