Anthropogenic Regional Adaptation in Multimodal Vision-Language Model
Samuel Cahyawijaya, Peerat Limkonchotiwat, Tack Hwa Wong, Carlos Rafael Catalan, Manuel Antonio Rufino, Hitesh Laxmichand Patel, Amit Agarwal, Muhammad Reza Qorib, Vicky Feliren, Holy Lovenia, Aye Hninn Khine, Frederikus Hudi, David Anugraha, Alham Fikri Aji, Romrawin Chumpu, Viet Thanh Pham, Minghan Wang, Mohamed Fazli Mohamed Imam, Ruochen Zhang, Joseph Marvin Imperial, Khumaisa Nur’aini, Do Xuan Long, Musa Izzanardi Wijanarko, Joel Ruben Antony Moniz, Patrick Amadeus Irawan, Hanif Muhammad Zhafran, Isaiah Edri W. Flores, Salsabila Zahirah Pranida, Jun Kevin, Jostin Jerico Rosal, Patricia Nicole Monderin, Kun Kerdthaisong, Ahmad Mustafid, My Chiffon Nguyen, Natchapon Jongwiriyanurak, Siva Worajitwannakul, Haochen Li, Adrian Xuan Wei Lim, Bin Wang, Muhammad Ravi Shulthan Habibi, Lynnette Hui Xian Ng, Mithil Bangera, Yeshil Bangera, Priyaranjan Pattnayak, Dun Li Chan, Sherissa Caren Djuniwar, Cho Chan Myei Oo, Ming Shan Hee
Correct Metadata for
Abstract
While the field of vision-language (VL) has achieved remarkable success in integrating visual and textual information across multiple languages and domains, there is no dedicated framework for assessing human-centric alignment in vision-language systems.We offer two contributions to address this gap. First, we introduce Anthropogenic Regional Adaptation: a regional-adaptation framework that aims to optimize model relevance to specific regional contexts while ensuring the retention of global generalization capabilities. Second, we present a simple, but effective adaptation method named Geographical-generalization-made-easy (GG-EZ), which utilizes regional data filtering and model merging.
Through comprehensive ablations on 3 VL architectures: large vision-language models (LVLM), text-to-image diffusion models, and vision-language embedding models, and a case study in Southeast Asia (SEA) regional adaptation, we demonstrate the importance of Anthropogenic Regional Adaptation and GG-EZ, showing 5-15% gains in cultural relevance metrics across SEA while maintaining over 98% of global performance and even occasionally surpassing it.
Our findings establish Anthropogenic Regional Alignment as a foundational paradigm towards applicability of multimodal VL architectures in diverse regions and demonstrate a simple-yet-effective baseline method that optimizes regional value alignment while preserving global generalization.
- Anthology ID:
- 2026.findings-aacl.36
- Volume:
- Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
- Month:
- November
- Year:
- 2026
- Address:
- Hengqin, China
- Editors:
- Yue Zhang, Ruifeng Xu, Ayu Purwarianti, Tong Rong, Jun Suzuki
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 695–712
- Language:
- URL:
- https://aclanthology.org/2026.findings-aacl.36/
- DOI:
- Bibkey:
- Cite (ACL):
- Samuel Cahyawijaya, Peerat Limkonchotiwat, Tack Hwa Wong, Carlos Rafael Catalan, Manuel Antonio Rufino, Hitesh Laxmichand Patel, Amit Agarwal, Muhammad Reza Qorib, Vicky Feliren, Holy Lovenia, Aye Hninn Khine, Frederikus Hudi, David Anugraha, Alham Fikri Aji, Romrawin Chumpu, Viet Thanh Pham, Minghan Wang, Mohamed Fazli Mohamed Imam, Ruochen Zhang, Joseph Marvin Imperial, Khumaisa Nur’aini, Do Xuan Long, Musa Izzanardi Wijanarko, Joel Ruben Antony Moniz, Patrick Amadeus Irawan, Hanif Muhammad Zhafran, Isaiah Edri W. Flores, Salsabila Zahirah Pranida, Jun Kevin, Jostin Jerico Rosal, Patricia Nicole Monderin, Kun Kerdthaisong, Ahmad Mustafid, My Chiffon Nguyen, Natchapon Jongwiriyanurak, Siva Worajitwannakul, Haochen Li, Adrian Xuan Wei Lim, Bin Wang, Muhammad Ravi Shulthan Habibi, Lynnette Hui Xian Ng, Mithil Bangera, Yeshil Bangera, Priyaranjan Pattnayak, Dun Li Chan, Sherissa Caren Djuniwar, Cho Chan Myei Oo, and Ming Shan Hee. 2026. Anthropogenic Regional Adaptation in Multimodal Vision-Language Model. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 695–712, Hengqin, China. Association for Computational Linguistics.
