@inproceedings{kundu-etal-2026-carve,
title = "{C}a{RVE}: Critiquing and Refining Visual Elaborations for Figurative Language Illustrations",
author = "Kundu, Manishit and
Padole, Tejomay Kishor and
Shekhar, Sumit and
Banerjee, Biplab and
Bhattacharyya, Pushpak",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.2122/",
doi = "10.18653/v1/2026.findings-acl.2122",
pages = "42760--42777",
ISBN = "979-8-89176-395-1",
abstract = "Illustrating figurative language remains challenging due to its non-literal semantics, and existing text-to-image frameworks rely heavily on proprietary models or human supervision to achieve adequate alignment. We introduce CaRVE, a lightweight and fully open-source critique-driven framework that employs VLM feedback to refine visual elaborations for figurative image generation. CaRVE bridges the semantic alignment gap even in sub-4B models by correcting visual and conceptual misalignments, reducing over-literalization, and improving robustness to complex figurative expressions. Using only open-source models, CaRVE achieves a 6.49{\%} improvement over prior baselines on intrinsic automatic evaluations and a +0.37 average rank gain in human preference. We further release MetaCaRVE, an enhanced figurative image dataset constructed by refining HAIVMet using CaRVE."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kundu-etal-2026-carve">
<titleInfo>
<title>CaRVE: Critiquing and Refining Visual Elaborations for Figurative Language Illustrations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Manishit</namePart>
<namePart type="family">Kundu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tejomay</namePart>
<namePart type="given">Kishor</namePart>
<namePart type="family">Padole</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sumit</namePart>
<namePart type="family">Shekhar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biplab</namePart>
<namePart type="family">Banerjee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Illustrating figurative language remains challenging due to its non-literal semantics, and existing text-to-image frameworks rely heavily on proprietary models or human supervision to achieve adequate alignment. We introduce CaRVE, a lightweight and fully open-source critique-driven framework that employs VLM feedback to refine visual elaborations for figurative image generation. CaRVE bridges the semantic alignment gap even in sub-4B models by correcting visual and conceptual misalignments, reducing over-literalization, and improving robustness to complex figurative expressions. Using only open-source models, CaRVE achieves a 6.49% improvement over prior baselines on intrinsic automatic evaluations and a +0.37 average rank gain in human preference. We further release MetaCaRVE, an enhanced figurative image dataset constructed by refining HAIVMet using CaRVE.</abstract>
<identifier type="citekey">kundu-etal-2026-carve</identifier>
<identifier type="doi">10.18653/v1/2026.findings-acl.2122</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.2122/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>42760</start>
<end>42777</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CaRVE: Critiquing and Refining Visual Elaborations for Figurative Language Illustrations
%A Kundu, Manishit
%A Padole, Tejomay Kishor
%A Shekhar, Sumit
%A Banerjee, Biplab
%A Bhattacharyya, Pushpak
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F kundu-etal-2026-carve
%X Illustrating figurative language remains challenging due to its non-literal semantics, and existing text-to-image frameworks rely heavily on proprietary models or human supervision to achieve adequate alignment. We introduce CaRVE, a lightweight and fully open-source critique-driven framework that employs VLM feedback to refine visual elaborations for figurative image generation. CaRVE bridges the semantic alignment gap even in sub-4B models by correcting visual and conceptual misalignments, reducing over-literalization, and improving robustness to complex figurative expressions. Using only open-source models, CaRVE achieves a 6.49% improvement over prior baselines on intrinsic automatic evaluations and a +0.37 average rank gain in human preference. We further release MetaCaRVE, an enhanced figurative image dataset constructed by refining HAIVMet using CaRVE.
%R 10.18653/v1/2026.findings-acl.2122
%U https://aclanthology.org/2026.findings-acl.2122/
%U https://doi.org/10.18653/v1/2026.findings-acl.2122
%P 42760-42777
Markdown (Informal)
[CaRVE: Critiquing and Refining Visual Elaborations for Figurative Language Illustrations](https://aclanthology.org/2026.findings-acl.2122/) (Kundu et al., Findings 2026)
ACL