ÚFAL MFF UK

Institute of Formal and Applied Linguistics (ÚFAL) is a departmet at the Computer Science School, Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic. The Institute is a primarily research department working on many topics in the areas of Language technologies and Computational linguistics, and on many research projects both nationally and internationally. In addition, the Institute carries a comprehensive teaching program both for the Master's degree (Mgr., or MSc.) as well as for a doctorate (Ph.D.) in Computational Linguistics.

AI v hudbě [ AI v kontextu #24 ]

AI v medicíně [ AI v kontextu #23 ]

Representational Capabilities of Transformers and Recurrent Architectures [ LingMon #225 ]

Behavioral studies of LLMs [ LingMon #224 ]

Workshop projektu PONK

Interpretive-structural coding and semantic analysis of Czech and Polish discourses [ LingMon #223 ]

LLMs and Factuality [ FJelinekSeries #59 ]

Psycholinguistic corpora, eye movements and meaning representations [ LingMon #222 ]

AI v kontextu filosofie jazyka [ AI v kontextu #22 ]

AI and democracy [ AI in context #21 ]

Investigating the valency of prefixed and particle verbs using SynSemClass [ LingMon #221 ]

The Most (Subjectively) Interesting Bottlenecks of LLM Development [ LingMon #220 ]

NameTag 3: A Tool and a Service for Multilingual/Multitagset NER [ACL 2025]

AI v kontextu: AI a létající drony [ AI v kontextu #20 ]

AI v kontextu: Emboddied AI. Je doopravdy vtělená? [ AI v kontextu #19 ]

Towards Natural Behaviour of Dialogue Systems with Explicit Dialogue Control [ LingMon #219 ]

Analysis of Student Interpreting: Annotation, Alignment, and Automation [ LingMon #218 ]

AI v kontextu poezie [ AI v kontextu #18 ]

AI v kontextu sportu [ AI v kontextu #17 ]

AI v kontextu zemědělství [ AI v kontextu #16 ]

MSDRT: A formal framework for a semantics and pragmatics of natural language [ FJelinekSeries #58 ]

EdUKate: Multilingual Learning Materials for Ukrainian Children in Czech Schools [ LingMon #217 ]

HiČKoK: History of Czech in Corpus Continuum [ LingMon #216 ]

The Unnatural Language of Poetic Meters, or Why I Am Scared of Systematicity [ LingMon #215 ]

2024 demo of AI Interpreting service

Lexically Grounded Subword Segmentation [EMNLP 2024]

Ten Years of Universal Dependencies [ FJelinekSeries #57 ]

Reading and Predicting Music History [ LingMon #214 ]

Tokenization for NLP is Hard [ LingMon #213 ]

Breaking the Curse of Multilinguality in Language Models [ LingMon #212 ]