Research

Publications

KAŠE, Vojtěch; LANG, Sarah; PAVLAS, Petr (2025). Embedded in the Labyrinth: Investigating Latin Word Senses through Transformer-Based Contextual Embeddings and Attention. In: Computational Humanities Research 2025, ed. by Taylor Arnold, Margherita Fantoli, and Ruben Ros. Vol. 3. Anthology of Computers and the Humanities, pp. 498–512, https://doi.org/10.63744/FuaAvdPMdtwW.

KAŠE, Vojtěch; ŠVADLENKOVÁ, Jana; TVRZ, Jan; HEDESAN, Georgiana; PAVLAS, Petr (2025). iWEEMS: Interactive Word Embeddings for Early Modern Science. Journal of Open Humanities Data, 11(53) [collection Data-Driven History of Ideas], pp. 1-9, https://doi.org/10.5334/johd.379.

PAVLAS, Petr (2023). Gott und Frieden. Biblische Sprache und kognitive Metaphern des Angelus Pacis. Comenius-Jahrbuch 31, pp. 87–110. https://doi.org/10.5771/9783985721481-87. Available for reading and downloading here.

PAVLAS, Petr (2024). Word as definition. A key principle of the Comenian project for universal language: its sources and contexts. Language & History, 67(2), 75–101. https://doi.org/10.1080/17597536.2024.2307681. The accepted version is made available for reading and downloading here after 24 November 2025 when the publisher’s embargo expires.

PAVLAS, Petr; ŘEZNÍKOVÁ, Lenka; STORCHOVÁ, Lucie (eds.) (2025). Cognitive Metaphors and Encyclopaedic Knowledge: Exploring Semantic Transformations in Early Modernity. Filosofický časopis special issue, 1/2025, 178 pp. Available for reading and downloading here.

STORCHOVÁ, Lucie (2025). Pulcherrimus ordo naturae in Crisis or How Bohemian Latin Poets Coped with Changes in Wittenberg Cosmology after 1574. Central European Cultures, 5(1), pp. 50–77. https://doi.org/10.47075/CEC.2025-1.04.

ŽEMLA, Martin (2025). Perly, balzám a signatury: Poznámka k Paracelsovu Herbariu [Pearls, Balm, and Signatures: A Note on Paracelsus’ Herbarius]. In: Paracelsus, Herbarius. O léčivých účincích čemeřice, rdesna, soli, pupavy, korálu a magnetu [Herbarius. On the Medicinal Effects of Hellebore, Knotweed, Salt, Dandelion, Coral, and Magnet], transl. Petr Babka, Praha: Malvern, pp. 87-106. Available for reading and downloading here.

Coding and transcribing

We follow the FAIR data principles and make all analyses code and data available for reuse by other scholars. All our computational analyses are available via GitHub.

TOME published datasets

TOME repositories

  • noscemus_ETF (data processing and analysis of textual data from the Noscemus database)
  • interactive-embeddings (development of a tool allowing interactive visual exploration of word embedding data based on corpora of early modern scientific literature in Latin)
  • latin-contextual-embeddings (explorations of Latin BERT model)
  • corpus-corporum (data processing and analysis of textual data from the Corpus Corporum database)