- Cite (Informal):
- Anthropogenic Regional Adaptation in Multimodal Vision-Language Model (Cahyawijaya et al., Findings 2026)
- Copy Citation:
- PDF:
- https://aclanthology.org/2026.findings-aacl.36.pdf
Export citation
@inproceedings{cahyawijaya-etal-2026-anthropogenic,
title = "Anthropogenic Regional Adaptation in Multimodal Vision-Language Model",
author = "Cahyawijaya, Samuel and
Limkonchotiwat, Peerat and
Wong, Tack Hwa and
Catalan, Carlos Rafael and
Rufino, Manuel Antonio and
Patel, Hitesh Laxmichand and
Agarwal, Amit and
Qorib, Muhammad Reza and
Feliren, Vicky and
Lovenia, Holy and
Khine, Aye Hninn and
Hudi, Frederikus and
Anugraha, David and
Aji, Alham Fikri and
Chumpu, Romrawin and
Pham, Viet Thanh and
Wang, Minghan and
Imam, Mohamed Fazli Mohamed and
Zhang, Ruochen and
Imperial, Joseph Marvin and
Nur{'}aini, Khumaisa and
Long, Do Xuan and
Wijanarko, Musa Izzanardi and
Moniz, Joel Ruben Antony and
Irawan, Patrick Amadeus and
Zhafran, Hanif Muhammad and
Flores, Isaiah Edri W. and
Pranida, Salsabila Zahirah and
Kevin, Jun and
Rosal, Jostin Jerico and
Monderin, Patricia Nicole and
Kerdthaisong, Kun and
Mustafid, Ahmad and
Nguyen, My Chiffon and
Jongwiriyanurak, Natchapon and
Worajitwannakul, Siva and
Li, Haochen and
Lim, Adrian Xuan Wei and
Wang, Bin and
Habibi, Muhammad Ravi Shulthan and
Ng, Lynnette Hui Xian and
Bangera, Mithil and
Bangera, Yeshil and
Pattnayak, Priyaranjan and
Chan, Dun Li and
Djuniwar, Sherissa Caren and
Oo, Cho Chan Myei and
Hee, Ming Shan",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {IJCNLP}-{AACL} 2026",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-aacl.36/",
pages = "695--712",
ISBN = "979-8-89176-444-6",
abstract = "While the field of vision-language (VL) has achieved remarkable success in integrating visual and textual information across multiple languages and domains, there is no dedicated framework for assessing human-centric alignment in vision-language systems.\par We offer two contributions to address this gap. First, we introduce Anthropogenic Regional Adaptation: a regional-adaptation framework that aims to optimize model relevance to specific regional contexts while ensuring the retention of global generalization capabilities. Second, we present a simple, but effective adaptation method named Geographical-generalization-made-easy (GG-EZ), which utilizes regional data filtering and model merging.\par Through comprehensive ablations on 3 VL architectures: large vision-language models (LVLM), text-to-image diffusion models, and vision-language embedding models, and a case study in Southeast Asia (SEA) regional adaptation, we demonstrate the importance of Anthropogenic Regional Adaptation and GG-EZ, showing 5-15{\%} gains in cultural relevance metrics across SEA while maintaining over 98{\%} of global performance and even occasionally surpassing it.\par Our findings establish Anthropogenic Regional Alignment as a foundational paradigm towards applicability of multimodal VL architectures in diverse regions and demonstrate a simple-yet-effective baseline method that optimizes regional value alignment while preserving global generalization."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cahyawijaya-etal-2026-anthropogenic">
<titleInfo>
<title>Anthropogenic Regional Adaptation in Multimodal Vision-Language Model</title>
</titleInfo>
<name type="personal">
<namePart type="given">Samuel</namePart>
<namePart type="family">Cahyawijaya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peerat</namePart>
<namePart type="family">Limkonchotiwat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tack</namePart>
<namePart type="given">Hwa</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carlos</namePart>
<namePart type="given">Rafael</namePart>
<namePart type="family">Catalan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manuel</namePart>
<namePart type="given">Antonio</namePart>
<namePart type="family">Rufino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hitesh</namePart>
<namePart type="given">Laxmichand</namePart>
<namePart type="family">Patel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amit</namePart>
<namePart type="family">Agarwal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="given">Reza</namePart>
<namePart type="family">Qorib</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vicky</namePart>
<namePart type="family">Feliren</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Holy</namePart>
<namePart type="family">Lovenia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aye</namePart>
<namePart type="given">Hninn</namePart>
<namePart type="family">Khine</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Frederikus</namePart>
<namePart type="family">Hudi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Anugraha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alham</namePart>
<namePart type="given">Fikri</namePart>
<namePart type="family">Aji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Romrawin</namePart>
<namePart type="family">Chumpu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viet</namePart>
<namePart type="given">Thanh</namePart>
