Towards General Document Understanding through Question Answering

Investor logo

Warning

This publication doesn't include Faculty of Arts. It includes Faculty of Informatics. Official publication website can be found on muni.cz.
Authors

ŠČAVNICKÁ Šárka ŠTEFÁNIK Michal KADLČÍK Marek GELETKA Martin SOJKA Petr

Year of publication 2022
Type Article in Proceedings
Conference Recent Advances in Slavonic Natural Language Processing (RASLAN 2022)
MU Faculty or unit

Faculty of Informatics

Citation
Web fulltext PDF
Keywords Question Answering; Visual Question Answering; Document Visual Question Answering
Description Document Visual Question Answering is a relatively new extension of Visual Question Answering. The aim is to understand the documents and to be able to obtain information that corresponds to the question that was asked. This proposition aims to approach the problem of the lack of datasets and a model for Slavic languages. Therefore we would like to create a model and dataset for Document VQA suitable for the non-English language. This paper overviews the field of Question Answering and also describes the first Czech Document VQA dataset and model.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.