Cover Image

SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Cassano, Enrico, Renzulli, Riccardo, Nurisso, Marco, Zaffaroni, Mirko, Perotti, Alan, Grangetto, Marco
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Capsule Networks Do Not Need to Model Everything
by: Renzulli, Riccardo, et al.
Published: (2022)

When Does Pruning Benefit Vision Representations?
by: Cassano, Enrico, et al.
Published: (2025)

MedSAE: Dissecting MedCLIP Representations with Sparse Autoencoders
by: Renzulli, Riccardo, et al.
Published: (2025)

AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data
by: Zaffaroni, Mirko, et al.
Published: (2024)

Sparse Autoencoder as a Zero-Shot Classifier for Concept Erasing in Text-to-Image Diffusion Models
by: Tian, Zhihua, et al.
Published: (2025)

The Geometry of Representational Failures in Vision Language Models
by: Savietto, Daniele, et al.
Published: (2026)

Hierarchical Object-Centric Learning with Capsule Networks
by: Renzulli, Riccardo
Published: (2024)

Boost Your NeRF: A Model-Agnostic Mixture of Experts Framework for High Quality and Efficient Rendering
by: Di Sario, Francesco, et al.
Published: (2024)

Attributes Shape the Embedding Space of Face Recognition Models
by: Leroy, Pierrick, et al.
Published: (2025)

Automated Prediction of Paravalvular Regurgitation before Transcatheter Aortic Valve Implantation
by: Cannito, Michele, et al.
Published: (2026)

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

AI-Assisted Diagnosis for Covid-19 CXR Screening: From Data Collection to Clinical Validation
by: Barbano, Carlo Alberto, et al.
Published: (2024)

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

GoDe: Gaussians on Demand for Progressive Level of Detail and Scalable Compression
by: Di Sario, Francesco, et al.
Published: (2025)

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

SAUCE: Selective Concept Unlearning in Vision-Language Models with Sparse Autoencoders
by: Li, Qing, et al.
Published: (2025)

SPARC: Concept-Aligned Sparse Autoencoders for Cross-Model and Cross-Modal Interpretability
by: Nasiri-Sarvi, Ali, et al.
Published: (2025)

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

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

WiGNet: Windowed Vision Graph Neural Network
by: Spadaro, Gabriele, et al.
Published: (2024)

Erase Diffusion: Empowering Object Removal Through Calibrating Diffusion Pathways
by: Liu, Yi, et al.
Published: (2025)

Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models
by: Yeung, Calvin, et al.
Published: (2026)

Improving Weakly-Supervised Object Localization Using Adversarial Erasing and Pseudo Label
by: Kang, Byeongkeun, et al.
Published: (2024)

Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates
by: Spadaro, Gabriele, et al.
Published: (2025)

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

A Framework for Evaluating Zero-Shot Image Generation in Concept-based Explainability
by: Astolfi, Giacomo, et al.
Published: (2026)

Erased but Exploitable: Black-box Embedding-Aware Prompting Against Unlearned Text-to-Image Diffusion Models
by: Koma, Arian Komaei, et al.
Published: (2026)

One-Step is Enough: Sparse Autoencoders for Text-to-Image Diffusion Models
by: Surkov, Viacheslav, et al.
Published: (2024)

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

Latent Diffusion Model without Variational Autoencoder
by: Shi, Minglei, et al.
Published: (2025)

Causal Interpretation of Sparse Autoencoder Features in Vision
by: Han, Sangyu, et al.
Published: (2025)

TIDE : Temporal-Aware Sparse Autoencoders for Interpretable Diffusion Transformers in Image Generation
by: Huang, Victor Shea-Jay, et al.
Published: (2025)

Flip Learning: Weakly Supervised Erase to Segment Nodules in Breast Ultrasound
by: Huang, Yuhao, et al.
Published: (2025)

Erasing with Precision: Evaluating Specific Concept Erasure from Text-to-Image Generative Models
by: Fuchi, Masane, et al.
Published: (2025)

Latent-Compressed Variational Autoencoder for Video Diffusion Models
by: Guan, Jiarui, et al.
Published: (2026)

Beyond Labels: A Self-Supervised Framework with Masked Autoencoders and Random Cropping for Breast Cancer Subtype Classification
by: Chiocchetti, Annalisa, et al.
Published: (2024)

How to Blend Concepts in Diffusion Models
by: Olearo, Lorenzo, et al.
Published: (2024)

HadaNorm: Diffusion Transformer Quantization through Mean-Centered Transformations
by: Federici, Marco, et al.
Published: (2025)

Disentangled Sparse Representations for Concept-Separated Diffusion Unlearning
by: Kim, Hyeonjin, et al.
Published: (2026)

Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models
by: Chen, Junyu, et al.
Published: (2024)