Word Position Aware Translation Memory for Neural Machine Translation

Autor:	Lemao Liu, Li Li, Guoping Huang, Qiuxiang He
Rok vydání:	2019
Předmět:	Machine translation business.industry Computer science Computation Translation (geometry) computer.software_genre Position (vector) Key (cryptography) Translation memory Artificial intelligence business computer Word (computer architecture) Decoding methods Natural language processing
Zdroj:	Natural Language Processing and Chinese Computing ISBN: 9783030322328 NLPCC (1)
Popis:	The approach based on translation pieces is appealing for neural machine translation with a translation memory (TM), owing to its efficiency in both computation and memory consumption. Unfortunately, it is incapable of capturing sufficient contextual translation leading to a limited translation performance. This paper thereby proposes a simple yet effective approach to address this issue. Its key idea is to employ the word position information from a TM as additional rewards to guide the decoding of neural machine translation (NMT). Experiments on seven tasks show that the proposed approach yields consistent gains particularly for those source sentences whose TM is very similar to themselves, while maintaining similar efficiency to the counterpart of translation pieces.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::4247b0472cbdec9a589ab6e2a58ed24a https://doi.org/10.1007/978-3-030-32233-5_29 Zobrazit plný text záznamu