@inproceedings{metinov-etal-2026-ca,
title = "{CA}-{C}limate{N}ews: A Human-Annotated Multilingual Benchmark for Climate-Relevance Classification of {C}entral {A}sian News",
author = "Metinov, Adilet and
Uvalieva, Zarina and
Uulu, Bektemir Kumarbai and
Alibekov, Nurtilek and
Tashbaltaev, Tynchtykbek",
editor = "Schimanski, Tobias and
Usmanova, Aida and
Morio, Gaku and
Ni, Jingwei and
Dutia, Kalyan and
Vaghefi, Seyed Saeid and
Stammbach, Dominik and
Leippold, Markus",
booktitle = "Proceedings of the 3rd Workshop on Natural Language Processing Meets Climate Change ({C}limate{NLP} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.climatenlp-1.2/",
pages = "20--28",
ISBN = "979-8-89176-465-1",
abstract = "Climate change is acute in Central Asia (glacier loss, the Aral Sea, and drought exposure), yet there are no NLP resources for climate discourse in the region{'}s low-resource Turkic languages. We present CA-ClimateNews, the first human-annotated multilingual dataset for climate-relevance classification of Central Asian news, covering Kyrgyz, Kazakh, Uzbek, and (regionally grounded) Russian. The corpus comprises 3,140 news items labeled at two levels (binary climate relevance and an 8-way topic category) by two annotators per language, with a 100-item blind double-annotation overlap adjudicated to gold. Inter-annotator agreement is almost perfect (Cohen{'}s $\kappa$ 0.86{--}0.98 on relevance). Because we used transparent LLM-assisted pre-labeling, we additionally run a between-subjects experiment showing that displaying model pre-labels induces anchoring: an independent group shown the pre-label overrode it 3{--}$7\times$ less than a blind group. Our released labels are the blind version. We report baselines: fine-tuned multilingual encoders reach 88{--}96 relevance macro-F1 on the balanced benchmark; a Russian-only model loses up to $\sim$7 points transferring zero-shot to the Turkic languages even at matched training size (least, $\sim$2, for Kazakh), and a related Turkic source is no better, isolating a cross-lingual gap that in-language supervision closes. The dataset, guidelines, and code are publicly released."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="metinov-etal-2026-ca">
<titleInfo>
<title>CA-ClimateNews: A Human-Annotated Multilingual Benchmark for Climate-Relevance Classification of Central Asian News</title>
</titleInfo>
<name type="personal">
<namePart type="given">Adilet</namePart>
<namePart type="family">Metinov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zarina</namePart>
<namePart type="family">Uvalieva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bektemir</namePart>
<namePart type="given">Kumarbai</namePart>
<namePart type="family">Uulu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nurtilek</namePart>
<namePart type="family">Alibekov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tynchtykbek</namePart>
<namePart type="family">Tashbaltaev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd Workshop on Natural Language Processing Meets Climate Change (ClimateNLP 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tobias</namePart>
<namePart type="family">Schimanski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aida</namePart>
<namePart type="family">Usmanova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gaku</namePart>
<namePart type="family">Morio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingwei</namePart>
<namePart type="family">Ni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalyan</namePart>
<namePart type="family">Dutia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seyed</namePart>
<namePart type="given">Saeid</namePart>
<namePart type="family">Vaghefi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dominik</namePart>
<namePart type="family">Stammbach</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Markus</namePart>
<namePart type="family">Leippold</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-465-1</identifier>
</relatedItem>
<abstract>Climate change is acute in Central Asia (glacier loss, the Aral Sea, and drought exposure), yet there are no NLP resources for climate discourse in the region’s low-resource Turkic languages. We present CA-ClimateNews, the first human-annotated multilingual dataset for climate-relevance classification of Central Asian news, covering Kyrgyz, Kazakh, Uzbek, and (regionally grounded) Russian. The corpus comprises 3,140 news items labeled at two levels (binary climate relevance and an 8-way topic category) by two annotators per language, with a 100-item blind double-annotation overlap adjudicated to gold. Inter-annotator agreement is almost perfect (Cohen’s ąppa 0.86–0.98 on relevance). Because we used transparent LLM-assisted pre-labeling, we additionally run a between-subjects experiment showing that displaying model pre-labels induces anchoring: an independent group shown the pre-label overrode it 3–7\times less than a blind group. Our released labels are the blind version. We report baselines: fine-tuned multilingual encoders reach 88–96 relevance macro-F1 on the balanced benchmark; a Russian-only model loses up to \sim7 points transferring zero-shot to the Turkic languages even at matched training size (least, \sim2, for Kazakh), and a related Turkic source is no better, isolating a cross-lingual gap that in-language supervision closes. The dataset, guidelines, and code are publicly released.</abstract>
<identifier type="citekey">metinov-etal-2026-ca</identifier>
<location>
<url>https://aclanthology.org/2026.climatenlp-1.2/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>20</start>
<end>28</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CA-ClimateNews: A Human-Annotated Multilingual Benchmark for Climate-Relevance Classification of Central Asian News
%A Metinov, Adilet
%A Uvalieva, Zarina
%A Uulu, Bektemir Kumarbai
%A Alibekov, Nurtilek
%A Tashbaltaev, Tynchtykbek
%Y Schimanski, Tobias
%Y Usmanova, Aida
%Y Morio, Gaku
%Y Ni, Jingwei
%Y Dutia, Kalyan
%Y Vaghefi, Seyed Saeid
%Y Stammbach, Dominik
%Y Leippold, Markus
%S Proceedings of the 3rd Workshop on Natural Language Processing Meets Climate Change (ClimateNLP 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-465-1
%F metinov-etal-2026-ca
%X Climate change is acute in Central Asia (glacier loss, the Aral Sea, and drought exposure), yet there are no NLP resources for climate discourse in the region’s low-resource Turkic languages. We present CA-ClimateNews, the first human-annotated multilingual dataset for climate-relevance classification of Central Asian news, covering Kyrgyz, Kazakh, Uzbek, and (regionally grounded) Russian. The corpus comprises 3,140 news items labeled at two levels (binary climate relevance and an 8-way topic category) by two annotators per language, with a 100-item blind double-annotation overlap adjudicated to gold. Inter-annotator agreement is almost perfect (Cohen’s ąppa 0.86–0.98 on relevance). Because we used transparent LLM-assisted pre-labeling, we additionally run a between-subjects experiment showing that displaying model pre-labels induces anchoring: an independent group shown the pre-label overrode it 3–7\times less than a blind group. Our released labels are the blind version. We report baselines: fine-tuned multilingual encoders reach 88–96 relevance macro-F1 on the balanced benchmark; a Russian-only model loses up to \sim7 points transferring zero-shot to the Turkic languages even at matched training size (least, \sim2, for Kazakh), and a related Turkic source is no better, isolating a cross-lingual gap that in-language supervision closes. The dataset, guidelines, and code are publicly released.
%U https://aclanthology.org/2026.climatenlp-1.2/
%P 20-28
Markdown (Informal)
[CA-ClimateNews: A Human-Annotated Multilingual Benchmark for Climate-Relevance Classification of Central Asian News](https://aclanthology.org/2026.climatenlp-1.2/) (Metinov et al., ClimateNLP 2026)
ACL