GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models

2025-11-16

AI Summary

This work addresses the problem of fine-grained, tuning-free concept erasure in text-to-image diffusion models. We propose an online erasure framework grounded in dynamic semantic graphs, which constructs an incrementally updatable concept-relational graph. Our method integrates multi-hop similarity decay traversal, adaptive cluster identification, and selective edge pruning to precisely and adaptively isolate harmful, inappropriate, or copyright-protected concepts while fully preserving non-target semantics. Unlike coarse-grained semantic separation approaches, our framework enables interpretable, graph-guided semantic reasoning—avoiding irrelevant concept degradation and eliminating the need for model retraining. Experiments demonstrate state-of-the-art performance on both Concept Similarity (CS) and Fréchet Inception Distance (FID), significantly outperforming existing training-free erasure methods.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

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. To alleviate this issue, we propose Graph-Guided Online Concept Erasure (GrOCE), a training-free framework that performs precise and adaptive concept removal through graph-based semantic reasoning. GrOCE models concepts and their interrelations as a dynamic semantic graph, enabling principled reasoning over dependencies and fine-grained isolation of undesired content. It comprises three components: (1) Dynamic Topological Graph Construction for incremental graph building, (2) Adaptive Cluster Identification for multi-hop traversal with similarity-decay scoring, and (3) Selective Edge Severing for targeted edge removal while preserving global semantics. Extensive experiments demonstrate that GrOCE achieves state-of-the-art performance on Concept Similarity (CS) and Fréchet Inception Distance (FID) metrics, offering efficient, accurate, and stable concept erasure without retraining.

Problem

• Removing harmful or copyrighted content from diffusion models
• Preventing degradation of unrelated concepts during erasure
• Enabling adaptive concept removal without model retraining

Innovation

• Training-free framework for concept erasure
• Dynamic semantic graph models concept relationships
• Selective edge removal preserves global semantics