@article{he-etal-2025-investigating,
title = "Investigating Idiomaticity in Word Representations",
author = "He, Wei and
Vieira, Tiago Kramer and
Garcia, Marcos and
Scarton, Carolina and
Idiart, Marco and
Villavicencio, Aline",
journal = "Computational Linguistics",
volume = "51",
month = jun,
year = "2025",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2025.cl-2.4/",
doi = "10.1162/coli_a_00546",
pages = "505--555",
abstract = "Idiomatic expressions are an integral part of human languages, often used to express complex ideas in compressed or conventional ways (e.g., eager beaver as a keen and enthusiastic person). However, their interpretations may not be straightforwardly linked to the meanings of their individual components in isolation and this may have an impact for compositional approaches. In this article, we investigate to what extent word representation models are able to go beyond compositional word combinations and capture multiword expression idiomaticity and some of the expected properties related to idiomatic meanings. We focus on noun compounds of varying levels of idiomaticity in two languages (English and Portuguese), presenting a dataset of minimal pairs containing human idiomaticity judgments for each noun compound at both type and token levels, their paraphrases and their occurrences in naturalistic and sense-neutral contexts, totalling 32,200 sentences. We propose this set of minimal pairs for evaluating how well a model captures idiomatic meanings, and define a set of fine-grained metrics of Affinity and Scaled Similarity, to determine how sensitive the models are to perturbations that may lead to changes in idiomaticity. Affinity is a comparative measure of the similarity between an experimental item, a target and a potential distractor, and Scaled Similarity incorporates a rescaling factor to magnify the meaningful similarities within the spaces defined by each specific model. The results obtained with a variety of representative and widely used models indicate that, despite superficial indications to the contrary in the form of high similarities, idiomaticity is not yet accurately represented in current models. Moreover, the performance of models with different levels of contextualization suggests that their ability to capture context is not yet able to go beyond more superficial lexical clues provided by the words and to actually incorporate the relevant semantic clues needed for idiomaticity. By proposing model-agnostic measures for assessing the ability of models to capture idiomaticity, this article contributes to determining limitations in the handling of non-compositional structures, which is one of the directions that needs to be considered for more natural, accurate, and robust language understanding. The source code and additional materials related to this paper are available at our GitHub repository.1"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="he-etal-2025-investigating">
<titleInfo>
<title>Investigating Idiomaticity in Word Representations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tiago</namePart>
<namePart type="given">Kramer</namePart>
<namePart type="family">Vieira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcos</namePart>
<namePart type="family">Garcia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolina</namePart>
<namePart type="family">Scarton</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Idiart</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aline</namePart>
<namePart type="family">Villavicencio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Computational Linguistics</title>
</titleInfo>
<originInfo>
<issuance>continuing</issuance>
<publisher>MIT Press</publisher>
<place>
<placeTerm type="text">Cambridge, MA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">periodical</genre>
<genre authority="bibutilsgt">academic journal</genre>
</relatedItem>
<abstract>Idiomatic expressions are an integral part of human languages, often used to express complex ideas in compressed or conventional ways (e.g., eager beaver as a keen and enthusiastic person). However, their interpretations may not be straightforwardly linked to the meanings of their individual components in isolation and this may have an impact for compositional approaches. In this article, we investigate to what extent word representation models are able to go beyond compositional word combinations and capture multiword expression idiomaticity and some of the expected properties related to idiomatic meanings. We focus on noun compounds of varying levels of idiomaticity in two languages (English and Portuguese), presenting a dataset of minimal pairs containing human idiomaticity judgments for each noun compound at both type and token levels, their paraphrases and their occurrences in naturalistic and sense-neutral contexts, totalling 32,200 sentences. We propose this set of minimal pairs for evaluating how well a model captures idiomatic meanings, and define a set of fine-grained metrics of Affinity and Scaled Similarity, to determine how sensitive the models are to perturbations that may lead to changes in idiomaticity. Affinity is a comparative measure of the similarity between an experimental item, a target and a potential distractor, and Scaled Similarity incorporates a rescaling factor to magnify the meaningful similarities within the spaces defined by each specific model. The results obtained with a variety of representative and widely used models indicate that, despite superficial indications to the contrary in the form of high similarities, idiomaticity is not yet accurately represented in current models. Moreover, the performance of models with different levels of contextualization suggests that their ability to capture context is not yet able to go beyond more superficial lexical clues provided by the words and to actually incorporate the relevant semantic clues needed for idiomaticity. By proposing model-agnostic measures for assessing the ability of models to capture idiomaticity, this article contributes to determining limitations in the handling of non-compositional structures, which is one of the directions that needs to be considered for more natural, accurate, and robust language understanding. The source code and additional materials related to this paper are available at our GitHub repository.1</abstract>
<identifier type="citekey">he-etal-2025-investigating</identifier>
<identifier type="doi">10.1162/coli_a_00546</identifier>
<location>
<url>https://aclanthology.org/2025.cl-2.4/</url>
</location>
<part>
<date>2025-06</date>
<detail type="volume"><number>51</number></detail>
<extent unit="page">
<start>505</start>
<end>555</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Investigating Idiomaticity in Word Representations
%A He, Wei
%A Vieira, Tiago Kramer
%A Garcia, Marcos
%A Scarton, Carolina
%A Idiart, Marco
%A Villavicencio, Aline
%J Computational Linguistics
%D 2025
%8 June
%V 51
%I MIT Press
%C Cambridge, MA
%F he-etal-2025-investigating
%X Idiomatic expressions are an integral part of human languages, often used to express complex ideas in compressed or conventional ways (e.g., eager beaver as a keen and enthusiastic person). However, their interpretations may not be straightforwardly linked to the meanings of their individual components in isolation and this may have an impact for compositional approaches. In this article, we investigate to what extent word representation models are able to go beyond compositional word combinations and capture multiword expression idiomaticity and some of the expected properties related to idiomatic meanings. We focus on noun compounds of varying levels of idiomaticity in two languages (English and Portuguese), presenting a dataset of minimal pairs containing human idiomaticity judgments for each noun compound at both type and token levels, their paraphrases and their occurrences in naturalistic and sense-neutral contexts, totalling 32,200 sentences. We propose this set of minimal pairs for evaluating how well a model captures idiomatic meanings, and define a set of fine-grained metrics of Affinity and Scaled Similarity, to determine how sensitive the models are to perturbations that may lead to changes in idiomaticity. Affinity is a comparative measure of the similarity between an experimental item, a target and a potential distractor, and Scaled Similarity incorporates a rescaling factor to magnify the meaningful similarities within the spaces defined by each specific model. The results obtained with a variety of representative and widely used models indicate that, despite superficial indications to the contrary in the form of high similarities, idiomaticity is not yet accurately represented in current models. Moreover, the performance of models with different levels of contextualization suggests that their ability to capture context is not yet able to go beyond more superficial lexical clues provided by the words and to actually incorporate the relevant semantic clues needed for idiomaticity. By proposing model-agnostic measures for assessing the ability of models to capture idiomaticity, this article contributes to determining limitations in the handling of non-compositional structures, which is one of the directions that needs to be considered for more natural, accurate, and robust language understanding. The source code and additional materials related to this paper are available at our GitHub repository.1
%R 10.1162/coli_a_00546
%U https://aclanthology.org/2025.cl-2.4/
%U https://doi.org/10.1162/coli_a_00546
%P 505-555
Markdown (Informal)
[Investigating Idiomaticity in Word Representations](https://aclanthology.org/2025.cl-2.4/) (He et al., CL 2025)
ACL