PureCC：面向文本到图像概念定制的纯学习框架

摘要

现有概念定制方法在高保真度和多概念定制方面已取得显著成果，但往往忽略了学习新个性化概念时对原始模型行为与能力的影响。为解决此问题，我们提出PureCC方法。该方法创新性地引入解耦学习目标，将目标概念的隐式引导与原始条件预测相结合。这种分离形式使PureCC在训练过程中能充分聚焦于原始模型特性。基于此目标，PureCC设计了双分支训练流程：包含提供纯净目标概念表征的冻结提取器作为隐式引导，以及生成原始条件预测的可训练流模型，二者协同实现个性化概念的纯净学习。此外，PureCC引入新型自适应引导系数λ*，动态调整目标概念的引导强度，平衡定制保真度与模型保护。大量实验表明，PureCC在实现高保真概念定制的同时，能保持原始模型行为与能力，达到业界领先水平。代码已开源：https://github.com/lzc-sg/PureCC。

English

Existing concept customization methods have achieved remarkable outcomes in high-fidelity and multi-concept customization. However, they often neglect the influence on the original model's behavior and capabilities when learning new personalized concepts. To address this issue, we propose PureCC. PureCC introduces a novel decoupled learning objective for concept customization, which combines the implicit guidance of the target concept with the original conditional prediction. This separated form enables PureCC to substantially focus on the original model during training. Moreover, based on this objective, PureCC designs a dual-branch training pipeline that includes a frozen extractor providing purified target concept representations as implicit guidance and a trainable flow model producing the original conditional prediction, jointly achieving pure learning for personalized concepts. Furthermore, PureCC introduces a novel adaptive guidance scale λ^star to dynamically adjust the guidance strength of the target concept, balancing customization fidelity and model preservation. Extensive experiments show that PureCC achieves state-of-the-art performance in preserving the original behavior and capabilities while enabling high-fidelity concept customization. The code is available at https://github.com/lzc-sg/PureCC.

PureCC：面向文本到图像概念定制的纯学习框架

PureCC: Pure Learning for Text-to-Image Concept Customization

摘要

Support