用单次推理完成图像风格迁移的高效建模，无需重复训练。
Compressing Image Style Training into a Single Model Forward
- 输入参考图后，一步生成LoRA权重实现风格迁移。
- 在多个数据集上提升风格保真度与提示词对齐效果。
- 适合需要快速风格化和多参考融合的应用场景。
基于扩散模型的风格迁移需权衡推理效率与风格保真度。基于适配器的方法虽高效，但将风格作为外部条件注入，可能弱化参考图像特定外观或复制其语义内容。优化类个性化方法如LoRA能更有效内化风格，但需为每种新风格单独训练。本文提出i2L（image-to-LoRA）框架，将风格LoRA训练过程压缩至单次前向传播。给定一个或多个参考图像，i2L可预测文本到图像模型的LoRA权重，实现无需逐风格优化的即时风格实例化。该架构结合图像编码器、可学习的LoRA查询及压缩解码头，生成适配矩阵。在语义多样化的风格对上训练，促使预测器保留外观特征同时抑制参考内容复制。在Z-Image、FLUX.2和Hidream-O1上的实验表明，i2L在风格保真度、提示词对齐和感知质量方面优于现有基线。由于i2L生成显式LoRA权重，还可支持非对称分类器无引导、多参考风格融合以及与可控生成模块的组合。
原文摘要 · Abstract (English)
Diffusion-based style transfer must balance inference efficiency with stylization fidelity. Adapter-based methods are efficient, but they inject style as an external condition and can either weaken reference-specific appearance or copy reference semantics into the generated image. Optimization-based personalization methods such as LoRA internalize style more effectively, but require a separate training process for every new style. We introduce i2L (image-to-LoRA), a framework that amortizes style LoRA training into a single forward pass. Given one or more reference images, i2L predicts LoRA weights for a text-to-image model, enabling immediate style instantiation without per-style optimization. The architecture combines an image encoder, learnable LoRA queries, and compressed decoding heads that generate adapted matrices. Training on semantically diverse style pairs encourages the predictor to preserve appearance cues while suppressing reference-content copying. Experiments on Z-Image, FLUX.2, and Hidream-O1 show that i2L improves style fidelity, prompt alignment, and perceptual quality over existing baselines. Because i2L produces explicit LoRA weights, it also supports asymmetric classifier-free guidance, multi-reference style fusion, and composition with controllable-generation modules.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。