<namePart type="family">Pham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohamed</namePart>
<namePart type="given">Fazli</namePart>
<namePart type="given">Mohamed</namePart>
<namePart type="family">Imam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruochen</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joseph</namePart>
<namePart type="given">Marvin</namePart>
<namePart type="family">Imperial</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khumaisa</namePart>
<namePart type="family">Nur’aini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Do</namePart>
<namePart type="given">Xuan</namePart>
<namePart type="family">Long</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Musa</namePart>
<namePart type="given">Izzanardi</namePart>
<namePart type="family">Wijanarko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joel</namePart>
<namePart type="given">Ruben</namePart>
<namePart type="given">Antony</namePart>
<namePart type="family">Moniz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Patrick</namePart>
<namePart type="given">Amadeus</namePart>
<namePart type="family">Irawan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanif</namePart>
<namePart type="given">Muhammad</namePart>
<namePart type="family">Zhafran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isaiah</namePart>
<namePart type="given">Edri</namePart>
<namePart type="given">W</namePart>
<namePart type="family">Flores</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salsabila</namePart>
<namePart type="given">Zahirah</namePart>
<namePart type="family">Pranida</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Kevin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jostin</namePart>
<namePart type="given">Jerico</namePart>
<namePart type="family">Rosal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Patricia</namePart>
<namePart type="given">Nicole</namePart>
<namePart type="family">Monderin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kun</namePart>
<namePart type="family">Kerdthaisong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ahmad</namePart>
<namePart type="family">Mustafid</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">My</namePart>
<namePart type="given">Chiffon</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Natchapon</namePart>
<namePart type="family">Jongwiriyanurak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siva</namePart>
<namePart type="family">Worajitwannakul</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haochen</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adrian</namePart>
<namePart type="given">Xuan</namePart>
<namePart type="given">Wei</namePart>
<namePart type="family">Lim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bin</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Muhammad</namePart>
<namePart type="given">Ravi</namePart>
<namePart type="given">Shulthan</namePart>
<namePart type="family">Habibi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lynnette</namePart>
<namePart type="given">Hui</namePart>
<namePart type="given">Xian</namePart>
<namePart type="family">Ng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mithil</namePart>
<namePart type="family">Bangera</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yeshil</namePart>
<namePart type="family">Bangera</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Priyaranjan</namePart>
<namePart type="family">Pattnayak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dun</namePart>
<namePart type="given">Li</namePart>
<namePart type="family">Chan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sherissa</namePart>
<namePart type="given">Caren</namePart>
<namePart type="family">Djuniwar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cho</namePart>
<namePart type="given">Chan</namePart>
<namePart type="given">Myei</namePart>
<namePart type="family">Oo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming</namePart>
<namePart type="given">Shan</namePart>
<namePart type="family">Hee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-444-6</identifier>
</relatedItem>
<abstract>While the field of vision-language (VL) has achieved remarkable success in integrating visual and textual information across multiple languages and domains, there is no dedicated framework for assessing human-centric alignment in vision-language systems.\par We offer two contributions to address this gap. First, we introduce Anthropogenic Regional Adaptation: a regional-adaptation framework that aims to optimize model relevance to specific regional contexts while ensuring the retention of global generalization capabilities. Second, we present a simple, but effective adaptation method named Geographical-generalization-made-easy (GG-EZ), which utilizes regional data filtering and model merging.\par Through comprehensive ablations on 3 VL architectures: large vision-language models (LVLM), text-to-image diffusion models, and vision-language embedding models, and a case study in Southeast Asia (SEA) regional adaptation, we demonstrate the importance of Anthropogenic Regional Adaptation and GG-EZ, showing 5-15% gains in cultural relevance metrics across SEA while maintaining over 98% of global performance and even occasionally surpassing it.\par Our findings establish Anthropogenic Regional Alignment as a foundational paradigm towards applicability of multimodal VL architectures in diverse regions and demonstrate a simple-yet-effective baseline method that optimizes regional value alignment while preserving global generalization.</abstract>
