使用双编码器改进场景文本编辑的扩散模型（英文版）

英文标题：Improving Diffusion Models for Scene Text Editing with Dual Encoders中文摘要：DIFFSTE 是一个改善预训练扩散模型性能的双编码器设计框架，通过指令调整训练，实现了场景文本编辑中正确文本渲染和风格控制的任务，使其具有零 - shot 泛化能力。英文摘要：Scene text editing is a challenging task that involves modifying or insertingspecified texts in an image while maintaining its natural and realisticappearance. Most previous approaches to this task rely on style-transfer modelsthat crop out text regions and feed them into image transfer models, such asGANs. However, these methods a