Rethinking the Evaluation of Alignment Methods: Insights into Diversity, Generalisation, and SafetyDP Janiak, JA Moska, D Motyka, PT Walkowiak, A Janz, K Seweryn, …EACL 2026DOIPDFLink2026
PLLuM-Align: Polish Preference Dataset for Large Language Model AlignmentK Seweryn, A Kołos, A Karlińska, K Lorenc, K Dziewulska, M Chrabąszcz, …, D Motyka, …, A JanzThe 2025 Conference on Empirical Methods in Natural Language Processing, EMNLP 2025, November 4-9 23879–23908DOIPDFLink2025
