Omni-Attribute：面向视觉概念个性化的开放词表属性编码器
Omni-Attribute: Open-vocabulary Attribute Encoder for Visual Concept Personalization
把「属性」当作可检索的开放词表嵌入来学习，用正负属性配对数据加对比解耦目标，替代通用编码器的整体嵌入；做个性化/风格迁移时可直接复用该编码器来隔离身份、光照、风格等单一因素，减少信息泄漏。
原文依据 the first open-vocabulary image attribute encoder designed to learn high-fidelity, attribute-specific representations
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉概念个性化的目标是将特定的图像属性（如身份、表情、光照和风格）迁移到未见过的场景中。然而，现有方法依赖通用图像编码器输出的整体嵌入，这类嵌入将多种视觉因素纠缠在一起，难以分离出单一属性，因而常常导致信息泄漏与合成结果不连贯。为解决这一局限，我们提出 Omni-Attribute，这是首个旨在学习高保真、属性特定表示的开放词表图像属性编码器。我们的方法对数据和模型进行了联合设计：(i) 我们构建了带有正、负属性标注的语义关联图像对，以显式地教会编码器应当保留或抑制哪些内容；(ii) 我们采用双目标训练范式，在生成保真度与对比解耦之间取得平衡。所得嵌入在开放词表属性检索、个性化以及组合式生成上均表现有效，在多个基准上取得了最优性能。
英文原文摘要 1099 字符
Visual concept personalization aims to transfer only specific image attributes, such as identity, expression, lighting, and style, into unseen contexts. However, existing methods rely on holistic embeddings from general-purpose image encoders, which entangle multiple visual factors and make it difficult to isolate a single attribute. This often leads to information leakage and incoherent synthesis. To address this limitation, we introduce Omni-Attribute, the first open-vocabulary image attribute encoder designed to learn high-fidelity, attribute-specific representations. Our approach jointly designs the data and model: (i) we curate semantically linked image pairs annotated with positive and negative attributes to explicitly teach the encoder what to preserve or suppress; and (ii) we adopt a dual-objective training paradigm that balances generative fidelity with contrastive disentanglement. The resulting embeddings prove effective for open-vocabulary attribute retrieval, personalization, and compositional generation, achieving state-of-the-art performance across multiple benchmarks.