Crina Madalina Tudor PhD Student
Contact
Name and title: Crina Madalina TudorPhD Student
ORCID0009-0009-9573-6945 Länk till annan webbplats.
Workplace: Department of Linguistics Länk till annan webbplats.
Visiting address Room C246Universitetsvägen 10 C, plan 2-3
Postal address Institutionen för lingvistik106 91 Stockholm
About me
I am a PhD student in computational linguistics with experience applying NLP and machine-learning methods to historical and multilingual text. My thesis develops and evaluates methods for named entity recognition (NER) in historical archival text, with the aim of making entity annotation accessible and reliable for historians, librarians and archivists. My research includes fine-tuning and domain-adaptive pre-training of language models and the evaluation of prompt-based LLM approaches. Alongside research and data analysis, I work with programme coordination and academic mentoring, and I am interested applied AI research and research coordination.
Master's thesis supervision (Master's Programme in AI and Language, spring 2026).
Guest lecture on NER, annotation and ethical considerations in archival text processing (Corpus Linguistics, autumn 2026).
- Crina Tudor, Beata Megyesi, and Robert Östling. 2025. Prompting the Past: Exploring Zero-Shot Learning for Named Entity Recognition in Historical Texts Using Prompt-Answering LLMs. In Proceedings of the 9th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2025), pages 216–226, Albuquerque, New Mexico. Association for Computational Linguistics.
- Crina Tudor and Eva Pettersson. 2024. People and Places of the Past - Named Entity Recognition in Swedish Labour Movement Documents from Historical Sources. In Proceedings of the 8th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2024), pages 185–195, St. Julians, Malta. Association for Computational Linguistics.
For a full list of my publications, please go to: Google Scholar .