A Universal Semantic Space

01/21/2018
by   Philipp Dufter, et al.
0

Multilingual embeddings build on the success of monolingual embeddings and have applications in crosslingual transfer, in machine translation and in the digital humanities. We present the first multilingual embedding space for thousands of languages, a much larger number of languages than in prior work.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/08/2022

MMTAfrica: Multilingual Machine Translation for African Languages

In this paper, we focus on the task of multilingual machine translation ...
research
10/09/2021

An Isotropy Analysis in the Multilingual BERT Embedding Space

Several studies have explored various advantages of multilingual pre-tra...
research
05/23/2023

Is a Prestigious Job the same as a Prestigious Country? A Case Study on Multilingual Sentence Embeddings and European Countries

We study how multilingual sentence representations capture European coun...
research
04/18/2021

Embedding-Enhanced Giza++: Improving Alignment in Low- and High- Resource Scenarios Using Embedding Space Geometry

A popular natural language processing task decades ago, word alignment h...
research
09/08/2019

MULE: Multimodal Universal Language Embedding

Existing vision-language methods typically support two languages at a ti...
research
05/22/2023

Discovering Universal Geometry in Embeddings with ICA

This study employs Independent Component Analysis (ICA) to uncover unive...
research
04/21/2020

Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation

We present an easy and efficient method to extend existing sentence embe...

Please sign up or login with your details

Forgot password? Click here to reset