@inproceedings{umakoshi-etal-2021-japanese-zero,
title = "{J}apanese Zero Anaphora Resolution Can Benefit from Parallel Texts Through Neural Transfer Learning",
author = "Umakoshi, Masato and
Murawaki, Yugo and
Kurohashi, Sadao",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
month = nov,
year = "2021",
address = "Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-emnlp.165",
doi = "10.18653/v1/2021.findings-emnlp.165",
pages = "1920--1934",
abstract = "Parallel texts of Japanese and a non-pro-drop language have the potential of improving the performance of Japanese zero anaphora resolution (ZAR) because pronouns dropped in the former are usually mentioned explicitly in the latter. However, rule-based cross-lingual transfer is hampered by error propagation in an NLP pipeline and the frequent lack of transparency in translation correspondences. In this paper, we propose implicit transfer by injecting machine translation (MT) as an intermediate task between pretraining and ZAR. We employ a pretrained BERT model to initialize the encoder part of the encoder-decoder model for MT, and eject the encoder part for fine-tuning on ZAR. The proposed framework empirically demonstrates that ZAR performance can be improved by transfer learning from MT. In addition, we find that the incorporation of the masked language model training into MT leads to further gains.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="umakoshi-etal-2021-japanese-zero">
<titleInfo>
<title>Japanese Zero Anaphora Resolution Can Benefit from Parallel Texts Through Neural Transfer Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Masato</namePart>
<namePart type="family">Umakoshi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yugo</namePart>
<namePart type="family">Murawaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sadao</namePart>
<namePart type="family">Kurohashi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2021</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Parallel texts of Japanese and a non-pro-drop language have the potential of improving the performance of Japanese zero anaphora resolution (ZAR) because pronouns dropped in the former are usually mentioned explicitly in the latter. However, rule-based cross-lingual transfer is hampered by error propagation in an NLP pipeline and the frequent lack of transparency in translation correspondences. In this paper, we propose implicit transfer by injecting machine translation (MT) as an intermediate task between pretraining and ZAR. We employ a pretrained BERT model to initialize the encoder part of the encoder-decoder model for MT, and eject the encoder part for fine-tuning on ZAR. The proposed framework empirically demonstrates that ZAR performance can be improved by transfer learning from MT. In addition, we find that the incorporation of the masked language model training into MT leads to further gains.</abstract>
<identifier type="citekey">umakoshi-etal-2021-japanese-zero</identifier>
<identifier type="doi">10.18653/v1/2021.findings-emnlp.165</identifier>
<location>
<url>https://aclanthology.org/2021.findings-emnlp.165</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>1920</start>
<end>1934</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Japanese Zero Anaphora Resolution Can Benefit from Parallel Texts Through Neural Transfer Learning
%A Umakoshi, Masato
%A Murawaki, Yugo
%A Kurohashi, Sadao
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Findings of the Association for Computational Linguistics: EMNLP 2021
%D 2021
%8 November
%I Association for Computational Linguistics
%C Punta Cana, Dominican Republic
%F umakoshi-etal-2021-japanese-zero
%X Parallel texts of Japanese and a non-pro-drop language have the potential of improving the performance of Japanese zero anaphora resolution (ZAR) because pronouns dropped in the former are usually mentioned explicitly in the latter. However, rule-based cross-lingual transfer is hampered by error propagation in an NLP pipeline and the frequent lack of transparency in translation correspondences. In this paper, we propose implicit transfer by injecting machine translation (MT) as an intermediate task between pretraining and ZAR. We employ a pretrained BERT model to initialize the encoder part of the encoder-decoder model for MT, and eject the encoder part for fine-tuning on ZAR. The proposed framework empirically demonstrates that ZAR performance can be improved by transfer learning from MT. In addition, we find that the incorporation of the masked language model training into MT leads to further gains.
%R 10.18653/v1/2021.findings-emnlp.165
%U https://aclanthology.org/2021.findings-emnlp.165
%U https://doi.org/10.18653/v1/2021.findings-emnlp.165
%P 1920-1934
Markdown (Informal)
[Japanese Zero Anaphora Resolution Can Benefit from Parallel Texts Through Neural Transfer Learning](https://aclanthology.org/2021.findings-emnlp.165) (Umakoshi et al., Findings 2021)
ACL