@inproceedings{anh-etal-2024-morphology,
title = "Morphology Matters: Probing the Cross-linguistic Morphological Generalization Abilities of Large Language Models through a Wug Test",
author = "Anh, Dang and
Raviv, Limor and
Galke, Lukas",
editor = "Kuribayashi, Tatsuki and
Rambelli, Giulia and
Takmaz, Ece and
Wicke, Philipp and
Oseki, Yohei",
booktitle = "Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.cmcl-1.15",
doi = "10.18653/v1/2024.cmcl-1.15",
pages = "177--188",
abstract = "We develop a multilingual version of the Wug Test, an artificial word completion experiment that is typically used to test the morphological knowledge of children, and apply it to the GPT family of large language models (LLMs). LLMs{'} performance on this test was evaluated by native speakers of six different languages, who judged whether the inflected and derived forms generated by the models conform to the morphological rules of their language. Our results show that LLMs can generalize their morphological knowledge to new, unfamiliar words, but that their success in generating the {``}correct{''} generalization (as judged by native human speakers) is predicted by a language{'}s morphological complexity (specifically, integrative complexity). We further find that the amount of training data has surprisingly little on LLMs{'} morphological generalization abilities within the scope of the analyzed languages. These findings highlight that {``}morphology matters{''}, and have important implications for improving low-resource language modeling.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="anh-etal-2024-morphology">
<titleInfo>
<title>Morphology Matters: Probing the Cross-linguistic Morphological Generalization Abilities of Large Language Models through a Wug Test</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dang</namePart>
<namePart type="family">Anh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Limor</namePart>
<namePart type="family">Raviv</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lukas</namePart>
<namePart type="family">Galke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tatsuki</namePart>
<namePart type="family">Kuribayashi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Giulia</namePart>
<namePart type="family">Rambelli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ece</namePart>
<namePart type="family">Takmaz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Philipp</namePart>
<namePart type="family">Wicke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yohei</namePart>
<namePart type="family">Oseki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We develop a multilingual version of the Wug Test, an artificial word completion experiment that is typically used to test the morphological knowledge of children, and apply it to the GPT family of large language models (LLMs). LLMs’ performance on this test was evaluated by native speakers of six different languages, who judged whether the inflected and derived forms generated by the models conform to the morphological rules of their language. Our results show that LLMs can generalize their morphological knowledge to new, unfamiliar words, but that their success in generating the “correct” generalization (as judged by native human speakers) is predicted by a language’s morphological complexity (specifically, integrative complexity). We further find that the amount of training data has surprisingly little on LLMs’ morphological generalization abilities within the scope of the analyzed languages. These findings highlight that “morphology matters”, and have important implications for improving low-resource language modeling.</abstract>
<identifier type="citekey">anh-etal-2024-morphology</identifier>
<identifier type="doi">10.18653/v1/2024.cmcl-1.15</identifier>
<location>
<url>https://aclanthology.org/2024.cmcl-1.15</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>177</start>
<end>188</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Morphology Matters: Probing the Cross-linguistic Morphological Generalization Abilities of Large Language Models through a Wug Test
%A Anh, Dang
%A Raviv, Limor
%A Galke, Lukas
%Y Kuribayashi, Tatsuki
%Y Rambelli, Giulia
%Y Takmaz, Ece
%Y Wicke, Philipp
%Y Oseki, Yohei
%S Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F anh-etal-2024-morphology
%X We develop a multilingual version of the Wug Test, an artificial word completion experiment that is typically used to test the morphological knowledge of children, and apply it to the GPT family of large language models (LLMs). LLMs’ performance on this test was evaluated by native speakers of six different languages, who judged whether the inflected and derived forms generated by the models conform to the morphological rules of their language. Our results show that LLMs can generalize their morphological knowledge to new, unfamiliar words, but that their success in generating the “correct” generalization (as judged by native human speakers) is predicted by a language’s morphological complexity (specifically, integrative complexity). We further find that the amount of training data has surprisingly little on LLMs’ morphological generalization abilities within the scope of the analyzed languages. These findings highlight that “morphology matters”, and have important implications for improving low-resource language modeling.
%R 10.18653/v1/2024.cmcl-1.15
%U https://aclanthology.org/2024.cmcl-1.15
%U https://doi.org/10.18653/v1/2024.cmcl-1.15
%P 177-188
Markdown (Informal)
[Morphology Matters: Probing the Cross-linguistic Morphological Generalization Abilities of Large Language Models through a Wug Test](https://aclanthology.org/2024.cmcl-1.15) (Anh et al., CMCL-WS 2024)
ACL