Piotr Przybyła

Piotr Przybyła is a tenure-track Assistant Professor at Universitat Pompeu Fabra in Barcelona, Spain, where he researches in the TALN (Natural Language Processing) Research Group. He maintains a significant affiliation with the Linguistic Engineering Group at the Institute of Computer Science, Polish Academy of Sciences (ICS PAS) in Warsaw, Poland, where he completed his PhD in Computer Science. Previously, he worked as a research fellow at the National Centre for Text Mining (NaCTeM) at the University of Manchester.

Przybyła's research focuses primarily on Natural Language Processing with particular emphasis on misinformation detection, adversarial attacks on text classifiers, text simplification, and Polish language processing. His work bridges theoretical NLP with practical applications for credibility assessment and language understanding. He has developed innovative approaches for testing the robustness of text classifiers against adversarial examples and has made significant contributions to Polish language resources and processing tools.

His recent publications demonstrate a strong trajectory in examining the robustness of NLP systems, particularly in the context of misinformation detection and credibility assessment. His work spans from foundational research on language model behavior to practical applications in Polish language processing and text simplification. The ERINIA project, funded by a prestigious Marie Skłodowska-Curie Postdoctoral Fellowship, represents a significant contribution to understanding how misinformation detection systems can be made more robust against adversarial attacks.

• Marie Skłodowska-Curie Postdoctoral Fellowship for the ERINIA project
• Computing grant of 10,000 hours on the Athena supercomputer for accelerating work in the ERINIA project

Przybyła actively contributes to the NLP community through conference organization, shared tasks (such as coordinating the InCrediblAE shared task for CheckThat! 2024), and developing open-source tools like Plainifier for multi-word lexical simplification. His work demonstrates a commitment to both advancing NLP research methodology and addressing practical challenges in misinformation detection and language understanding across multiple languages, with special attention to Polish language processing.