Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

LAMBERT: Layout-Aware (Language) Modeling for information extraction

About

We introduce a simple new approach to the problem of understanding documents where non-trivial layout influences the local semantics. To this end, we modify the Transformer encoder architecture in a way that allows it to use layout features obtained from an OCR system, without the need to re-learn language semantics from scratch. We only augment the input of the model with the coordinates of token bounding boxes, avoiding, in this way, the use of raw images. This leads to a layout-aware language model which can then be fine-tuned on downstream tasks. The model is evaluated on an end-to-end information extraction task using four publicly available datasets: Kleister NDA, Kleister Charity, SROIE and CORD. We show that our model achieves superior performance on datasets consisting of visually rich documents, while also outperforming the baseline RoBERTa on documents with flat layout (NDA \(F_{1}\) increase from 78.50 to 80.42). Our solution ranked first on the public leaderboard for the Key Information Extraction from the SROIE dataset, improving the SOTA \(F_{1}\)-score from 97.81 to 98.17.

{\L}ukasz Garncarek, Rafa{\l} Powalski, Tomasz Stanis{\l}awek, Bartosz Topolski, Piotr Halama, Micha{\l} Turski, Filip Grali\'nski• 2020

Related benchmarks

TaskDatasetResultRank
Information ExtractionCORD (test)
F1 Score96.06
133
Information ExtractionSROIE (test)
F1 Score98.17
58
Semantic Entity RecognitionCORD
F1 Score94.41
55
Document UnderstandingDUE Benchmark--
24
Entity extractionKleister-NDA (val)
Entity-level F175.7
21
Information ExtractionCORD
F1 Score94.41
18
Information ExtractionSROIE
F1 Score98.17
16
Document UnderstandingDUE-Benchmark (test)--
10
Information ExtractionKleister NDA
F1 Score80.42
6
Information ExtractionKleister Charity
F1 Score81.34
6
Showing 10 of 11 rows

Other info

Code

Follow for update