@inproceedings{komurcu-temel-2026-3k2t,
title = "3{K}2{T} at {MWE}-2026 {A}d{MIR}e 2: {CARIM}{--} Category-Aware Reasoning for Idiomatic Multimodality",
author = {K{\"o}m{\"u}rc{\"u}, Kubilay Ka{\u{g}}an and
Temel, Tugce},
editor = {Ojha, Atul Kr. and
Mititelu, Verginica Barbu and
Constant, Mathieu and
Stoyanova, Ivelina and
Do{\u{g}}ru{\"o}z, A. Seza and
Rademaker, Alexandre},
booktitle = "Proceedings of the 22nd Workshop on Multiword Expressions ({MWE} 2026)",
month = mar,
year = "2026",
address = "Rabat, Marocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.mwe-1.21/",
pages = "160--164",
ISBN = "979-8-89176-363-0",
abstract = "Idiomatic expressions pose a fundamental challenge for multimodal understanding due to their non-compositional semantics, while pretrained vision{--}language models tend to over-rely on literal visual alignments. We address this issue in the context of the AdMIRe 2.0 multimodal idiomatic image ranking task by introducing CARIM (Category-Aware Reasoning for Idiomatic Multimodality), an inference-time framework that injects structured semantic reasoning without end-to-end retraining.Experiments on the official Codabench leaderboard demonstrate that CARIM achieves competitive Top-1 Accuracy and nDCG across multiple languages. Additional post-competition evaluation on the released test annotations further shows that CARIM maintains robust multilingual performance, highlighting the effectiveness of inference-time category-aware reasoning for multimodal idiomatic grounding."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="komurcu-temel-2026-3k2t">
<titleInfo>
<title>3K2T at MWE-2026 AdMIRe 2: CARIM– Category-Aware Reasoning for Idiomatic Multimodality</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kubilay</namePart>
<namePart type="given">Kağan</namePart>
<namePart type="family">Kömürcü</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tugce</namePart>
<namePart type="family">Temel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 22nd Workshop on Multiword Expressions (MWE 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Atul</namePart>
<namePart type="given">Kr.</namePart>
<namePart type="family">Ojha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Verginica</namePart>
<namePart type="given">Barbu</namePart>
<namePart type="family">Mititelu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mathieu</namePart>
<namePart type="family">Constant</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivelina</namePart>
<namePart type="family">Stoyanova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">A</namePart>
<namePart type="given">Seza</namePart>
<namePart type="family">Doğruöz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandre</namePart>
<namePart type="family">Rademaker</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Marocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-363-0</identifier>
</relatedItem>
<abstract>Idiomatic expressions pose a fundamental challenge for multimodal understanding due to their non-compositional semantics, while pretrained vision–language models tend to over-rely on literal visual alignments. We address this issue in the context of the AdMIRe 2.0 multimodal idiomatic image ranking task by introducing CARIM (Category-Aware Reasoning for Idiomatic Multimodality), an inference-time framework that injects structured semantic reasoning without end-to-end retraining.Experiments on the official Codabench leaderboard demonstrate that CARIM achieves competitive Top-1 Accuracy and nDCG across multiple languages. Additional post-competition evaluation on the released test annotations further shows that CARIM maintains robust multilingual performance, highlighting the effectiveness of inference-time category-aware reasoning for multimodal idiomatic grounding.</abstract>
<identifier type="citekey">komurcu-temel-2026-3k2t</identifier>
<location>
<url>https://aclanthology.org/2026.mwe-1.21/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>160</start>
<end>164</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T 3K2T at MWE-2026 AdMIRe 2: CARIM– Category-Aware Reasoning for Idiomatic Multimodality
%A Kömürcü, Kubilay Kağan
%A Temel, Tugce
%Y Ojha, Atul Kr.
%Y Mititelu, Verginica Barbu
%Y Constant, Mathieu
%Y Stoyanova, Ivelina
%Y Doğruöz, A. Seza
%Y Rademaker, Alexandre
%S Proceedings of the 22nd Workshop on Multiword Expressions (MWE 2026)
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Marocco
%@ 979-8-89176-363-0
%F komurcu-temel-2026-3k2t
%X Idiomatic expressions pose a fundamental challenge for multimodal understanding due to their non-compositional semantics, while pretrained vision–language models tend to over-rely on literal visual alignments. We address this issue in the context of the AdMIRe 2.0 multimodal idiomatic image ranking task by introducing CARIM (Category-Aware Reasoning for Idiomatic Multimodality), an inference-time framework that injects structured semantic reasoning without end-to-end retraining.Experiments on the official Codabench leaderboard demonstrate that CARIM achieves competitive Top-1 Accuracy and nDCG across multiple languages. Additional post-competition evaluation on the released test annotations further shows that CARIM maintains robust multilingual performance, highlighting the effectiveness of inference-time category-aware reasoning for multimodal idiomatic grounding.
%U https://aclanthology.org/2026.mwe-1.21/
%P 160-164
Markdown (Informal)
[3K2T at MWE-2026 AdMIRe 2: CARIM– Category-Aware Reasoning for Idiomatic Multimodality](https://aclanthology.org/2026.mwe-1.21/) (Kömürcü & Temel, MWE 2026)
ACL