Pavel Pecina

I am an associate professor working in the area of computational linguistics, natural language processing and other related areas at the Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic. My research interests include machine translation, information retrieval and extraction, multimodal data interpretation, and optical music recognition.

We are co-organizing IWSLT Dialectal and Low-resource track this year!
I am seeking prospective students to pursue PhD in Natural Language Processing and related areas.

Filip Danielsson, Pavel Pecina (2024). Towards a Dataset for Estimation of Keyboard Fingerings. Accepted to Proceedings of the 15th International Workshop on Machine Learning and Music - MML 2024, Vilnius, Lithuania.
Mayer Jiří, Straka Milan, Jan Hajič jr., Pecina Pavel (2024). Practical End-to-End Optical Music Recognition for Pianoform Music. In Document Analysis and Recognition - ICDAR 2024., pp. 55–73, Lecture Notes in Computer Science, vol 14809. Springer, Cham (bib).
Vojtěch Lanz, Pavel Pecina (2024). Paragraph Retrieval for Enhanced Question Answering in Clinical Documents. In Proceedings of the 23rd Workshop on Biomedical Natural Language Processing, pp. 580–590, Bangkok, Thailand (bib).
Christopher Brückner, Leixin Zhang, Pavel Pecina (2024). Similarity-Based Cluster Merging for Semantic Change Modeling. In Proceedings of the 5th Workshop on Computational Approaches to Historical Language Change, pp. 23–28, Bangkok, Thailand (bib).
Ibrahim Said Ahmad et al. (2024). Findings of the IWSLT 2024 Evaluation Campaign. In Proceedings of the 21st International Conference on Spoken Language Translation (IWSLT 2024), pp. 1–11, Bangkok, Thailand (bib).
Mateusz Krubiński, Pavel Pecina (2024). Towards Unified Uni- and Multi-modal News Headline Generation. In Findings of the Association for Computational Linguistics: EACL 2024, pp. 437-450. St. Julian's, Malta (bib).

GI-Insight: New methods for stomach examination using artificial intelligence: Utilization of deep learning for assisted gastroscopy, LUABA24136.
RES-Q+: Comprehensive solutions of healthcare improvement based on the global Registry of Stroke Care Quality, HORIZON-HLTH-2021-TOOL-06/101057603.
MEMORISE: Virtualisation and Multimodal Exploration of Heritage on Nazi Persecution, HORIZON-CL2-2021-HERITAGE-01/101061016.

Klára Tauchmanová - Named entity recognition in historical texts (MSc)
Igbal Huseynov - Topical segmentation of spoken narratives (MSc)
Nair Amrita Harikrishnan - Unsupervised open information extraction with large language models (MSc, co-supervised)
Karunakaran Goutham Venkatesh - Automatic relation extraction from clinical documents (MSc, co-supervised)
Vojtěch Lanz - Information extraction from clinical documents (PhD)
Christopher Brückner - Information extraction from historical documents (PhD)
Filip Danielsson - Towards a dataset for estimation of keyboard fingerings (Bc, external)
Jiří Mayer - Optical music recognition (PhD)
Mateusz Krubiński - Multimodal summarization (PhD)
Michal Auersperger - Neural representations (PhD)

Institute of Formal and Applied Linguistics