GrOCE：面向文本到图像扩散模型的图引导在线概念擦除
GrOCE : Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models
提供一种免训练、在提示词层面做概念擦除的推理时干预手段：用动态语义图定位目标概念簇并选择性切断其语义成分，可直接复用于扩散模型的内容安全过滤与个性化生成中的概念抑制，无需微调权重。
原文依据 a training-free framework that performs precise and context-aware online removal of target concepts
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论概念擦除旨在从文本到图像扩散模型中移除有害、不当或受版权保护的内容，同时保留非目标语义。然而，现有方法要么依赖代价高昂的微调，要么仅做粗粒度的语义分离，往往会损害无关概念，且难以适应不断变化的概念集合。本文提出图引导在线概念擦除（GrOCE），这是一个免训练框架，可对目标概念执行精确且上下文感知的在线移除。GrOCE 构建动态语义图以识别目标概念簇，并在文本提示词中选择性地抑制其影响。它由三个协同组件构成：（1）动态语义图构建（Construct）在词汇概念上增量式地建立加权图，以刻画语义亲和度；（2）自适应簇识别（Identify）通过多跳遍历与基于扩散的打分提取目标概念簇，以量化语义影响；（3）选择性切断（Sever）从文本提示词中移除与目标簇相关的语义成分，同时保留非目标语义与全局句子结构。大量实验表明，GrOCE 在概念相似度（CS）与 Frechet Inception Distance（FID）指标上取得最优性能，实现了高效、准确且稳定的概念擦除。
英文原文摘要 1420 字符
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 Frechet Inception Distance (FID) metrics, offering efficient, accurate, and stable concept erasure.