统一个性化理解、生成与编辑
Unified Personalized Understanding, Generating and Editing
提出结构化解耦的概念令牌与显式知识回放机制，可在统一多模态模型中实现理解、生成与编辑的个性化，避免跨任务干扰；其跨任务评估协议也为个性化统一模型提供了可复用的评测思路。
原文依据 OmniPersona introduces structurally decoupled concept tokens, allocating dedicated subspaces for different tasks to minimize interference
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论统一大型多模态模型（LMMs）在通用多模态理解与生成方面取得了显著进展。然而，它们仍运行在“一刀切”范式下，难以以一致且可控的方式对用户特定概念（例如生成一张 \texttt{<maeve>} 的照片）进行建模。现有的个性化方法通常依赖外部检索，效率低下且难以良好地集成到统一多模态流程中。近期个性化统一模型引入可学习软提示来编码概念信息，但它们要么将理解与生成耦合，要么依赖复杂的多阶段训练，导致跨任务干扰，并最终产生模糊或错位的个性化知识。我们提出 OmniPersona，一个面向统一 LMMs 的端到端个性化框架，首次在单一架构中集成个性化理解、生成与图像编辑。OmniPersona 引入结构化解耦的概念令牌，为不同任务分配专用子空间以最小化干扰，并纳入显式知识回放机制，跨任务传播个性化属性知识，从而实现一致的个性化行为。为系统评估统一个性化，我们提出 \texttt{OmniPBench}，在公开的 UnifyBench 概念集基础上扩展了个性化编辑任务以及集成理解、生成与编辑的跨任务评估协议。实验结果表明，OmniPersona 在多样化个性化任务上提供了有竞争力且稳健的性能。我们希望 OmniPersona 能作为一个强基线，并推动可控、统一个性化的进一步研究。
英文原文摘要 1774 字符
Unified large multimodal models (LMMs) have achieved remarkable progress in general-purpose multimodal understanding and generation. However, they still operate under a "one-size-fits-all" paradigm and struggle to model user-specific concepts (e.g., generate a photo of \texttt \<maeve> ) in a consistent and controllable manner. Existing personalization methods typically rely on external retrieval, which is inefficient and poorly integrated into unified multimodal pipelines. Recent personalized unified models introduce learnable soft prompts to encode concept information, yet they either couple understanding and generation or depend on complex multi-stage training, leading to cross-task interference and ultimately to fuzzy or misaligned personalized knowledge.We present OmniPersona, an end-to-end personalization framework for unified LMMs that, for the first time, integrates personalized understanding, generation, and image editing within a single architecture. OmniPersona introduces structurally decoupled concept tokens, allocating dedicated subspaces for different tasks to minimize interference, and incorporates an explicit knowledge replay mechanism that propagates personalized attribute knowledge across tasks, enabling consistent personalized behavior.To systematically evaluate unified personalization, we propose \texttt OmniPBench , extending the public UnifyBench concept set with personalized editing tasks and cross-task evaluation protocols integrating understanding, generation, and editing. Experimental results demonstrate that OmniPersona delivers competitive and robust performance across diverse personalization tasks. We hope OmniPersona will serve as a strong baseline and spur further research on controllable, unified personalization.