# Concept Steerers: Leveraging K-Sparse Autoencoders for Test-Time Controllable Generations

arXiv: https://arxiv.org/abs/2501.19066 · HTML: https://mlsanity.com/paper/2501.19066/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Interpreting and steering diffusion models, one of 13 directions in Mechanistic interpretability.
- Builds on 58 of its 76 references that are in the map.
- First posted on arXiv 31 Jan 2025.

## Placed in

- Mechanistic interpretability / [Interpreting and steering diffusion models](https://mlsanity.com/mechinterp/interpreting-and-steering-diffusion-models/)

## Builds on (most cited first, 30 of 58)

- [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)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [Diffusion Model Alignment Using Direct Preference Optimization](https://mlsanity.com/paper/2311.12908/) (2023)
- [Adversarial Diffusion Distillation](https://mlsanity.com/paper/2311.17042/) (2023)
- [Scaling and evaluating sparse autoencoders](https://mlsanity.com/paper/2406.04093/) (2024)
- [Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2303.07345/) (2023)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [Universal Guidance for Diffusion Models](https://mlsanity.com/paper/2302.07121/) (2023)
- [Unified Concept Editing in Diffusion Models](https://mlsanity.com/paper/2308.14761/) (2023)
- [SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation](https://mlsanity.com/paper/2310.12508/) (2023)
- [Ablating Concepts in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.13516/) (2023)
- [MACE: Mass Concept Erasure in Diffusion Models](https://mlsanity.com/paper/2403.06135/) (2024)
- [Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.17591/) (2023)
- [Human Preference Score: Better Aligning Text-to-Image Models with Human Preference](https://mlsanity.com/paper/2303.14420/) (2023)
- [Diffusion Models for Medical Image Analysis: A Comprehensive Survey](https://mlsanity.com/paper/2211.07804/) (2022)
- [Red-Teaming the Stable Diffusion Safety Filter](https://mlsanity.com/paper/2210.04610/) (2022)
- [Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?](https://mlsanity.com/paper/2310.10012/) (2023)
- [A General Framework for Inference-time Scaling and Steering of Diffusion Models](https://mlsanity.com/paper/2501.06848/) (2025)

## Built on by (most cited first, 7 of 7)

- [SAEdit: Token-level control for continuous image editing via Sparse AutoEncoder](https://mlsanity.com/paper/2510.05081/) (2025)
- [Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models](https://mlsanity.com/paper/2605.27813/) (2026)
- [PEAK: Precise and Persistent Concept Erasure via k-Sparse Autoencoders](https://mlsanity.com/paper/2608.10985/) (2026)
- [EDGE-Shield: Efficient Denoising-staGE Shield for Violative Content Filtering via Scalable Reference-Based Matching](https://mlsanity.com/paper/2604.06063/) (2026)
- [Robust and Generalizable Safety Steering for Text-to-Image Diffusion Transformers](https://mlsanity.com/paper/2605.30049/) (2026)
- [D-Scope: Decomposing and Steering Diffusion Transformers with Sparse Autoencoders](https://mlsanity.com/paper/2609.39625/) (2026)
- [Beyond the Linear Representation Hypothesis: Non-Linear Activation Steering in Text-to-Image Models](https://mlsanity.com/paper/2610.06945/) (2026)

## Abstract (arXiv)

Despite the remarkable progress in text-to-image generative models, they are prone to adversarial attacks and inadvertently generate unsafe, unethical content. Existing approaches often rely on fine-tuning models to remove specific concepts, which is computationally expensive, lacks scalability, and/or compromises generation quality. In this work, we propose a novel framework leveraging k-sparse autoencoders (k-SAEs) to enable efficient and interpretable concept manipulation in diffusion models. Specifically, we first identify interpretable monosemantic concepts in the latent space of text embeddings and leverage them to precisely steer the generation away or towards a given concept (e.g., nudity) or to introduce a new concept (e.g., photographic style) -- all during test time. Through extensive experiments, we demonstrate that our approach is very simple, requires no retraining of the base model nor LoRA adapters, does not compromise the generation quality, and is robust to adversarial prompt manipulations. Our method yields an improvement of 20.01% in unsafe concept removal, is effective in style manipulation, and is ~5x faster than the current state-of-the-art. Code is available at: https://github.com/kim-dahye/steerers