xlm-roberta-base-ft-udpos28-sk Open-source Multilingual Part-of-Speech Tagging Model

Home

Xlm Roberta Base Ft Udpos28 Sk

Developed by wietsedv

Multilingual POS tagging model based on XLM-RoBERTa, specifically optimized for Slovak

Sequence Labeling

Transformers

OtherOpen Source License:Apache-2.0 #Multilingual POS tagging #High accuracy #Cross-lingual transfer

Downloads 14

Release Time : 3/2/2022

Model Overview

This model is a multilingual POS tagging model based on the XLM-RoBERTa architecture, fine-tuned on Universal Dependencies v2.8, with special optimization for Slovak language.

Model Features

Multilingual support

Based on XLM-RoBERTa architecture, capable of handling multiple languages

High accuracy

Achieves 97.5% accuracy on Slovak POS tagging tasks

Universal Dependencies compatibility

Trained and evaluated using Universal Dependencies v2.8 standards

Model Capabilities

POS tagging

Multilingual text processing

Token classification

Use Cases

Natural Language Processing

Text analysis

Perform POS tagging on Slovak text

Accurately identifies the POS categories of words in the text

Linguistic research

Supports linguistic researchers in analyzing Slovak grammar structures

Provides accurate POS tagging data

🚀 XLM-RoBERTa base Universal Dependencies v2.8 POS tagging: Slovak

This model is designed for part - of - speech tagging in Slovak, leveraging the XLM - RoBERTa base architecture and Universal Dependencies v2.8 dataset, offering high - quality tagging results.

🚀 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.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForTokenClassification

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

📄 License

The model is released under the Apache - 2.0 license.

📋 Model Information

Property	Details
Library Name	transformers
Tags	part - of - speech, token - classification
Datasets	universal_dependencies
Metrics	accuracy
Model Name	xlm - roberta - base - ft - udpos28 - sk

Model Results

The following table shows the accuracy metrics of the model on different languages:

Language	Test Accuracy
English	82.6
Dutch	84.2
German	79.4
Italian	82.0
French	83.9
Spanish	87.9
Russian	90.5
Swedish	84.6
Norwegian	77.9
Danish	82.2
Low Saxon	53.9
Akkadian	35.8
Armenian	83.8
Welsh	64.8
Old East Slavic	74.9
Albanian	77.9
Slovenian	87.7
Guajajara	36.6
Kurmanji	76.5
Turkish	75.1
Finnish	79.5
Indonesian	81.3
Ukrainian	92.0
Polish	93.3
Portuguese	85.1
Kazakh	79.5
Latin	77.1
Old French	58.0
Buryat	60.6
Kaapor	22.1
Korean	57.4
Estonian	80.7
Croatian	93.7
Gothic	28.3
Swiss German	44.1
Assyrian	14.8
North Sami	40.6
Naija	39.9
Latvian	84.2
Chinese	42.5
Tagalog	70.8
Bambara	28.8
Lithuanian	85.8
Galician	86.1
Vietnamese	67.4
Greek	84.6
Catalan	85.8
Czech	94.3
Erzya	49.8
Bhojpuri	48.1
Thai	58.1
Marathi	87.7
Basque	74.0
Slovak	97.5
Kiche	33.9
Yoruba	26.9
Warlpiri	42.1
Tamil	83.0
Maltese	29.1
Ancient Greek	59.0
Icelandic	77.4
Mbya Guarani	33.1
Urdu	62.2
Romanian	81.4
Persian	77.9
Apurina	46.7
Japanese	27.4
Hungarian	81.9
Hindi	65.3
Classical Chinese	30.2
Komi Permyak	48.7
Faroese	75.4
Sanskrit	36.3
Livvi	64.9
Arabic	79.6
Wolof	39.0
Bulgarian	90.5
Akuntsu	39.1
Makurap	24.7
Kangri	49.9
Breton	61.8
Telugu	79.6
Cantonese	45.6
Old Church Slavonic	45.9
Karelian	67.9
Upper Sorbian	78.6
South Levantine Arabic	66.7
Komi Zyrian	44.2
Irish	67.2
Nayini	43.6
Munduruku	27.3
Manx	36.8
Skolt Sami	41.3
Afrikaans	79.2
Old Turkish	38.0
Tupinamba	40.3
Belarusian	89.8
Serbian	94.6
Moksha	48.2
Western Armenian	76.0
Scottish Gaelic	57.0
Khunsari	37.8
Hebrew	81.2
Uyghur	72.4
Chukchi	37.0

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