多模态文档理解综述：Pix2Struct

Pix2Struct

总览

题目: Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding
机构：谷歌
论文: https://arxiv.org/pdf/2210.03347.pdf
代码: https://github.com/google-research/pix2struct
任务:
特点:
方法:
前置相关工作：

摘要

视觉定位的语言（Visually-situated la