Trans-gram, Fast Cross-lingual Word-embeddings

We introduce Trans-gram, a simple and computationally-efficient method to simultaneously learn and align wordembeddings for a variety of languages, using only monolingual data and a smaller set of sentence-aligned data. We use our new method to compute aligned wordembeddings for twenty-one languages...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2016-01
Main Authors: Coulmance, Jocelyn, Marty, Jean-Marc, Wenzek, Guillaume, Benhalloum, Amine
Format: Article
Language:English
Subjects:
Alignment
Languages
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We introduce Trans-gram, a simple and computationally-efficient method to simultaneously learn and align wordembeddings for a variety of languages, using only monolingual data and a smaller set of sentence-aligned data. We use our new method to compute aligned wordembeddings for twenty-one languages using English as a pivot language. We show that some linguistic features are aligned across languages for which we do not have aligned data, even though those properties do not exist in the pivot language. We also achieve state of the art results on standard cross-lingual text classification and word translation tasks.
ISSN:2331-8422