Mova0.5 pre-release⌕
☰
home / docs / papers / English / Cross-Lingual and Cross-Domain Transfer Learning for POS Tagging in Historical Germanic Low-Resource Languages

Cross-Lingual and Cross-Domain Transfer Learning for POS Tagging in Historical Germanic Low-Resource Languages

What it is about

The paper tags parts of speech in four historical Germanic languages (Old English, Old Norse, Old High German, Old Saxon), each in a poetry and a prose corpus. The original tag sets are mapped to UPOS. The authors train monolingual baselines, then multilingual models (mBERT in MaChAmp) on poetry only, prose only, or both genres, at two dataset sizes; the smaller size is 444 training sentences, the size of the smallest corpus. Poetry-only multilingual models beat the monolingual baselines on poetry test sets, with gains of up to 6 F1 points in the reduced setting. Models trained on all data are worse than the baselines on prose test sets. Old Saxon poetry-trained models reach 0.62 accuracy on prose (original sizes, Table 4), against 0.891 for the prose-trained model. The authors conclude that genre differences in structure limit cross-lingual transfer.

Ideas

For Mova

The period and texts are in 1. The period, the register and the texts Mova reads, the sample texts in 14-sample-texts.md, and the archaic forms in 09. Early Modern and archaic forms. The code is en/src/emode.rs (closed table of Early Modern forms) with the tagger en/src/ptag.rs. The paper's tagger and model are not for Mova: the design is. Old Norse, Old Saxon and Old High German are not part of Mova; the paper matters for English only as a case of pooled versus genre-specific training. Measure: the two-genre split on EModE lessons, with the negative control above.

Topics: Part-of-speech tagging

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.loreslm-1.47

@inproceedings{miani2026cross,
  title = {{Cross-Lingual and Cross-Domain Transfer Learning for POS Tagging in Historical Germanic Low-Resource Languages}},
  author = {Miani, Irene and Stymne, Sara and Darwin, Gregory R.},
  booktitle = {Proceedings of the Second Workshop on Language Models for Low-Resource Languages (LoResLM 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.loreslm-1.47},
  url = {https://doi.org/10.18653/v1/2026.loreslm-1.47}
}

References

21 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Momen Abo Mokh, Houda Bouamor, and Nizar Habash. 2022. Improving POS tagging for Arabic dialects on out-of-domain texts. In Proceedings of the Seventh Arabic Natural Language Processing Workshop (WANLP), pages 205–215. Association for Computational Linguistics. find
  2. Hirona Jacqueline Arai. 2021. Optimizing an automatic part of speech tagger for poetry text using data augmentation. Undergraduate thesis, Middlebury College, Computer Science Department. find
  3. Nomi Baruah and Pritom Jyoti Goutom. 2025. A comparative analysis of deep learning and machine learning for pos tagging. Expert Systems with Applications, 288:128026. find
  4. Kristin Bech and Kristine Eide. 2014. The iswoc corpus. Department of Literature, Area Studies and European Languages, University of Oslo. find
  5. Shabnam Behzad and Amir Zeldes. 2020. A cross-genre ensemble approach to robust reddit part of speech tagging. arXiv e-prints. find
  6. Marie-Catherine de Marneffe, Christopher D. Manning, Joakim Nivre, and Daniel Zeman. 2021. Universal dependencies. Computational Linguistics, 47(2):255– 308. find our card
  7. Wietse de Vries, Martijn Wieling, and Malvina Nissim. 2022. Make the best of cross-lingual transfer: Evidence from POS tagging with over 100 languages. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7676–7685, Dublin, Ireland. Association for Computational Linguistics. find
  8. Brendan Ferreri Hanberry. 2015. Application of a pos tagger to a novel chronological division of early modern german text. Master’s thesis, University of North Carolina at Chapel Hill. find
  9. Kia Kirstein Hansen and Rob van der Goot. 2023. Cross-domain evaluation of pos taggers: From wall street journal to fandom wiki. arXiv e-prints. find
  10. Odd Einar Haugen and Fartein Th. Øverland. 2014. Guidelines for morphological and syntactic annotation of old norwegian texts. Bergen Language and Linguistics Studies (BeLLS), 4(2). find
  11. Kyle P. Johnson, Patrick J. Burns, John Stewart, Todd Cook, Clément Besnier, and William J. B. Mattingly. 2021. The Classical Language Toolkit: An NLP framework for pre-modern languages. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: System Demonstrations, pages 20–29, Online. Association for Computational Linguistics. find
  12. Mariya Koleva, Melissa Farasyn, Bart Desmet, Anne Breitbarth, and Veronique Hoste. 2017. An automatic part-of-speech tagger for middle low german. International Journal of Corpus Linguistics, 22(1):108– 141. find
  13. Irene Miani, Sara Stymne, and Gregory R. Darwin. 2025. Cross-genre learning for Old English poetry POS tagging. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop), pages 708–724, Vienna, Austria. Association for Computational Linguistics. find our card
  14. Ercong Nie, Helmut Schmid, and Hinrich Schütze. 2023. Cross-lingual constituency parsing for middle high german: A delexicalized approach. arXiv preprint arXiv:2308.04645. ↗
  15. Enora Rice, Ali Marashian, Hannah Haynie, Katharina von der Wense, and Alexis Palmer. 2025. Untangling the influence of typology, data and model architecture on ranking transfer languages for cross-lingual pos tagging. arXiv e-prints. find
  16. Victoria Rosén, Koenraad De Smedt, Paul Meurer, and Helge Dyvik. 2012. An open infrastructure for advanced treebanking. In META-RESEARCH Workshop on Advanced Treebanking at LREC2012, pages 22–29, Istanbul, Turkey. find
  17. Silke Scheible, Richard J. Whitt, Martin Durrell, and Paul Bennett. 2011. Evaluating an ‘off-the-shelf’ pos-tagger on early modern german text. In Proceedings of the 5th ACL-HLT Workshop on Language Technology for Cultural Heritage, Social Sciences, and Humanities, pages 19–23, Portland, OR, USA. Association for Computational Linguistics. find
  18. Matthias Schöffel, Esteban Garces Arias, Marinus Wiedner, Paula Ruppert, Meimingwei Li, Christian Heumann, and Matthias Aßenmacher. 2025a. Unveiling factors for enhanced pos tagging: A study of low-resource medieval romance languages. Preprint, arXiv:2506.17715. ↗
  19. Matthias Schöffel, Marinus Wiedner, Esteban Garces Arias, Paula Ruppert, Christian Heumann, and Matthias Aßenmacher. 2025b. Modern models, medieval texts: A pos tagging study of old occitan. arXiv preprint arXiv:2503.07827. ↗
  20. Rob van der Goot, Ahmet Üstün, Alan Ramponi, Ibrahim Sharaf, and Barbara Plank. 2021. Massive choice, ample tasks (MaChAmp): A toolkit for multi-task learning in NLP. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations, pages 176–197, Online. Association for Computational Linguistics. ↗
  21. Lars Erik Zeige, Gohar Schnelle, Martin Klotz, Karin Donhauser, and and Lühr Rosemarie Gippert, Jost. 2025. Deutsch Diachron Digital - Referenzkorpus Altdeutsch (Version 1.2). Humboldt-Universität zu Berlin. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/miani-2026-cross-lingual-cross-domain-transfer/