@inproceedings{wei-etal-2022-compositional,
title = "Compositional Generalization for Kinship Prediction through Data Augmentation",
author = "Wei, Kangda and
Ghosh, Sayan and
Srivastava, Shashank",
editor = "Clark, Elizabeth and
Brahman, Faeze and
Iyyer, Mohit",
booktitle = "Proceedings of the 4th Workshop of Narrative Understanding (WNU2022)",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.wnu-1.2/",
doi = "10.18653/v1/2022.wnu-1.2",
pages = "13--19",
abstract = "Transformer-based models have shown promising performance in numerous NLP tasks. However, recent work has shown the limitation of such models in showing compositional generalization, which requires models to generalize to novel compositions of known concepts. In this work, we explore two strategies for compositional generalization on the task of kinship prediction from stories, (1) data augmentation and (2) predicting and using intermediate structured representation (in form of kinship graphs). Our experiments show that data augmentation boosts generalization performance by around 20{\%} on average relative to a baseline model from prior work not using these strategies. However, predicting and using intermediate kinship graphs leads to a deterioration in the generalization of kinship prediction by around 50{\%} on average relative to models that only leverage data augmentation."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wei-etal-2022-compositional">
<titleInfo>
<title>Compositional Generalization for Kinship Prediction through Data Augmentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kangda</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sayan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shashank</namePart>
<namePart type="family">Srivastava</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 4th Workshop of Narrative Understanding (WNU2022)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Elizabeth</namePart>
<namePart type="family">Clark</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Faeze</namePart>
<namePart type="family">Brahman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Iyyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Transformer-based models have shown promising performance in numerous NLP tasks. However, recent work has shown the limitation of such models in showing compositional generalization, which requires models to generalize to novel compositions of known concepts. In this work, we explore two strategies for compositional generalization on the task of kinship prediction from stories, (1) data augmentation and (2) predicting and using intermediate structured representation (in form of kinship graphs). Our experiments show that data augmentation boosts generalization performance by around 20% on average relative to a baseline model from prior work not using these strategies. However, predicting and using intermediate kinship graphs leads to a deterioration in the generalization of kinship prediction by around 50% on average relative to models that only leverage data augmentation.</abstract>
<identifier type="citekey">wei-etal-2022-compositional</identifier>
<identifier type="doi">10.18653/v1/2022.wnu-1.2</identifier>
<location>
<url>https://aclanthology.org/2022.wnu-1.2/</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>13</start>
<end>19</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Compositional Generalization for Kinship Prediction through Data Augmentation
%A Wei, Kangda
%A Ghosh, Sayan
%A Srivastava, Shashank
%Y Clark, Elizabeth
%Y Brahman, Faeze
%Y Iyyer, Mohit
%S Proceedings of the 4th Workshop of Narrative Understanding (WNU2022)
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, United States
%F wei-etal-2022-compositional
%X Transformer-based models have shown promising performance in numerous NLP tasks. However, recent work has shown the limitation of such models in showing compositional generalization, which requires models to generalize to novel compositions of known concepts. In this work, we explore two strategies for compositional generalization on the task of kinship prediction from stories, (1) data augmentation and (2) predicting and using intermediate structured representation (in form of kinship graphs). Our experiments show that data augmentation boosts generalization performance by around 20% on average relative to a baseline model from prior work not using these strategies. However, predicting and using intermediate kinship graphs leads to a deterioration in the generalization of kinship prediction by around 50% on average relative to models that only leverage data augmentation.
%R 10.18653/v1/2022.wnu-1.2
%U https://aclanthology.org/2022.wnu-1.2/
%U https://doi.org/10.18653/v1/2022.wnu-1.2
%P 13-19
Markdown (Informal)
[Compositional Generalization for Kinship Prediction through Data Augmentation](https://aclanthology.org/2022.wnu-1.2/) (Wei et al., WNU 2022)
ACL