cs.CVNov 16, 2026

Example-Based Spatial Guidance for Training-Free Concept Erasure in Diffusion Models

Authors: Younghwan Kil, Giung Nam, Joonhyeong Park, Jinwoo Shin, Juho Lee

Organizations: Korea Advanced Institute of Science and Technology (KAIST)

Abstract

Text-to-image (T2I) diffusion models often synthesize policy-violating content, necessitating robust safeguards and rigorous evaluation frameworks. However, current approaches remain largely monolithic in both execution and evaluation, limited to coarse-grained categorization within broad unsafe concepts. Such approaches fail to account for the specific constituent factors of a violation, resulting in weak learning signals and allowing model collapse to bypass safety checks. To overcome these limitations, we propose Example-Based Spatial Guidance (EBSG), a training-free method that utilizes user-editable exemplar packs to provide granular spatial guidance for precise concept steering. Unlike existing methods that coarsely steer the entire image away from a broad concept, EBSG decomposes these categories into specific sub-concepts via exemplar text-image pairs, providing explicit, localized steering signals for more robust safety control. Furthermore, we introduce a vision language model (VLM)-based evaluation protocol that provides a fine-grained assessment, avoiding the pitfall of conventional binary evaluators that permit model collapse to bypass safety checks. Empirically, EBSG achieves state-of-the-art erasure performance across diverse safety datasets and concept categories, spanning four nudity benchmarks, seven I2P harmful-concept slices, and four MJA metaphor categories, while supporting multi-concept removal and transferring to a modern backbone, SD3.

Figures & tables

Appendix figures & tables47 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Erasing Thousands of Concepts: Towards Scalable and Practical Concept Erasure for Text-to-Image Diffusion Models

    Apr 12, 2026Hoigi Seo, Byung Hyun Lee, Jaehyun Cho +2Safe T2I GenerationDiffusion Model Unlearning

  2. DSS: Dynamic Semantic Steering for Robust Concept Erasure in Diffusion Models

    Apr 13, 2026Qinghui Gong, Zhengchun Zhou, Hua Meng +2Diffusion Model UnlearningT2I Diffusion Models

  3. EGLOCE: Training-Free Energy-Guided Latent Optimization for Concept Erasure

    Apr 10, 2026Junyeong Ahn, Seojin Yoon, Sungyong BaikDiffusion Model UnlearningSafe T2I Generation