<identifier type="citekey">cahyawijaya-etal-2026-anthropogenic</identifier>
<location>
<url>https://aclanthology.org/2026.findings-aacl.36/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>695</start>
<end>712</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings %T Anthropogenic Regional Adaptation in Multimodal Vision-Language Model %A Cahyawijaya, Samuel %A Limkonchotiwat, Peerat %A Wong, Tack Hwa %A Catalan, Carlos Rafael %A Rufino, Manuel Antonio %A Patel, Hitesh Laxmichand %A Agarwal, Amit %A Qorib, Muhammad Reza %A Feliren, Vicky %A Lovenia, Holy %A Khine, Aye Hninn %A Hudi, Frederikus %A Anugraha, David %A Aji, Alham Fikri %A Chumpu, Romrawin %A Pham, Viet Thanh %A Wang, Minghan %A Imam, Mohamed Fazli Mohamed %A Zhang, Ruochen %A Imperial, Joseph Marvin %A Nur’aini, Khumaisa %A Long, Do Xuan %A Wijanarko, Musa Izzanardi %A Moniz, Joel Ruben Antony %A Irawan, Patrick Amadeus %A Zhafran, Hanif Muhammad %A Flores, Isaiah Edri W. %A Pranida, Salsabila Zahirah %A Kevin, Jun %A Rosal, Jostin Jerico %A Monderin, Patricia Nicole %A Kerdthaisong, Kun %A Mustafid, Ahmad %A Nguyen, My Chiffon %A Jongwiriyanurak, Natchapon %A Worajitwannakul, Siva %A Li, Haochen %A Lim, Adrian Xuan Wei %A Wang, Bin %A Habibi, Muhammad Ravi Shulthan %A Ng, Lynnette Hui Xian %A Bangera, Mithil %A Bangera, Yeshil %A Pattnayak, Priyaranjan %A Chan, Dun Li %A Djuniwar, Sherissa Caren %A Oo, Cho Chan Myei %A Hee, Ming Shan %Y Zhang, Yue %Y Xu, Ruifeng %Y Purwarianti, Ayu %Y Rong, Tong %Y Suzuki, Jun %S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026 %D 2026 %8 November %I Association for Computational Linguistics %C Hengqin, China %@ 979-8-89176-444-6 %F cahyawijaya-etal-2026-anthropogenic %X While the field of vision-language (VL) has achieved remarkable success in integrating visual and textual information across multiple languages and domains, there is no dedicated framework for assessing human-centric alignment in vision-language systems.\par We offer two contributions to address this gap. First, we introduce Anthropogenic Regional Adaptation: a regional-adaptation framework that aims to optimize model relevance to specific regional contexts while ensuring the retention of global generalization capabilities. Second, we present a simple, but effective adaptation method named Geographical-generalization-made-easy (GG-EZ), which utilizes regional data filtering and model merging.\par Through comprehensive ablations on 3 VL architectures: large vision-language models (LVLM), text-to-image diffusion models, and vision-language embedding models, and a case study in Southeast Asia (SEA) regional adaptation, we demonstrate the importance of Anthropogenic Regional Adaptation and GG-EZ, showing 5-15% gains in cultural relevance metrics across SEA while maintaining over 98% of global performance and even occasionally surpassing it.\par Our findings establish Anthropogenic Regional Alignment as a foundational paradigm towards applicability of multimodal VL architectures in diverse regions and demonstrate a simple-yet-effective baseline method that optimizes regional value alignment while preserving global generalization. %U https://aclanthology.org/2026.findings-aacl.36/ %P 695-712
Markdown (Informal)
[Anthropogenic Regional Adaptation in Multimodal Vision-Language Model](https://aclanthology.org/2026.findings-aacl.36/) (Cahyawijaya et al., Findings 2026)
- Anthropogenic Regional Adaptation in Multimodal Vision-Language Model (Cahyawijaya et al., Findings 2026)
ACL
- Samuel Cahyawijaya, Peerat Limkonchotiwat, Tack Hwa Wong, Carlos Rafael Catalan, Manuel Antonio Rufino, Hitesh Laxmichand Patel, Amit Agarwal, Muhammad Reza Qorib, Vicky Feliren, Holy Lovenia, Aye Hninn Khine, Frederikus Hudi, David Anugraha, Alham Fikri Aji, Romrawin Chumpu, Viet Thanh Pham, Minghan Wang, Mohamed Fazli Mohamed Imam, Ruochen Zhang, Joseph Marvin Imperial, Khumaisa Nur’aini, Do Xuan Long, Musa Izzanardi Wijanarko, Joel Ruben Antony Moniz, Patrick Amadeus Irawan, Hanif Muhammad Zhafran, Isaiah Edri W. Flores, Salsabila Zahirah Pranida, Jun Kevin, Jostin Jerico Rosal, Patricia Nicole Monderin, Kun Kerdthaisong, Ahmad Mustafid, My Chiffon Nguyen, Natchapon Jongwiriyanurak, Siva Worajitwannakul, Haochen Li, Adrian Xuan Wei Lim, Bin Wang, Muhammad Ravi Shulthan Habibi, Lynnette Hui Xian Ng, Mithil Bangera, Yeshil Bangera, Priyaranjan Pattnayak, Dun Li Chan, Sherissa Caren Djuniwar, Cho Chan Myei Oo, and Ming Shan Hee. 2026. Anthropogenic Regional Adaptation in Multimodal Vision-Language Model. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 695–712, Hengqin, China. Association for Computational Linguistics.