<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lu-etal-2021-parameter-efficient">
    <titleInfo>
        <title>Parameter-Efficient Domain Knowledge Integration from Multiple Sources for Biomedical Pre-trained Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Qiuhao</namePart>
        <namePart type="family">Lu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dejing</namePart>
        <namePart type="family">Dou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Thien</namePart>
        <namePart type="given">Huu</namePart>
        <namePart type="family">Nguyen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2021</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Marie-Francine</namePart>
            <namePart type="family">Moens</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Xuanjing</namePart>
            <namePart type="family">Huang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Lucia</namePart>
            <namePart type="family">Specia</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Scott</namePart>
            <namePart type="given">Wen-tau</namePart>
            <namePart type="family">Yih</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Punta Cana, Dominican Republic</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Domain-specific pre-trained language models (PLMs) have achieved great success over various downstream tasks in different domains. However, existing domain-specific PLMs mostly rely on self-supervised learning over large amounts of domain text, without explicitly integrating domain-specific knowledge, which can be essential in many domains. Moreover, in knowledge-sensitive areas such as the biomedical domain, knowledge is stored in multiple sources and formats, and existing biomedical PLMs either neglect them or utilize them in a limited manner. In this work, we introduce an architecture to integrate domain knowledge from diverse sources into PLMs in a parameter-efficient way. More specifically, we propose to encode domain knowledge via adapters, which are small bottleneck feed-forward networks inserted between intermediate transformer layers in PLMs. These knowledge adapters are pre-trained for individual domain knowledge sources and integrated via an attention-based knowledge controller to enrich PLMs. Taking the biomedical domain as a case study, we explore three knowledge-specific adapters for PLMs based on the UMLS Metathesaurus graph, the Wikipedia articles for diseases, and the semantic grouping information for biomedical concepts. Extensive experiments on different biomedical NLP tasks and datasets demonstrate the benefits of the proposed architecture and the knowledge-specific adapters across multiple PLMs.</abstract>
    <identifier type="citekey">lu-etal-2021-parameter-efficient</identifier>
    <identifier type="doi">10.18653/v1/2021.findings-emnlp.325</identifier>
    <location>
        <url>https://aclanthology.org/2021.findings-emnlp.325</url>
    </location>
    <part>
        <date>2021-11</date>
        <extent unit="page">
            <start>3855</start>
            <end>3865</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Parameter-Efficient Domain Knowledge Integration from Multiple Sources for Biomedical Pre-trained Language Models
%A Lu, Qiuhao
%A Dou, Dejing
%A Nguyen, Thien Huu
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Findings of the Association for Computational Linguistics: EMNLP 2021
%D 2021
%8 November
%I Association for Computational Linguistics
%C Punta Cana, Dominican Republic
%F lu-etal-2021-parameter-efficient
%X Domain-specific pre-trained language models (PLMs) have achieved great success over various downstream tasks in different domains. However, existing domain-specific PLMs mostly rely on self-supervised learning over large amounts of domain text, without explicitly integrating domain-specific knowledge, which can be essential in many domains. Moreover, in knowledge-sensitive areas such as the biomedical domain, knowledge is stored in multiple sources and formats, and existing biomedical PLMs either neglect them or utilize them in a limited manner. In this work, we introduce an architecture to integrate domain knowledge from diverse sources into PLMs in a parameter-efficient way. More specifically, we propose to encode domain knowledge via adapters, which are small bottleneck feed-forward networks inserted between intermediate transformer layers in PLMs. These knowledge adapters are pre-trained for individual domain knowledge sources and integrated via an attention-based knowledge controller to enrich PLMs. Taking the biomedical domain as a case study, we explore three knowledge-specific adapters for PLMs based on the UMLS Metathesaurus graph, the Wikipedia articles for diseases, and the semantic grouping information for biomedical concepts. Extensive experiments on different biomedical NLP tasks and datasets demonstrate the benefits of the proposed architecture and the knowledge-specific adapters across multiple PLMs.
%R 10.18653/v1/2021.findings-emnlp.325
%U https://aclanthology.org/2021.findings-emnlp.325
%U https://doi.org/10.18653/v1/2021.findings-emnlp.325
%P 3855-3865

Markdown (Informal)

[Parameter-Efficient Domain Knowledge Integration from Multiple Sources for Biomedical Pre-trained Language Models](https://aclanthology.org/2021.findings-emnlp.325) (Lu et al., Findings 2021)

• Parameter-Efficient Domain Knowledge Integration from Multiple Sources for Biomedical Pre-trained Language Models (Lu et al., Findings 2021)

ACL

• Qiuhao Lu, Dejing Dou, and Thien Huu Nguyen. 2021. Parameter-Efficient Domain Knowledge Integration from Multiple Sources for Biomedical Pre-trained Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 3855–3865, Punta Cana, Dominican Republic. Association for Computational Linguistics.