Views
No views yet
FacebookAI/xlm-roberta-base as its base encoder and is trained on UD_Latin-ITTB (UD v2.17).| Metric | Tokens | Sentences | Words | UPOS | XPOS | UFeats | AllTags | Lemmas |
|---|---|---|---|---|---|---|---|---|
| Full-text (F1) | 99.99 | 95.32 | 99.99 | 99.26 | 97.12 | 97.86 | 96.32 | 99.42 |
| Aligned accuracy | 0.00 | 0.00 | 0.00 | 99.26 | 97.12 | 97.86 | 96.32 | 99.42 |
| Metric | UAS | LAS | CLAS | MLAS | BLEX |
|---|---|---|---|---|---|
| Full-text (F1) | 92.10 | 90.50 | 88.88 | 85.40 | 88.49 |
| Aligned accuracy | 92.10 | 90.50 | 88.80 | 85.32 | 88.40 |
pip install combo-nlpcombo-nlp, so
no extra install step is needed.1from combo import COMBO
2
3# Load a pre-trained model with the corresponding combo-seg segmenter
4nlp = COMBO("Latin")
5
6# Parse raw text (handles sentence splitting + tokenization)
7result = nlp("Vulpes brunnea velox super canem pigrum salit.")
8
9# Inspect results
10for sentence in result:
11 for token in sentence:
12 print(f"{token.form:<15} {token.lemma:<15} {token.upos:<8} head={token.head} {token.deprel}")LICENSE.txt file in the treebank repository:1@software{combo_nlp_2026,
2 author = {Ulewicz, Michał and Jabłońska, Maja and Klimaszewski, Mateusz and Przybyła, Piotr and Pszenny, Łukasz and Rybak, Piotr and Wiącek, Martyna and Wróblewska, Alina},
3 title = {{COMBO-NLP} Models Trained on {UD} v2.17},
4 year = {2026},
5 publisher = {Zenodo},
6 doi = {10.5281/zenodo.19650523},
7 url = {https://doi.org/10.5281/zenodo.19650523}
8}1@article{morpharmo,
2 author = {Gamba, Federica and Zeman, Daniel},
3 title = {Latin Morphology through the Centuries: Ensuring Consistency for Better Language Processing},
4 journal = {Proceedings of the Ancient Language Processing Workshop associated with the 14th International Conference on Recent Advances in Natural Language Processing RANLP 2023},
5 year = {2023},
6 address = {Varna, Bulgaria},
7url = {https://ufal.mff.cuni.cz/biblio/attachments/2023-gamba-p3787387064232511302.pdf}
8}
9
10@article{harmo,
11 author = {Gamba, Federica and Zeman, Daniel},
12 title = {Universalising Latin Universal Dependencies: a harmonisation of Latin treebanks in UD},
13 journal = {Proceedings of the Sixth Workshop on Universal Dependencies (UDW, GURT/SyntaxFest 2023)}},
14 year = {2023},
15 address = {Washington, DC, USA},
16 url = {https://aclanthology.org/2023.udw-1.2/}
17}
18
19@article{lait-ud,
20 author = {Cecchini, Flavio Massimiliano and Passarotti, Marco and Marongiu, Paola and Zeman, Daniel},
21 title = {Challenges in Converting the \emph{Index Thomisticus} treebank into Universal Dependencies},
22 journal = {Proceedings of the Universal Dependencies Workshop 2018 (UDW 2018)},
23 year = {2018},
24 address = {Brussels, Belgium}
25}
26
27@article{lait,
28 author = {Passarotti, Marco and Dell’Orletta, Felice},
29 title = {Improvements in parsing the index thomisticus treebank. Revision, combination and a feature model for medieval Latin},
30 journal = {Training},
31 volume = {2},
32 pages = {61--024},
33 year = {2010}
34}