GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908956248506368 |
|---|---|
| author | Han, Ning Ge, Zhenyu Han, Feng Sun, Yuhua Li, Chengqing Chen, Jingjing |
| author_facet | Han, Ning Ge, Zhenyu Han, Feng Sun, Yuhua Li, Chengqing Chen, Jingjing |
| contents | Concept erasure aims to remove harmful, inappropriate, or copyrighted content from text-to-image diffusion models while preserving non-target semantics. However, existing methods either rely on costly fine-tuning or apply coarse semantic separation, often degrading unrelated concepts and lacking adaptability to evolving concept sets. In this paper, we propose Graph-Guided Online Concept Erasure (GrOCE), a training-free framework that performs precise and context-aware online removal of target concepts. GrOCE constructs dynamic semantic graphs to identify clusters of target concepts and selectively suppress their influence within text prompts. It consists of three synergistic components: (1) dynamic semantic graph construction (Construct) incrementally builds a weighted graph over vocabulary concepts to capture semantic affinities; (2) adaptive cluster identification (Identify) extracts a target concept cluster through multi-hop traversal and diffusion-based scoring to quantify semantic influence; and (3) selective severing (Sever) removes semantic components associated with the target cluster from the text prompt while retaining non-target semantics and the global sentence structure. Extensive experiments demonstrate that GrOCE achieves state-of-the-art performance on the Concept Similarity (CS) and Fréchet Inception Distance (FID) metrics, offering efficient, accurate, and stable concept erasure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_12968 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models Han, Ning Ge, Zhenyu Han, Feng Sun, Yuhua Li, Chengqing Chen, Jingjing Computer Vision and Pattern Recognition Computer Vision and Pattern Recognition (cs.CV) Concept erasure aims to remove harmful, inappropriate, or copyrighted content from text-to-image diffusion models while preserving non-target semantics. However, existing methods either rely on costly fine-tuning or apply coarse semantic separation, often degrading unrelated concepts and lacking adaptability to evolving concept sets. In this paper, we propose Graph-Guided Online Concept Erasure (GrOCE), a training-free framework that performs precise and context-aware online removal of target concepts. GrOCE constructs dynamic semantic graphs to identify clusters of target concepts and selectively suppress their influence within text prompts. It consists of three synergistic components: (1) dynamic semantic graph construction (Construct) incrementally builds a weighted graph over vocabulary concepts to capture semantic affinities; (2) adaptive cluster identification (Identify) extracts a target concept cluster through multi-hop traversal and diffusion-based scoring to quantify semantic influence; and (3) selective severing (Sever) removes semantic components associated with the target cluster from the text prompt while retaining non-target semantics and the global sentence structure. Extensive experiments demonstrate that GrOCE achieves state-of-the-art performance on the Concept Similarity (CS) and Fréchet Inception Distance (FID) metrics, offering efficient, accurate, and stable concept erasure. |
| title | GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models |
| topic | Computer Vision and Pattern Recognition Computer Vision and Pattern Recognition (cs.CV) |
| url | https://arxiv.org/abs/2511.12968 |