<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="clark-etal-2019-bam">
    <titleInfo>
        <title>BAM! Born-Again Multi-Task Networks for Natural Language Understanding</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Kevin</namePart>
        <namePart type="family">Clark</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Minh-Thang</namePart>
        <namePart type="family">Luong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Urvashi</namePart>
        <namePart type="family">Khandelwal</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Christopher</namePart>
        <namePart type="given">D</namePart>
        <namePart type="family">Manning</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Quoc</namePart>
        <namePart type="given">V</namePart>
        <namePart type="family">Le</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2019-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Anna</namePart>
            <namePart type="family">Korhonen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">David</namePart>
            <namePart type="family">Traum</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Lluís</namePart>
            <namePart type="family">Màrquez</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Florence, Italy</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>It can be challenging to train multi-task neural networks that outperform or even match their single-task counterparts. To help address this, we propose using knowledge distillation where single-task models teach a multi-task model. We enhance this training with teacher annealing, a novel method that gradually transitions the model from distillation to supervised learning, helping the multi-task model surpass its single-task teachers. We evaluate our approach by multi-task fine-tuning BERT on the GLUE benchmark. Our method consistently improves over standard single-task and multi-task training.</abstract>
    <identifier type="citekey">clark-etal-2019-bam</identifier>
    <identifier type="doi">10.18653/v1/P19-1595</identifier>
    <location>
        <url>https://aclanthology.org/P19-1595</url>
    </location>
    <part>
        <date>2019-07</date>
        <extent unit="page">
            <start>5931</start>
            <end>5937</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T BAM! Born-Again Multi-Task Networks for Natural Language Understanding
%A Clark, Kevin
%A Luong, Minh-Thang
%A Khandelwal, Urvashi
%A Manning, Christopher D.
%A Le, Quoc V.
%Y Korhonen, Anna
%Y Traum, David
%Y Màrquez, Lluís
%S Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
%D 2019
%8 July
%I Association for Computational Linguistics
%C Florence, Italy
%F clark-etal-2019-bam
%X It can be challenging to train multi-task neural networks that outperform or even match their single-task counterparts. To help address this, we propose using knowledge distillation where single-task models teach a multi-task model. We enhance this training with teacher annealing, a novel method that gradually transitions the model from distillation to supervised learning, helping the multi-task model surpass its single-task teachers. We evaluate our approach by multi-task fine-tuning BERT on the GLUE benchmark. Our method consistently improves over standard single-task and multi-task training.
%R 10.18653/v1/P19-1595
%U https://aclanthology.org/P19-1595
%U https://doi.org/10.18653/v1/P19-1595
%P 5931-5937

Markdown (Informal)

[BAM! Born-Again Multi-Task Networks for Natural Language Understanding](https://aclanthology.org/P19-1595) (Clark et al., ACL 2019)

• BAM! Born-Again Multi-Task Networks for Natural Language Understanding (Clark et al., ACL 2019)

ACL

• Kevin Clark, Minh-Thang Luong, Urvashi Khandelwal, Christopher D. Manning, and Quoc V. Le. 2019. BAM! Born-Again Multi-Task Networks for Natural Language Understanding. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5931–5937, Florence, Italy. Association for Computational Linguistics.