English

Low-Resource Syntactic Transfer with Unsupervised Source Reordering

Computation and Language 2019-03-15 v1

Abstract

We describe a cross-lingual transfer method for dependency parsing that takes into account the problem of word order differences between source and target languages. Our model only relies on the Bible, a considerably smaller parallel data than the commonly used parallel data in transfer methods. We use the concatenation of projected trees from the Bible corpus, and the gold-standard treebanks in multiple source languages along with cross-lingual word representations. We demonstrate that reordering the source treebanks before training on them for a target language improves the accuracy of languages outside the European language family. Our experiments on 68 treebanks (38 languages) in the Universal Dependencies corpus achieve a high accuracy for all languages. Among them, our experiments on 16 treebanks of 12 non-European languages achieve an average UAS absolute improvement of 3.3% over a state-of-the-art method.

Keywords

Cite

@article{arxiv.1903.05683,
  title  = {Low-Resource Syntactic Transfer with Unsupervised Source Reordering},
  author = {Mohammad Sadegh Rasooli and Michael Collins},
  journal= {arXiv preprint arXiv:1903.05683},
  year   = {2019}
}

Comments

Accepted in NAACL 2019

R2 v1 2026-06-23T08:07:25.186Z