# PEAK: Precise and Persistent Concept Erasure via k-Sparse Autoencoders

arXiv: https://arxiv.org/abs/2608.10985 · HTML: https://mlsanity.com/paper/2608.10985/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models.
- Builds on 25 of its 32 references that are in the map.
- First posted on arXiv 11 Aug 2026.

## Placed in

- Diffusion models / [Robustness, privacy and adversarial aspects](https://mlsanity.com/diffusion/robustness-privacy-and-adversarial-aspects/)
- Mechanistic interpretability / [Interpreting and steering diffusion models](https://mlsanity.com/mechinterp/interpreting-and-steering-diffusion-models/)

## Builds on (most cited first, 25 of 25)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [Unified Concept Editing in Diffusion Models](https://mlsanity.com/paper/2308.14761/) (2023)
- [MACE: Mass Concept Erasure in Diffusion Models](https://mlsanity.com/paper/2403.06135/) (2024)
- [Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?](https://mlsanity.com/paper/2310.10012/) (2023)
- [k-Sparse Autoencoders](https://mlsanity.com/paper/1312.5663/) (2013)
- [MMA-Diffusion: MultiModal Attack on Diffusion Models](https://mlsanity.com/paper/2311.17516/) (2023)
- [One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications](https://mlsanity.com/paper/2312.16145/) (2023)
- [Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models](https://mlsanity.com/paper/2405.15234/) (2024)
- [Localizing and Editing Knowledge in Text-to-Image Generative Models](https://mlsanity.com/paper/2310.13730/) (2023)
- [Interpreting Attention Layer Outputs with Sparse Autoencoders](https://mlsanity.com/paper/2406.17759/) (2024)
- [Precise, Fast, and Low-cost Concept Erasure in Value Space: Orthogonal Complement Matters](https://mlsanity.com/paper/2412.06143/) (2024)
- [SPEED: Scalable, Precise, and Efficient Concept Erasure for Diffusion Models](https://mlsanity.com/paper/2503.07392/) (2025)
- [Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?](https://mlsanity.com/paper/2509.03516/) (2025)
- [TraSCE: Trajectory Steering for Concept Erasure](https://mlsanity.com/paper/2412.07658/) (2024)
- [SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders](https://mlsanity.com/paper/2509.21379/) (2025)
- [Concept Steerers: Leveraging K-Sparse Autoencoders for Test-Time Controllable Generations](https://mlsanity.com/paper/2501.19066/) (2025)
- [OrthoEraser: Coupled-Neuron Orthogonal Projection for Concept Erasure](https://mlsanity.com/paper/2603.11493/) (2026)
- [Co-occurring Associated REtained concepts in Diffusion Unlearning](https://mlsanity.com/paper/2606.24192/) (2026)

## Built on by (most cited first, 1 of 1)

- [SharedSAE: a unified sparse dictionary across u-net blocks and denoising time for diffusion models](https://mlsanity.com/paper/2610.13997/) (2026)

## Abstract (arXiv)

Erasing concepts from large-scale text-to-image (T2I) diffusion models has become increasingly crucial due to the growing concerns over copyright infringement, privacy violations, and offensive content. Existing approaches struggle to achieve both precise and persistent concept erasure: inaccurate localization of concept-related representations may cause unintended semantic interference, while incomplete removal of the underlying concept knowledge allows adversarial recovery. To address this dilemma, we propose PEAK, a textbftextitprecise and textbftextitpersistent concept erasure framework via k-Sparse Autoencoders (kSAEs). PEAK first trains a kSAE on internal activations of the diffusion denoising network to decompose dense representations into interpretable sparse features. By contrasting sparse activations induced by target and non-target prompts, PEAK identifies a compact set of target-specific features according to both activation strength and frequency. These localized features are then used for parameter optimization, where PEAK selectively suppresses target-related activations while preserving complementary non-target ones towards the original model. This feature-guided optimization embeds concept erasure directly into diffusion parameters, eliminating the need for additional inference-time intervention and facilitating effective persistence against adversarial attacks. Extensive experiments demonstrate that PEAK achieves effective and robust concept erasure. On the I2P benchmark, PEAK reduces NudeNet detections from 582 to 6, lowers the average attack success rate (ASR) from 96.52% to 5.63%, and preserves general generation quality on MS-COCO with a near-zero KID. Our code and models are available at: https://github.com/manmanTAT/PEAK