<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="volkov-2026-trust">
    <titleInfo>
        <title>Before You Trust the Count: Auditing Multilingual Sustainable Development Goal News</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sergey</namePart>
        <namePart type="family">Volkov</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 Fifth Workshop on NLP for Positive Impact (NLP4PI)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Katherine</namePart>
            <namePart type="family">Atwell</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Angana</namePart>
            <namePart type="family">Borah</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Daryna</namePart>
            <namePart type="family">Dementieva</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Elisa</namePart>
            <namePart type="family">Kreiss</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Neema</namePart>
            <namePart type="family">Kotonya</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jiarui</namePart>
            <namePart type="family">Liu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Liz</namePart>
            <namePart type="family">Olson</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ruyuan</namePart>
            <namePart type="family">Wan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jieyu</namePart>
            <namePart type="family">Zhao</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">979-8-89176-495-8</identifier>
    </relatedItem>
    <abstract>News monitoring of the Sustainable Development Goals (SDGs) can be distorted by a missing archive interval, a broad lexical rule, or duplicated evidence. We study how to audit such conclusions when target-domain relevance labels are unavailable. Our case comprises 16,060,006 URL-unique candidates retrieved from English, Spanish, Russian, and Chinese GDELT Article List (GAL) titles observed in 2024–2025. We replace whole-day gap accounting with a minute-interval ledger, correct public-label processing to distinguish accepted, rejected, and unassessed text–goal pairs, and compare weak signals on source-grouped public benchmarks. We find that weak fusion does not satisfy our predefined retention rule. We therefore use a claim-level audit with three outcomes: supported under tested assumptions, reversed by a specific measurement choice, or blocked by missing evidence. Paired translated headlines reveal disagreement between language-specific instruments. Controlled engineering and semantic streams let us measure false releases, retained correct claims, and release rate against separate references. The audit lowers false-release rates on semantic tracks but abstains too often on the engineering track. The released artifact makes these tradeoffs reproducible. Agreement across specifications shows robustness to the tested checks; it does not prove semantic correctness or accuracy on unlabeled news titles.</abstract>
    <identifier type="citekey">volkov-2026-trust</identifier>
    <location>
        <url>https://aclanthology.org/2026.nlp4pi-1.7/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>67</start>
            <end>76</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Before You Trust the Count: Auditing Multilingual Sustainable Development Goal News
%A Volkov, Sergey
%Y Atwell, Katherine
%Y Borah, Angana
%Y Dementieva, Daryna
%Y Kreiss, Elisa
%Y Kotonya, Neema
%Y Liu, Jiarui
%Y Olson, Liz
%Y Wan, Ruyuan
%Y Zhao, Jieyu
%S Proceedings of the Fifth Workshop on NLP for Positive Impact (NLP4PI)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-495-8
%F volkov-2026-trust
%X News monitoring of the Sustainable Development Goals (SDGs) can be distorted by a missing archive interval, a broad lexical rule, or duplicated evidence. We study how to audit such conclusions when target-domain relevance labels are unavailable. Our case comprises 16,060,006 URL-unique candidates retrieved from English, Spanish, Russian, and Chinese GDELT Article List (GAL) titles observed in 2024–2025. We replace whole-day gap accounting with a minute-interval ledger, correct public-label processing to distinguish accepted, rejected, and unassessed text–goal pairs, and compare weak signals on source-grouped public benchmarks. We find that weak fusion does not satisfy our predefined retention rule. We therefore use a claim-level audit with three outcomes: supported under tested assumptions, reversed by a specific measurement choice, or blocked by missing evidence. Paired translated headlines reveal disagreement between language-specific instruments. Controlled engineering and semantic streams let us measure false releases, retained correct claims, and release rate against separate references. The audit lowers false-release rates on semantic tracks but abstains too often on the engineering track. The released artifact makes these tradeoffs reproducible. Agreement across specifications shows robustness to the tested checks; it does not prove semantic correctness or accuracy on unlabeled news titles.
%U https://aclanthology.org/2026.nlp4pi-1.7/
%P 67-76

Markdown (Informal)

[Before You Trust the Count: Auditing Multilingual Sustainable Development Goal News](https://aclanthology.org/2026.nlp4pi-1.7/) (Volkov, NLP4PI 2026)

• Before You Trust the Count: Auditing Multilingual Sustainable Development Goal News (Volkov, NLP4PI 2026)

ACL

• Sergey Volkov. 2026. Before You Trust the Count: Auditing Multilingual Sustainable Development Goal News. In Proceedings of the Fifth Workshop on NLP for Positive Impact (NLP4PI), pages 67–76, Budapest, Hungary. Association for Computational Linguistics.