Business Document Information Extraction: Towards Practical Benchmarks

Skalický, Matyáš; Šimsa, Štěpán; Uřičář, Michal; Šulc, Milan

doi:10.1007/978-3-031-13643-6_8

Cited by 8 publications

(3 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The goal of the first track is to localize key information of pre-defined categories (field types) in the document. It is derived from the task of Key Information Localization and Extraction (KILE), as defined in [26].…”

Section: Track 1: Key Information Localization and Extractionmentioning

confidence: 99%

“…This track is derived from the task of Line Item Recognition (LIR) [26] and is related to Table Understanding [13] and Table Extraction [8,35] -problems where the tabular structure is also crucial for IE. Unlike these tasks, LIR does not explicitly rely on the structure but rather reflects the information to be extracted and stored.…”

Section: Track 2: Line Item Recognitionmentioning

confidence: 99%

“…Information extraction approaches must handle varying layouts, semantic fields and multiple input modalities at the intersection of computer vision, natural language processing and information retrieval. While there has been progress on the task [4,7,14,15,18,19,25,34], there is no publicly available large-scale benchmark to train and compare these approaches, an issue that has been noted by several authors [5,16,24,26,29]. Existing approaches are trained on privately collected datasets, hindering their reproducibility, fair comparisons and tracking field progression [11,23,24].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

DocILE 2023 Teaser: Document Information Localization and Extraction

Šimsa¹,

Müller²,

Skalický³

et al. 2023

Preprint

View full text Add to dashboard Cite

The lack of data for information extraction (IE) from semistructured business documents is a real problem for the IE community. Publications relying on large-scale datasets use only proprietary, unpublished data due to the sensitive nature of such documents. Publicly available datasets are mostly small and domain-specific. The absence of a large-scale public dataset or benchmark hinders the reproducibility and cross-evaluation of published methods. The DocILE 2023 competition, hosted as a lab at the CLEF 2023 conference and as an ICDAR 2023 competition, will run the first major benchmark for the tasks of Key Information Localization and Extraction (KILE) and Line Item Recognition (LIR) from business documents. With thousands of annotated real documents from open sources, a hundred thousand of generated synthetic documents, and nearly a million unlabeled documents, the DocILE lab comes with the largest publicly available dataset for KILE and LIR. We are looking forward to contributions from the Computer Vision, Natural Language Processing, Information Retrieval, and other communities. The data, baselines, code and up-to-date information about the lab and competition are available at https://docile.rossum.ai/.

show abstract

Section: Track 1: Key Information Localization and Extractionmentioning

confidence: 99%