Neural Machine Translation without Embeddings

Uri Shaham, Omer Levy

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Scopus citations

Abstract

Many NLP models operate over sequences of subword tokens produced by hand-crafted tokenization rules and heuristic subword induction algorithms. A simple universal alternative is to represent every computerized text as a sequence of bytes via UTF-8, obviating the need for an embedding layer since there are fewer token types (256) than dimensions. Surprisingly, replacing the ubiquitous embedding layer with one-hot representations of each byte does not hurt performance; experiments on byte-to-byte machine translation from English to 10 different languages show a consistent improvement in BLEU, rivaling character-level and even standard subword-level models. A deeper investigation reveals that the combination of embeddingless models with decoder-input dropout amounts to token dropout, which benefits byte-to-byte models in particular.

Original languageEnglish
Title of host publicationNAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics
Subtitle of host publicationHuman Language Technologies, Proceedings of the Conference
PublisherAssociation for Computational Linguistics (ACL)
Pages181-186
Number of pages6
ISBN (Electronic)9781954085466
DOIs
StatePublished - 2021
Event2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021 - Virtual, Online
Duration: 6 Jun 202111 Jun 2021

Publication series

NameNAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference

Conference

Conference2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021
CityVirtual, Online
Period6/06/2111/06/21

Fingerprint

Dive into the research topics of 'Neural Machine Translation without Embeddings'. Together they form a unique fingerprint.

Cite this