<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhukova-etal-2025-contrastive">
    <titleInfo>
        <title>Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Anastasia</namePart>
        <namePart type="family">Zhukova</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jonas</namePart>
        <namePart type="family">Luehrs</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Christian</namePart>
        <namePart type="family">Matt</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Bela</namePart>
        <namePart type="family">Gipp</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Saloni</namePart>
            <namePart type="family">Potdar</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Lina</namePart>
            <namePart type="family">Rojas-Barahona</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sebastien</namePart>
            <namePart type="family">Montella</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou (China)</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-333-3</identifier>
    </relatedItem>
    <abstract>Recent trends in NLP utilize knowledge graphs (KGs) to enhance pretrained language models by incorporating additional knowledge from the graph structures to learn domain-specific terminology or relationships between documents that might otherwise be overlooked. This paper explores how SciNCL, a graph-aware neighborhood contrastive learning methodology originally designed for scientific publications, can be applied to the process industry domain, where text logs contain crucial information about daily operations and are often structured as sparse KGs. Our experiments demonstrate that language models fine-tuned with triplets derived from graph embeddings (GE) outperform a state-of-the-art mE5-large text encoder by 9.8-14.3% (5.45-7.96p) on the proprietary process industry text embedding benchmark (PITEB) while having 3 times fewer parameters.</abstract>
    <identifier type="citekey">zhukova-etal-2025-contrastive</identifier>
    <location>
        <url>https://aclanthology.org/2025.emnlp-industry.103/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>1472</start>
            <end>1484</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry
%A Zhukova, Anastasia
%A Luehrs, Jonas
%A Matt, Christian
%A Gipp, Bela
%Y Potdar, Saloni
%Y Rojas-Barahona, Lina
%Y Montella, Sebastien
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou (China)
%@ 979-8-89176-333-3
%F zhukova-etal-2025-contrastive
%X Recent trends in NLP utilize knowledge graphs (KGs) to enhance pretrained language models by incorporating additional knowledge from the graph structures to learn domain-specific terminology or relationships between documents that might otherwise be overlooked. This paper explores how SciNCL, a graph-aware neighborhood contrastive learning methodology originally designed for scientific publications, can be applied to the process industry domain, where text logs contain crucial information about daily operations and are often structured as sparse KGs. Our experiments demonstrate that language models fine-tuned with triplets derived from graph embeddings (GE) outperform a state-of-the-art mE5-large text encoder by 9.8-14.3% (5.45-7.96p) on the proprietary process industry text embedding benchmark (PITEB) while having 3 times fewer parameters.
%U https://aclanthology.org/2025.emnlp-industry.103/
%P 1472-1484

Markdown (Informal)

[Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry](https://aclanthology.org/2025.emnlp-industry.103/) (Zhukova et al., EMNLP 2025)

• Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry (Zhukova et al., EMNLP 2025)

ACL

• Anastasia Zhukova, Jonas Luehrs, Christian Matt, and Bela Gipp. 2025. Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 1472–1484, Suzhou (China). Association for Computational Linguistics.