Cross-Lingual Dependency Parsing with Late Decoding for Truly Low-Resource Languages

2017/01/06 by Michael Sejr Schlichtkrull, Michael Schlichtkrull, Anders Søgaard +2 · 1 citation
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Text Readability and Simplification #Topic Modeling #cs.CL

paper · pdf · doi:10.48550/arxiv.1701.01623

To be published at EACL 2017

arxiv created 2017/01/06 · openalex publication_date 2017/01/06 · arxiv updated 2017/01/09 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28

Abstract

In cross-lingual dependency annotation projection, information is often lost during transfer because of early decoding. We present an end-to-end graph-based neural network dependency parser that can be trained to reproduce matrices of edge scores, which can be directly projected across word alignments. We show that our approach to cross-lingual dependency parsing is not only simpler, but also achieves an absolute improvement of 2.25% averaged across 10 languages compared to the previous state of the art.

Cited by

• Database Views as Explanations for Relational Deep Learning
• Cross-Lingual Syntactic Transfer with Limited Resources

Related

• Many Languages, One Parser
• Mutlitask Learning for Cross-Lingual Transfer of Semantic Dependencies
• Dependency Language Models for Transition-based Dependency Parsing
• Fill it up: Exploiting partial dependency annotations in a minimum spanning tree parser
• Cross-lingual Universal Dependency Parsing Only from One Monolingual Treebank