Contact - Document UnderstanDing of Everything 😎

For any questions about this challenge, please contact us at

Jordy Van Landeghem is a Ph.D. student at KU Leuven and the lead AI researcher at, a European SaaS company that offers an intelligent document processing solution. His industrial Ph.D. project entitled Intelligent Automation for AI-driven Document Understanding focuses on the fundamentals of probabilistic deep learning to obtain more reliable ML systems.

Rubèn Tito is a Ph.D. student at the Computer Vision Center and Universitat Autònoma de Barcelona. His research is focused on text-based visual question answering, having been part of the organization of the Scene Text Visual Question Answering Challenge and the two Document Visual Question Answering Challenges.

Tomasz Stanisławek is working as a Senior Research Scientist at Snowflake on problems related to the Document Understanding domain. He is a co-author of the LAMBERT model, Kleister dataset, and DUE benchmark.

Łukasz Borchmann is a Document Understanding professional from Snowflake, where he develops new architectures and conducts experiments to tackle previously unexplored business problems end-to-end. He is a co-creator of the TILT model and DUE Benchmark. His Ph.D. focused, among others, on the problem of Key Information Extraction.

Dawid Jurkiewicz is a Ph.D. student at Adam Mickiewicz University and Senior Research Scientist at Snowflake. His research focuses on developing novel models and datasets for Intelligent Document Processing.

Michał Pietruszka is performing research that bridges the gap between theoretical advances in NLP and real-world, large-scale applications at Snowflake, with a focus on improving model understanding of structured data, reducing computational costs of deep learning models, and improving model capabilities to populate databases.

Rafał Powalski is a Document Understanding Researcher from Instabase, where he develops new architectures and solutions to solve complex end-to-end business problems. He is a co-creator of the LAMBERT and TILT models.

Pawel Joziak is a Research Scientist in Snowflake Poland and an Assistant Professor at Warsaw University of Technology. He has experience in delivering Document Understanding ML/AI-based solutions for customers. Scientifically, his main area of interest is related to the mathematical foundations of Deep Learning.

Sanket Biswas is a Ph.D. student at the Computer Vision Center and Universitat Autònoma de Barcelona. Although in the early stage of his Ph.D., he has contributed to the field of document understanding with contributions in conferences like ICDAR 2021, ICPR 2022, and ECCV 2022.

Mickaël Coustaty is an Associate Professor in L3i laboratory from La Rochelle Université, France. His research focuses on multimodal analysis of documents for different tasks (document classification, information extraction). He co-organized 5 competitions in the previous ICDAR (from 2015 to 2019) and was co-chair of 5 satellite workshops of ICDAR from 2015 to 2021.



We thank Thierry Deruyttere for allowing us to take over his account to set up DUDE data annotations. 

Important Dates

Registration open: Christmas 2022

Competition Q&A period: 20 December - 10 January

Sample dataset available: 6 January 2023

Training/validation dataset available: 30 January 2023

Test set submissions open (Task 1 & evaluation phase2): 9 March 2023

General submission deadline: 20 April 2023

Method description submissions deadline: 20 April 2023 

Notification to authors: 1 May 2023

All dates are 23:59 AoE and subject to change.


Note on the registration for the DUDE challenge:

There is no need to register explicitly for the DUDE challenge. As long as you are registered to the RRC portal you will be able to submit your results when the submission is open.

Any questions, contact the DUDEs at