2022 Pix2StructScreenshotParsingAsPr

From GM-RKB
(Redirected from Lee, Joshi et al., 2022)
Jump to navigation Jump to search

Subject Headings: Pix2Struct.

Notes

Cited By

Quotes

Abstract

Visually-situated language is ubiquitous -- sources range from textbooks with diagrams to web pages with images and tables, to mobile apps with buttons and forms. Perhaps due to this diversity, previous work has typically relied on domain-specific recipes with limited sharing of the underlying data, model architectures, and objectives. We present Pix2Struct, a pretrained image-to-text model for purely visual language understanding, which can be finetuned on tasks containing visually-situated language. Pix2Struct is pretrained by learning to parse masked screenshots of web pages into simplified HTML. The web, with its richness of visual elements cleanly reflected in the HTML structure, provides a large source of pretraining data well suited to the diversity of downstream tasks. Intuitively, this objective subsumes common pretraining signals such as OCR, language modeling, image captioning. In addition to the novel pretraining strategy, we introduce a variable-resolution input representation and a more flexible integration of language and vision inputs, where language prompts such as questions are rendered directly on top of the input image. For the first time, we show that a single pretrained model can achieve state-of-the-art results in six out of nine tasks across four domains: documents, illustrations, user interfaces, and natural images.

References

;

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2022 Pix2StructScreenshotParsingAsPrMing-Wei Chang
Urvashi Khandelwal
Kristina Toutanova
Kenton Lee
Mandar Joshi
Peter Shaw
Iulia Turc
Hexiang Hu
Fangyu Liu
Julian Eisenschlos
Pix2Struct: Screenshot Parsing As Pretraining for Visual Language Understanding10.48550/arXiv.2210.033472022