@inproceedings{tang-etal-2019-understanding,
title = "Understanding Neural Machine Translation by Simplification: The Case of Encoder-free Models",
author = "Tang, Gongbo and
Sennrich, Rico and
Nivre, Joakim",
editor = "Mitkov, Ruslan and
Angelova, Galia",
booktitle = "Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)",
month = sep,
year = "2019",
address = "Varna, Bulgaria",
publisher = "INCOMA Ltd.",
url = "https://aclanthology.org/R19-1136",
doi = "10.26615/978-954-452-056-4_136",
pages = "1186--1193",
abstract = "In this paper, we try to understand neural machine translation (NMT) via simplifying NMT architectures and training encoder-free NMT models. In an encoder-free model, the sums of word embeddings and positional embeddings represent the source. The decoder is a standard Transformer or recurrent neural network that directly attends to embeddings via attention mechanisms. Experimental results show (1) that the attention mechanism in encoder-free models acts as a strong feature extractor, (2) that the word embeddings in encoder-free models are competitive to those in conventional models, (3) that non-contextualized source representations lead to a big performance drop, and (4) that encoder-free models have different effects on alignment quality for German-English and Chinese-English.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tang-etal-2019-understanding">
<titleInfo>
<title>Understanding Neural Machine Translation by Simplification: The Case of Encoder-free Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gongbo</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rico</namePart>
<namePart type="family">Sennrich</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joakim</namePart>
<namePart type="family">Nivre</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ruslan</namePart>
<namePart type="family">Mitkov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Galia</namePart>
<namePart type="family">Angelova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>INCOMA Ltd.</publisher>
<place>
<placeTerm type="text">Varna, Bulgaria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In this paper, we try to understand neural machine translation (NMT) via simplifying NMT architectures and training encoder-free NMT models. In an encoder-free model, the sums of word embeddings and positional embeddings represent the source. The decoder is a standard Transformer or recurrent neural network that directly attends to embeddings via attention mechanisms. Experimental results show (1) that the attention mechanism in encoder-free models acts as a strong feature extractor, (2) that the word embeddings in encoder-free models are competitive to those in conventional models, (3) that non-contextualized source representations lead to a big performance drop, and (4) that encoder-free models have different effects on alignment quality for German-English and Chinese-English.</abstract>
<identifier type="citekey">tang-etal-2019-understanding</identifier>
<identifier type="doi">10.26615/978-954-452-056-4_136</identifier>
<location>
<url>https://aclanthology.org/R19-1136</url>
</location>
<part>
<date>2019-09</date>
<extent unit="page">
<start>1186</start>
<end>1193</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Understanding Neural Machine Translation by Simplification: The Case of Encoder-free Models
%A Tang, Gongbo
%A Sennrich, Rico
%A Nivre, Joakim
%Y Mitkov, Ruslan
%Y Angelova, Galia
%S Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)
%D 2019
%8 September
%I INCOMA Ltd.
%C Varna, Bulgaria
%F tang-etal-2019-understanding
%X In this paper, we try to understand neural machine translation (NMT) via simplifying NMT architectures and training encoder-free NMT models. In an encoder-free model, the sums of word embeddings and positional embeddings represent the source. The decoder is a standard Transformer or recurrent neural network that directly attends to embeddings via attention mechanisms. Experimental results show (1) that the attention mechanism in encoder-free models acts as a strong feature extractor, (2) that the word embeddings in encoder-free models are competitive to those in conventional models, (3) that non-contextualized source representations lead to a big performance drop, and (4) that encoder-free models have different effects on alignment quality for German-English and Chinese-English.
%R 10.26615/978-954-452-056-4_136
%U https://aclanthology.org/R19-1136
%U https://doi.org/10.26615/978-954-452-056-4_136
%P 1186-1193
Markdown (Informal)
[Understanding Neural Machine Translation by Simplification: The Case of Encoder-free Models](https://aclanthology.org/R19-1136) (Tang et al., RANLP 2019)
ACL