@inproceedings{wang-etal-2023-hybrid,
title = "Hybrid-Regressive Paradigm for Accurate and Speed-Robust Neural Machine Translation",
author = "Wang, Qiang and
Hu, Xinhui and
Chen, Ming",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.367",
doi = "10.18653/v1/2023.findings-acl.367",
pages = "5931--5945",
abstract = "This work empirically confirms that non-autoregressive translation (NAT) is less robust in decoding batch size and hardware settings than autoregressive translation (AT). To address this issue, we demonstrate that prompting a small number of AT predictions can significantly reduce the performance gap between AT and NAT through synthetic experiments. Following this line, we propose hybrid-regressive translation (HRT), a two-stage translation prototype that combines the strengths of AT and NAT. Specifically, HRT first generates discontinuous sequences via autoregression (e.g., make a prediction for every $k$ tokens, $k>1$) and then fills in all previously skipped tokens at once in a non-autoregressive manner. Experiments on five translation tasks show that HRT achieves comparable translation quality with AT while having at least 1.5x faster inference regardless of batch size and device. Additionally, HRT successfully inherits the sound characteristics of AT in the deep-encoder-shallow-decoder architecture, allowing for further speedup without BLEU loss.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2023-hybrid">
<titleInfo>
<title>Hybrid-Regressive Paradigm for Accurate and Speed-Robust Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Qiang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinhui</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This work empirically confirms that non-autoregressive translation (NAT) is less robust in decoding batch size and hardware settings than autoregressive translation (AT). To address this issue, we demonstrate that prompting a small number of AT predictions can significantly reduce the performance gap between AT and NAT through synthetic experiments. Following this line, we propose hybrid-regressive translation (HRT), a two-stage translation prototype that combines the strengths of AT and NAT. Specifically, HRT first generates discontinuous sequences via autoregression (e.g., make a prediction for every k tokens, k>1) and then fills in all previously skipped tokens at once in a non-autoregressive manner. Experiments on five translation tasks show that HRT achieves comparable translation quality with AT while having at least 1.5x faster inference regardless of batch size and device. Additionally, HRT successfully inherits the sound characteristics of AT in the deep-encoder-shallow-decoder architecture, allowing for further speedup without BLEU loss.</abstract>
<identifier type="citekey">wang-etal-2023-hybrid</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.367</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.367</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>5931</start>
<end>5945</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Hybrid-Regressive Paradigm for Accurate and Speed-Robust Neural Machine Translation
%A Wang, Qiang
%A Hu, Xinhui
%A Chen, Ming
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F wang-etal-2023-hybrid
%X This work empirically confirms that non-autoregressive translation (NAT) is less robust in decoding batch size and hardware settings than autoregressive translation (AT). To address this issue, we demonstrate that prompting a small number of AT predictions can significantly reduce the performance gap between AT and NAT through synthetic experiments. Following this line, we propose hybrid-regressive translation (HRT), a two-stage translation prototype that combines the strengths of AT and NAT. Specifically, HRT first generates discontinuous sequences via autoregression (e.g., make a prediction for every k tokens, k>1) and then fills in all previously skipped tokens at once in a non-autoregressive manner. Experiments on five translation tasks show that HRT achieves comparable translation quality with AT while having at least 1.5x faster inference regardless of batch size and device. Additionally, HRT successfully inherits the sound characteristics of AT in the deep-encoder-shallow-decoder architecture, allowing for further speedup without BLEU loss.
%R 10.18653/v1/2023.findings-acl.367
%U https://aclanthology.org/2023.findings-acl.367
%U https://doi.org/10.18653/v1/2023.findings-acl.367
%P 5931-5945
Markdown (Informal)
[Hybrid-Regressive Paradigm for Accurate and Speed-Robust Neural Machine Translation](https://aclanthology.org/2023.findings-acl.367) (Wang et al., Findings 2023)
ACL