@inproceedings{kudo-etal-2023-deep,
title = "Do Deep Neural Networks Capture Compositionality in Arithmetic Reasoning?",
author = "Kudo, Keito and
Aoki, Yoichi and
Kuribayashi, Tatsuki and
Brassard, Ana and
Yoshikawa, Masashi and
Sakaguchi, Keisuke and
Inui, Kentaro",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.eacl-main.98",
doi = "10.18653/v1/2023.eacl-main.98",
pages = "1351--1362",
abstract = "Compositionality is a pivotal property of symbolic reasoning. However, how well recent neural models capture compositionality remains underexplored in the symbolic reasoning tasks. This study empirically addresses this question by systematically examining recently published pre-trained seq2seq models with a carefully controlled dataset of multi-hop arithmetic symbolic reasoning. We introduce a skill tree on compositionality in arithmetic symbolic reasoning that defines the hierarchical levels of complexity along with three compositionality dimensions: systematicity, productivity, and substitutivity. Our experiments revealed that among the three types of composition, the models struggled most with systematicity, performing poorly even with relatively simple compositions. That difficulty was not resolved even after training the models with intermediate reasoning steps.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kudo-etal-2023-deep">
<titleInfo>
<title>Do Deep Neural Networks Capture Compositionality in Arithmetic Reasoning?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Keito</namePart>
<namePart type="family">Kudo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yoichi</namePart>
<namePart type="family">Aoki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tatsuki</namePart>
<namePart type="family">Kuribayashi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ana</namePart>
<namePart type="family">Brassard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Masashi</namePart>
<namePart type="family">Yoshikawa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Keisuke</namePart>
<namePart type="family">Sakaguchi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isabelle</namePart>
<namePart type="family">Augenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dubrovnik, Croatia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Compositionality is a pivotal property of symbolic reasoning. However, how well recent neural models capture compositionality remains underexplored in the symbolic reasoning tasks. This study empirically addresses this question by systematically examining recently published pre-trained seq2seq models with a carefully controlled dataset of multi-hop arithmetic symbolic reasoning. We introduce a skill tree on compositionality in arithmetic symbolic reasoning that defines the hierarchical levels of complexity along with three compositionality dimensions: systematicity, productivity, and substitutivity. Our experiments revealed that among the three types of composition, the models struggled most with systematicity, performing poorly even with relatively simple compositions. That difficulty was not resolved even after training the models with intermediate reasoning steps.</abstract>
<identifier type="citekey">kudo-etal-2023-deep</identifier>
<identifier type="doi">10.18653/v1/2023.eacl-main.98</identifier>
<location>
<url>https://aclanthology.org/2023.eacl-main.98</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>1351</start>
<end>1362</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Do Deep Neural Networks Capture Compositionality in Arithmetic Reasoning?
%A Kudo, Keito
%A Aoki, Yoichi
%A Kuribayashi, Tatsuki
%A Brassard, Ana
%A Yoshikawa, Masashi
%A Sakaguchi, Keisuke
%A Inui, Kentaro
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F kudo-etal-2023-deep
%X Compositionality is a pivotal property of symbolic reasoning. However, how well recent neural models capture compositionality remains underexplored in the symbolic reasoning tasks. This study empirically addresses this question by systematically examining recently published pre-trained seq2seq models with a carefully controlled dataset of multi-hop arithmetic symbolic reasoning. We introduce a skill tree on compositionality in arithmetic symbolic reasoning that defines the hierarchical levels of complexity along with three compositionality dimensions: systematicity, productivity, and substitutivity. Our experiments revealed that among the three types of composition, the models struggled most with systematicity, performing poorly even with relatively simple compositions. That difficulty was not resolved even after training the models with intermediate reasoning steps.
%R 10.18653/v1/2023.eacl-main.98
%U https://aclanthology.org/2023.eacl-main.98
%U https://doi.org/10.18653/v1/2023.eacl-main.98
%P 1351-1362
Markdown (Informal)
[Do Deep Neural Networks Capture Compositionality in Arithmetic Reasoning?](https://aclanthology.org/2023.eacl-main.98) (Kudo et al., EACL 2023)
ACL
- Keito Kudo, Yoichi Aoki, Tatsuki Kuribayashi, Ana Brassard, Masashi Yoshikawa, Keisuke Sakaguchi, and Kentaro Inui. 2023. Do Deep Neural Networks Capture Compositionality in Arithmetic Reasoning?. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 1351–1362, Dubrovnik, Croatia. Association for Computational Linguistics.