From SVM models to SOTA: how to be innovative in NLP domain

First time onsite! Join us at the 18th meeting of the Polish NLP Group and networking beer after the talk!

Abstract:

During the presentation, I will introduce the Document Understanding domain. I will start from very simple models that we can use and later on the move to the state-of-the-art LAMBERT and TILT methods developed in Applica. I will also discuss further directions of development of the field.

Bio:

Tomasz is currently working as an Engineering Manager at Snowflake, where he takes care of the NLP research group, which builds best-in-class solutions for extracting information from unstructured data. In 2022 he received a Ph.D. degree in machine learning from the Warsaw University of Technology. His research interests include Natural Language Processing and machine learning applications.

Talk will be held in Polish. Registration is mandatory.