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

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

Downloads last month: 7

Safetensors

Model size

277M params

Tensor type

I64

F32

Inference Providers NEW

Token Classification

This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

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

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

Evaluation results

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

76.600
Dutch Test accuracy on Universal Dependencies v2.8
self-reported

73.700
German Test accuracy on Universal Dependencies v2.8
self-reported

70.500
Italian Test accuracy on Universal Dependencies v2.8
self-reported

75.100
French Test accuracy on Universal Dependencies v2.8
self-reported

71.300
Spanish Test accuracy on Universal Dependencies v2.8
self-reported

74.500
Russian Test accuracy on Universal Dependencies v2.8
self-reported

80.300
Swedish Test accuracy on Universal Dependencies v2.8
self-reported

79.300
Norwegian Test accuracy on Universal Dependencies v2.8
self-reported

75.700
Danish Test accuracy on Universal Dependencies v2.8
self-reported

80.400

View on Papers With Code