Document Visual Question Answering with CIVQA: Czech Invoice Visual Question Answering Dataset

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F23%3A00132396" target="_blank" >RIV/00216224:14330/23:00132396 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.fi.muni.cz/usr/sojka/papers/scavnicka-stefanik-sojka-raslan-2023.pdf" target="_blank" >https://www.fi.muni.cz/usr/sojka/papers/scavnicka-stefanik-sojka-raslan-2023.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Document Visual Question Answering with CIVQA: Czech Invoice Visual Question Answering Dataset

Popis výsledku v původním jazyce

Applications of document processing become increasingly popular across multiple industries, resulting in a growing amount of research on the applications of artificial intelligence in document processing (Document AI). This paper focuses on a subtask of Document AI, Document Visual Question Answering (DVQA), recently getting well-deserved attention thanks to its universality. However, the limited availability of data sources for languages outside English restrains the applicability of DVQA in non-English languages. <p> For this reason, we created the CIVQA (Czech Inovice Visual Question Answering) dataset covering 15 entities of financial documents, consisting of more than 6,000 invoices in the Czech language. </p> <p> We used the CIVQA dataset to create the first-of-its-kind DVQA models specifically tailored for applications to Czech documents. Striving to create DVQA models able to generalize, we specifically evaluate our models on the entities not covered in the training mix and find that multilingual LayoutLM models are able to respond to questions about previously unseen entities substantially more accurately than other models. </p> <p> The CIVQA dataset and experiment observations offer new opportunities for Document AI in the Czech Republic, with potential applications in research and commercial fields.</p>

Název v anglickém jazyce

Document Visual Question Answering with CIVQA: Czech Invoice Visual Question Answering Dataset

Popis výsledku anglicky

Applications of document processing become increasingly popular across multiple industries, resulting in a growing amount of research on the applications of artificial intelligence in document processing (Document AI). This paper focuses on a subtask of Document AI, Document Visual Question Answering (DVQA), recently getting well-deserved attention thanks to its universality. However, the limited availability of data sources for languages outside English restrains the applicability of DVQA in non-English languages. <p> For this reason, we created the CIVQA (Czech Inovice Visual Question Answering) dataset covering 15 entities of financial documents, consisting of more than 6,000 invoices in the Czech language. </p> <p> We used the CIVQA dataset to create the first-of-its-kind DVQA models specifically tailored for applications to Czech documents. Striving to create DVQA models able to generalize, we specifically evaluate our models on the entities not covered in the training mix and find that multilingual LayoutLM models are able to respond to questions about previously unseen entities substantially more accurately than other models. </p> <p> The CIVQA dataset and experiment observations offer new opportunities for Document AI in the Czech Republic, with potential applications in research and commercial fields.</p>

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/EG21_374%2F0026711" target="_blank" >EG21_374/0026711: Inteligentní back office</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Recent Advances in Slavonic Natural Language Processing (RASLAN 2023)

ISBN

9788026317937

ISSN

2336-4289

e-ISSN

—

Počet stran výsledku

12

Strana od-do

23-34

Název nakladatele

Tribun EU

Místo vydání

Brno

Místo konání akce

Kouty nad Desnou

Datum konání akce

8. 12. 2023

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

—