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

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

from transformers import AutoTokenizer, AutoModelForTokenClassification

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

Downloads last month: 11

Safetensors

Model size

277M params

Tensor type

I64

F32

Inference Examples

Token Classification

This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train wietsedv/xlm-roberta-base-ft-udpos28-ja

Space using wietsedv/xlm-roberta-base-ft-udpos28-ja 1

Evaluation results

English Test accuracy on Universal Dependencies v2.8
self-reported

47.700
Dutch Test accuracy on Universal Dependencies v2.8
self-reported

49.800
German Test accuracy on Universal Dependencies v2.8
self-reported

55.700
Italian Test accuracy on Universal Dependencies v2.8
self-reported

52.000
French Test accuracy on Universal Dependencies v2.8
self-reported

47.200
Spanish Test accuracy on Universal Dependencies v2.8
self-reported

48.200
Russian Test accuracy on Universal Dependencies v2.8
self-reported

62.700
Swedish Test accuracy on Universal Dependencies v2.8
self-reported

52.600
Norwegian Test accuracy on Universal Dependencies v2.8
self-reported

48.600
Danish Test accuracy on Universal Dependencies v2.8
self-reported

54.300

View on Papers With Code