English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Key information extraction (KIE) from document images requires understandingthe contextual and spatial semantics of texts in two-dimensional (2D) space.Many recent studies try to solve the task by developing pre-trained languagemodels focusing on combining visual features from document images with textsand their layout. On the other hand, this paper tackles the problem by goingback to the basic: effective combination of text and layout. Specifically, wepropose a pre-trained language model, named BROS (BERT Relying On Spatiality),that encodes relative positions of texts in 2D space and learns from unlabeleddocuments with area-masking strategy. With this optimized training scheme forunderstanding texts in 2D space, BROS shows comparable or better performancecompared to previous methods on four KIE benchmarks (FUNSD, SROIE*, CORD, andSciTSR) without relying on visual features. This paper also reveals tworeal-world challenges in KIE tasks-(1) minimizing the error from incorrect textordering and (2) efficient learning from fewer downstream examples-anddemonstrates the superiority of BROS over previous methods. Code is availableat https://github.com/clovaai/bros.

BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documents