@inproceedings{bubphasawan-rutherford-2026-mirrored,
title = "Mirrored Frames: {LLM}-Based Narrative Framing Analysis of {T}hai-Cambodian Border Conflict Coverage",
author = "Bubphasawan, Siwakorn and
Rutherford, Attapol",
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.45/",
pages = "871--884",
ISBN = "979-8-89176-444-6",
abstract = "Framing in news media shapes how events, actors, and causal responsibility are interpreted, yet computational framing analysis remains limited for lower-resource languages and fast-evolving international conflicts. In this work, we analyze narrative framing in news coverage of the 2025 Thailand{--}Cambodia border conflict across English, Thai, and Khmer news articles. To support benchmarking for this task, we manually annotate 51 conflict-related articles using a stakeholder-oriented Hero{--}Villain{--}Victim (HVV) framework and source attribution annotations. We benchmark proprietary and open-weight large language models (LLMs), compare original-language and translated input, evaluate salience-focused prompt variants, and test supervised cross-dataset transfer. Original-language inference and explicit article-level salience instructions improve the best micro-averaged F1 from 0.66 to 0.73. Applying this improved pipeline across active-conflict and post-conflict periods suggests largely mirrored partisan framing patterns in Thai and Cambodian national media."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bubphasawan-rutherford-2026-mirrored">
<titleInfo>
<title>Mirrored Frames: LLM-Based Narrative Framing Analysis of Thai-Cambodian Border Conflict Coverage</title>
</titleInfo>
<name type="personal">
<namePart type="given">Siwakorn</namePart>
<namePart type="family">Bubphasawan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Attapol</namePart>
<namePart type="family">Rutherford</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>Framing in news media shapes how events, actors, and causal responsibility are interpreted, yet computational framing analysis remains limited for lower-resource languages and fast-evolving international conflicts. In this work, we analyze narrative framing in news coverage of the 2025 Thailand–Cambodia border conflict across English, Thai, and Khmer news articles. To support benchmarking for this task, we manually annotate 51 conflict-related articles using a stakeholder-oriented Hero–Villain–Victim (HVV) framework and source attribution annotations. We benchmark proprietary and open-weight large language models (LLMs), compare original-language and translated input, evaluate salience-focused prompt variants, and test supervised cross-dataset transfer. Original-language inference and explicit article-level salience instructions improve the best micro-averaged F1 from 0.66 to 0.73. Applying this improved pipeline across active-conflict and post-conflict periods suggests largely mirrored partisan framing patterns in Thai and Cambodian national media.</abstract>
<identifier type="citekey">bubphasawan-rutherford-2026-mirrored</identifier>
<location>
<url>https://aclanthology.org/2026.findings-aacl.45/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>871</start>
<end>884</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Mirrored Frames: LLM-Based Narrative Framing Analysis of Thai-Cambodian Border Conflict Coverage
%A Bubphasawan, Siwakorn
%A Rutherford, Attapol
%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 bubphasawan-rutherford-2026-mirrored
%X Framing in news media shapes how events, actors, and causal responsibility are interpreted, yet computational framing analysis remains limited for lower-resource languages and fast-evolving international conflicts. In this work, we analyze narrative framing in news coverage of the 2025 Thailand–Cambodia border conflict across English, Thai, and Khmer news articles. To support benchmarking for this task, we manually annotate 51 conflict-related articles using a stakeholder-oriented Hero–Villain–Victim (HVV) framework and source attribution annotations. We benchmark proprietary and open-weight large language models (LLMs), compare original-language and translated input, evaluate salience-focused prompt variants, and test supervised cross-dataset transfer. Original-language inference and explicit article-level salience instructions improve the best micro-averaged F1 from 0.66 to 0.73. Applying this improved pipeline across active-conflict and post-conflict periods suggests largely mirrored partisan framing patterns in Thai and Cambodian national media.
%U https://aclanthology.org/2026.findings-aacl.45/
%P 871-884
Markdown (Informal)
[Mirrored Frames: LLM-Based Narrative Framing Analysis of Thai-Cambodian Border Conflict Coverage](https://aclanthology.org/2026.findings-aacl.45/) (Bubphasawan & Rutherford, Findings 2026)
ACL