Cross-lingual Models of Word Embeddings: An Empirical Comparison

2016/04/01 by Upadhyay, Shyam, Faruqui, Manaal, Dyer, Chris +1 · 1 citation
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1604.00425

Abstract

Despite interest in using cross-lingual knowledge to learn word embeddings for various tasks, a systematic comparison of the possible approaches is lacking in the literature. We perform an extensive evaluation of four popular approaches of inducing cross-lingual embeddings, each requiring a different form of supervision, on four typographically different language pairs. Our evaluation setup spans four different tasks, including intrinsic evaluation on mono-lingual and cross-lingual similarity, and extrinsic evaluation on downstream semantic and syntactic applications. We show that models which require expensive cross-lingual knowledge almost always perform better, but cheaply supervised models often prove competitive on certain tasks.

Cited by

• Mafoko Companion Dataset for Mafoko Open Multilingual Terminologies Paper

Related

• On the Robustness of Unsupervised and Semi-supervised Cross-lingual Word Embedding Learning
• Learning Crosslingual Word Embeddings without Bilingual Corpora
• Machine Translation with Cross-lingual Word Embeddings
• Best Practices for Learning Domain-Specific Cross-Lingual Embeddings
• Interactive Refinement of Cross-Lingual Word Embeddings