Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Models
Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated features can serve as controllable intervention points. In this work, we systematically evaluate this assumption in the context of object erasure and steering in diffusion models. We show that while SAEs reliably detect and localize semantic concepts within diffusion model activations, direct intervention in their latent space frequently induces out-of-distribution activations, resulting in severe visual artifacts. To disentangle detection from intervention, we use SAE activations purely as semantic detectors to identify image regions containing the target object, and replace those patch embeddings with the ones that do not contain it. This detection-based replacement preserves the diffusion model's activation statistics and produces significantly cleaner erasure results than latent steering. Our findings reveal a fundamental gap between concept detection and concept intervention in diffusion models: monosemantic or sparse features are not inherently suitable as control knobs for steering. These results position SAEs as powerful interpretability tools for analyzing generative models, but highlight important limitations when used for direct manipulation, such as unlearning.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders
Diffusion models, while powerful, can inadvertently generate harmful or undesirable content, raising significant ethical and safety concerns. Recent machine unlearning approaches offer potential solutions but often lack …
Adversarial AttackDenoisingfeature selectionMachine UnlearningSAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders
Concept unlearning in diffusion models is hampered by feature splitting, where concepts are distributed across many latent features, making their removal challenging and computationally expensive. We introduce SAEmnesia,…
SAUCE: Selective Concept Unlearning in Vision-Language Models with Sparse Autoencoders
Unlearning methods for vision-language models (VLMs) have primarily adapted techniques from large language models (LLMs), relying on weight updates that demand extensive annotated forget sets. Moreover, these methods per…
TouchAnything: Diffusion-Guided 3D Reconstruction from Sparse Robot Touches
Accurate object geometry estimation is essential for many downstream tasks, including robotic manipulation and physical interaction. Although vision is the dominant modality for shape perception, it becomes unreliable un…
3D ReconstructionSAEs $\textit{Can}$ Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMs
Machine unlearning is a promising approach to improve LLM safety by removing unwanted knowledge from the model. However, prevailing gradient-based unlearning methods suffer from issues such as high computational costs, h…
Computational Efficiencyfeature selectionMachine Unlearning