Karolina Seweryn
Research interests: Natural language processing (NLP), computer vision, multimodal models, action spotting
Karolina Seweryn is a PhD student at the Information and Biomedical Technologies Doctoral School of the Polish Academy of Sciences. Her doctoral research work conducted at NASK involves the analysis of recordings of football matches, especially combining different modalities in training machine learning models. In addition, she is developing a tool to detect domains used for phishing during their registration.
As a member of a research grant team at the Warsaw University of Technology, she worked on automatic act analysis, NLI inference in Polish and model training on small sets. She also taught project classes on Natural Language Processing.
Selected Publications
Articles
Karolina Seweryn, Anna Kołos, Agnieszka Karlińska, Katarzyna Lorenc, Katarzyna Dziewulska, Maciej Chrabaszcz, Aleksandra Krasnodebska, Paula Betscher, Zofia Cieślińska, Katarzyna Kowol, Julia Moska, Dawid Motyka, Paweł Walkowiak, Bartosz Żuk, Arkadiusz Janz, "PLLuM-Align: Polish Preference Dataset for Large Language Model Alignment", Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, , 2025, 23890–23919.
Aleksandra Krasnodębska, Karolina Seweryn, Szymon Łukasik, Wojciech Kusa, "PL-Guard: Benchmarking Language Model Safety for Polish", Proceedings of the 10th Workshop on Slavic Natural Language Processing, Slavic NLP, 2025, 25–37.
Mateusz Kowalczyk, Karolina Seweryn, Joanna Kołodziej, Mateusz Krzysztoń, "Adversarial Robustness of Multimodal Machine Learning Models", Proceedings of the 39th ECMS International Conference on Modelling and Simulation, ECMS 2025, , 248-254.
Dawid Motyka, Paweł Walkowiak, Julia Moska, Bartosz Żuk, Karolina Seweryn, Arkadiusz Janz, "Is Heterogeneous Model Soup Tasty? A Multidimensional Evaluation of Diverse Model Soups in Language Model Alignment", Computational Science – ICCS 2025, Proceedings, Part IV, 2025, 154-162.
Karolina Seweryn, Gabriel Chęć, Szymon Łukasik, Anna Wróblewska, "Improving Object Detection Quality in Football Through Super-Resolution Techniques", Computational Science – ICCS 2025, Proceedings, Part I, 2025, 151-163.
Daniel Ziembicki, Karolina Seweryn, Anna Wróblewska, "Polish Natural Language Inference and Factivity – an Expert-based Dataset and Benchmarks", Natural Language Engineering, 30(2), 2023, 385–416.
Anna Wróblewska, Bartosz Pieliński, Karolina Seweryn, Sylwia Sysko-Romańczuk, Karol Saputa, Aleksandra Wichrowska, Hanna Schreiber, "Automating the Analysis of Institutional Design in International Agreements", Computational Science – ICCS 2023, Lecture Notes in Computer Science, vol. 10475, red. Jiří Mikyška, Clélia de Mulatier, Maciej Paszynski, Valeria V. Krzhizhanovskaya, Jack J. Dongarra, Peter M.A. Sloot, 2023, Springer, Cham.