clarin-pl/combo-nlp-xlm-roberta-base-ancient-greek-proiel-ud2.17

COMBO-NLP Model for Ancient_Greek

Model Description

This is a Ancient_Greek-language model based on COMBO-NLP, an open-source natural language preprocessing system. It performs:

• sentence segmentation (via LAMBO)
• tokenisation (via LAMBO)
• part-of-speech tagging
• morphological analysis
• lemmatisation
• dependency parsing

The Ancient_Greek model uses FacebookAI/xlm-roberta-base as its base encoder and is trained on UD_Ancient_Greek-PROIEL (UD v2.17).

Evaluation

Evaluation was performed on the UD_Ancient_Greek-PROIEL test split using the standard CoNLL 2018 eval script.

Two evaluation modes are reported (all values are F1 scores):

• Full-text: raw text is segmented by LAMBO, then parsed and compared against gold — measures end-to-end pipeline performance including segmentation quality.
• Aligned: gold tokenization is used as input — measures parsing quality independent of segmentation (upper bound).

Morphosyntactic Tagging

Metric	Tokens	Sentences	Words	UPOS	XPOS	UFeats	AllTags	Lemmas
Full-text	99.90	67.37	99.90	97.53	97.66	90.15	89.09	97.47
Aligned	n/a	n/a	n/a	97.85	97.90	90.45	89.58	97.66

Dependency Parsing

Metric	UAS	LAS	CLAS	MLAS	BLEX
Full-text	85.14	81.63	75.69	64.85	73.51
Aligned	88.45	84.87	80.01	68.74	77.72

Usage

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

pip install combo-nlp

Install the Lambo segmenter - only needed when passing raw text strings to COMBO:

pip install --index-url https://pypi.clarin-pl.eu/ lambo

Usage with default segmenter

from combo import COMBO

# Load a pre-trained model with corresponding Lambo segmenter
nlp = COMBO("Ancient_Greek")

# Parse raw text (handles sentence splitting + tokenization)
result = nlp("Ἡ ταχεῖα φαιὰ ἀλώπηξ ὑπὲρ τοῦ ἀργοῦ κυνὸς πηδᾷ.")

# 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:

• https://gitlab.clarin-pl.eu/syntactic-tools/combo-nlp
• https://gitlab.clarin-pl.eu/syntactic-tools/lambo

Citation

Resources

• COMBO-NLP: https://gitlab.clarin-pl.eu/syntactic-tools/combo-nlp
• LAMBO: https://gitlab.clarin-pl.eu/syntactic-tools/lambo
• UD_Ancient_Greek-PROIEL: https://github.com/UniversalDependencies/UD_Ancient_Greek-PROIEL