Full-text (F1): raw text is segmented by combo-seg, then parsed and compared against gold — measures end-to-end pipeline performance including segmentation quality.
Aligned accuracy: accuracy on correctly segmented (aligned) tokens — measures parsing quality on tokens that were correctly identified by the segmenter.

Morphosyntactic Tagging

Metric	Tokens	Sentences	Words	UPOS	XPOS	UFeats	AllTags	Lemmas
Full-text (F1)	98.79	100.00	80.69	69.73	65.13	65.49	57.95	74.90
Aligned accuracy	0.00	0.00	0.00	86.42	80.72	81.17	71.81	92.83

Dependency Parsing

Metric	UAS	LAS	CLAS	MLAS	BLEX
Full-text (F1)	44.35	38.98	38.79	28.61	35.95
Aligned accuracy	54.96	48.30	46.55	34.33	43.14

Usage

Install the library from PyPI (assuming you have a virtual environment created):

pip install combo-nlp

The combo-seg segmenter (used to split and tokenise raw text) is installed automatically as a dependency of combo-nlp, so no extra install step is needed.

python

from combo import COMBO

# Load a pre-trained model with the corresponding combo-seg segmenter
nlp = COMBO("Tamil")

# Parse raw text (handles sentence splitting + tokenization)
result = nlp("வேகமான பழுப்பு நரி சோம்பேறி நாயின் மீது குதிக்கிறது.")

# Inspect results
for sentence in result:
    for token in sentence:
        print(f"{token.form:<15} {token.lemma:<15} {token.upos:<8} head={token.head}  {token.deprel}")

Refer to the combo-nlp documentation for installation and usage instructions:

License

The training data license: cc-by-nc-sa-3.0 is derived from the Universal Dependencies treebank. For the full license terms of each treebank, please refer to the corresponding LICENSE.txt file in the treebank repository:

UD_Tamil-TTB LICENSE.txt

Citation

If you use this model, please cite:

Ulewicz, M., Jabłońska, M., Klimaszewski, M., Przybyła, P., Pszenny, Ł., Rybak, P., Wiącek, M., & Wróblewska, A. (2026). COMBO-NLP Models Trained on UD v2.17. Zenodo. https://doi.org/10.5281/zenodo.19650523

bibtex

@software{combo_nlp_2026,
  author    = {Ulewicz, Michał and Jabłońska, Maja and Klimaszewski, Mateusz and Przybyła, Piotr and Pszenny, Łukasz and Rybak, Piotr and Wiącek, Martyna and Wróblewska, Alina},
  title     = {{COMBO-NLP} Models Trained on {UD} v2.17},
  year      = {2026},
  publisher = {Zenodo},
  doi       = {10.5281/zenodo.19650523},
  url       = {https://doi.org/10.5281/zenodo.19650523}
}

Treebank References

If you use the Tamil TTB treebank data, please also cite:

bibtex

@inproceedings{ta,
  author    = {Ramasamy, Loganathan and \v{Z}abokrtsk\'{y}, Zden\v{e}k},
  year      = {2012},
  title     = {Prague Dependency Style Treebank for {Tamil}},
  booktitle = {Proceedings of Eighth International Conference on Language Resources and Evaluation ({LREC} 2012)},
  address   = {\.{I}stanbul, Turkey},
  editor    = {Nicoletta Calzolari (Conference Chair) and Khalid Choukri and Thierry Declerck and Mehmet Uğur Doğan and Bente Maegaard and Joseph Mariani and Asuncion Moreno and Jan Odijk and Stelios Piperidis},
  isbn      = {978-2-9517408-7-7},
  pages     = {1888--1894},
  url       = {http://www.lrec-conf.org/proceedings/lrec2012/summaries/456.html}
}

Resources

combo-nlp: https://pypi.org/project/combo-nlp/
combo-seg: https://pypi.org/project/combo-seg/
UD_Tamil-TTB: https://github.com/UniversalDependencies/UD_Tamil-TTB

clarin-pl/combo-nlp-xlm-roberta-base-tamil-ttb-ud2.17

COMBO-NLP Model for Tamil

Model Description

Evaluation