xlm-roberta-base-ft-udpos28-orv Open-source Model - Multilingual Part-of-Speech Tagging for Old East Slavic

Xlm Roberta Base Ft Udpos28 Orv

Developed by wietsedv

A multilingual POS tagging model based on XLM-RoBERTa, supporting POS tagging tasks for Old East Slavic and other languages.

Sequence Labeling

Transformers

OtherOpen Source License:Apache-2.0 #Multilingual POS tagging #Old East Slavic optimization #Universal Dependencies parsing

Downloads 17

Release Time : 3/2/2022

Model Overview

This model is a multilingual POS tagging model based on the XLM-RoBERTa architecture, fine-tuned specifically on the Universal Dependencies v2.8 dataset, with outstanding performance on Old East Slavic.

Model Features

Multilingual support

Supports POS tagging for multiple languages including Old East Slavic.

High accuracy

Achieves 91.0% accuracy on the Old East Slavic test set.

Based on Universal Dependencies dataset

Trained and evaluated using the Universal Dependencies v2.8 dataset.

Model Capabilities

POS tagging

Multilingual text processing

Token classification

Use Cases

Linguistic analysis

Old East Slavic text analysis

Performs POS tagging on Old East Slavic texts.

Achieves 91.0% accuracy.

Multilingual text processing

Handles POS tagging tasks for multiple languages.

Supports POS tagging for over 60 languages.

🚀 XLM-RoBERTa base Universal Dependencies v2.8 POS tagging: Old East Slavic

This model is designed for part - of - speech tagging, leveraging the XLM - RoBERTa base architecture on the Universal Dependencies v2.8 dataset. It provides valuable insights into cross - lingual transfer in POS tagging across over 100 languages.

🚀 Quick Start

This model is part of our paper called:

Make the Best of Cross-lingual Transfer: Evidence from POS Tagging with over 100 Languages

Check the Space for more details.

✨ Features

Multilingual Support: Capable of performing part - of - speech tagging on over 100 languages.
High - performance Metrics: Demonstrates varying levels of accuracy across different languages, as shown in the results section.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForTokenClassification

tokenizer = AutoTokenizer.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-orv")
model = AutoModelForTokenClassification.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-orv")

📚 Documentation

Model Information

Property	Details
Model Type	xlm - roberta - base - ft - udpos28 - orv
Training Data	Universal Dependencies v2.8
Library Name	transformers
Tags	part - of - speech, token - classification

Results

The model has been evaluated on multiple languages with the accuracy metric. Here are some of the results:

Language	Test Accuracy
English	79.4
Dutch	77.8
German	79.3
Italian	77.5
French	75.2
Spanish	77.2
Russian	87.9
Swedish	83.0
Norwegian	78.6
Danish	82.9
Low Saxon	58.9
Akkadian	41.8
Armenian	82.7
Welsh	64.3
Old East Slavic	91.0
Albanian	73.4
Slovenian	73.8
Guajajara	41.7
Kurmanji	76.7
Turkish	73.5
Finnish	83.0
Indonesian	78.9
Ukrainian	86.7
Polish	85.5
Portuguese	79.5
Kazakh	79.7
Latin	80.9
Old French	60.5
Buryat	59.8
Kaapor	27.1
Korean	61.0
Estonian	83.9
Croatian	84.7
Gothic	33.1
Swiss German	53.5
Assyrian	15.7
North Sami	39.9
Naija	41.9
Latvian	85.7
Chinese	42.7
Tagalog	73.5
Bambara	29.5
Lithuanian	86.1
Galician	77.7
Vietnamese	64.8
Greek	73.8
Catalan	74.2
Czech	85.0
Erzya	46.1
Bhojpuri	56.8
Thai	60.6
Marathi	84.0
Basque	77.2
Slovak	84.3
Kiche	35.3
Yoruba	29.9
Warlpiri	33.6
Tamil	84.3
Maltese	32.0
Ancient Greek	65.7
Icelandic	81.6
Mbya Guarani	33.2
Urdu	66.2
Romanian	80.9
Persian	74.6
Apurina	44.6
Japanese	35.7
Hungarian	73.3
Hindi	75.3
Classical Chinese	41.5
Komi Permyak	49.0
Faroese	78.3
Sanskrit	43.3
Livvi	70.2
Arabic	79.8
Wolof	39.8
Bulgarian	85.8
Akuntsu	36.5
Makurap	14.4
Kangri	52.0
Breton	58.1
Telugu	79.9
Cantonese	50.8
Old Church Slavonic	78.2
Karelian	73.5
Upper Sorbian	76.0
South Levantine Arabic	70.0
Komi Zyrian	43.1
Irish	61.1
Nayini	53.8
Munduruku	26.4
Manx	44.6
Skolt Sami	45.2
Afrikaans	76.9
Old Turkish	2.7
Tupinamba	39.0
Belarusian	89.5
Serbian	85.1
Moksha	42.8
Western Armenian	77.0
Scottish Gaelic	51.6
Khunsari	54.1
Hebrew	85.4
Uyghur	74.4
Chukchi	34.5

📄 License

The model is released under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご