Rethinking the Evaluation of Alignment Methods: Insights into Diversity, Generalisation, and Safety
Denis P. Janiak, Julia A Moska, Dawid Motyka, Paweł T Walkowiak, Arkadiusz Janz, Karolina Seweryn, Bartosz Żuk
EACL 2026
2026
Profile
Julia Moska
PhD Student
Education
- 2017
B.Eng. in Computer Science
Wroclaw University of Science and Technology
- 2019
MsC in Computer Science
Wroclaw University of Science and Technology
Publications6
Think like a researcher: A dataset for scientific ideation with large language models
Julia A. Moska, Maciej Piasecki, Arkadiusz Janz
Springer Nature Switzerland AG
2026
PLLuM-Align: Polish Preference Dataset for Large Language Model Alignment
Julia A Moska, Dawid Motyka, Paweł T Walkowiak, Arkadiusz Janz, Karolina Seweryn, Anna Kołos, Agnieszka Karlińska, Katarzyna Lorenc, Katarzyna Dziewulska, Maciej Chrabąszcz, Aleksandra Krasnodębska, Paula Betscher, Zofia Cieślińska, Katarzyna Kowol, Bartosz Żuk
Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
2025
OBSR: Open Benchmark for spatial representations
Julia A. Moska, Oleksii Furman, Piotr Ł Gramacki, Piotr Szymański, Kacper Kozaczko, Szymon Leszkiewicz, Jakub Polczyk
ACM
2025
Is heterogeneous model soup tasty? A multidimensional evaluation of diverse model soups in language model alignment
Dawid Motyka, Paweł T Walkowiak, Julia A Moska, Arkadiusz Janz, Bartosz Żuk, Karolina Seweryn
Springer Nature Switzerland AG
2025
Refining natural language inferences using cross-document structure theory
Arkadiusz Janz, Dominik Kurowski, Joanna Baran, Julia A Moska, Tomasz Bernaś, Marcin Ł Oleksy
Springer
2024