DE eng

Search in the Catalogues and Directories

Hits 1 – 2 of 2

1
Inducing Language-Agnostic Multilingual Representations ...
Abstract: Cross-lingual representations have the potential to make NLP techniques available to the vast majority of languages in the world. However, they currently require large pretraining corpora or access to typologically similar languages. In this work, we address these obstacles by removing language identity signals from multilingual embeddings. We examine three approaches for this: (i) re-aligning the vector spaces of target languages (all together) to a pivot source language; (ii) removing language-specific means and variances, which yields better discriminativeness of embeddings as a by-product; and (iii) increasing input similarity across languages by removing morphological contractions and sentence reordering. We evaluate on XNLI and reference-free MT across 19 typologically diverse languages. Our findings expose the limitations of these approaches -- unlike vector normalization, vector space re-alignment and text normalization do not achieve consistent gains across encoders and languages. Due to the ...
Keyword: Computational Linguistics; Data Management System; FOS Languages and literature; Linguistics; Natural Language Processing; Semantics; Translation Studies
URL: https://underline.io/lecture/29780-inducing-language-agnostic-multilingual-representations
https://dx.doi.org/10.48448/158a-7m34
BASE
Hide details
2
From Natural Language Requirements to Executable Models of Software Components
In: DTIC AND NTIS (2003)
BASE
Show details

Catalogues
0
0
0
0
0
0
0
Bibliographies
0
0
0
0
0
0
0
0
0
Linked Open Data catalogues
0
Online resources
0
0
0
0
Open access documents
2
0
0
0
0
© 2013 - 2024 Lin|gu|is|tik | Imprint | Privacy Policy | Datenschutzeinstellungen ändern