<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="schuster-etal-2021-consistent">
    <titleInfo>
        <title>Consistent Accelerated Inference via Confident Adaptive Transformers</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Tal</namePart>
        <namePart type="family">Schuster</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Adam</namePart>
        <namePart type="family">Fisch</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tommi</namePart>
        <namePart type="family">Jaakkola</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Regina</namePart>
        <namePart type="family">Barzilay</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>Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing</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">Online and Punta Cana, Dominican Republic</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>We develop a novel approach for confidently accelerating inference in the large and expensive multilayer Transformers that are now ubiquitous in natural language processing (NLP). Amortized or approximate computational methods increase efficiency, but can come with unpredictable performance costs. In this work, we present CATs – Confident Adaptive Transformers – in which we simultaneously increase computational efficiency, while guaranteeing a specifiable degree of consistency with the original model with high confidence. Our method trains additional prediction heads on top of intermediate layers, and dynamically decides when to stop allocating computational effort to each input using a meta consistency classifier. To calibrate our early prediction stopping rule, we formulate a unique extension of conformal prediction. We demonstrate the effectiveness of this approach on four classification and regression tasks.</abstract>
    <identifier type="citekey">schuster-etal-2021-consistent</identifier>
    <identifier type="doi">10.18653/v1/2021.emnlp-main.406</identifier>
    <location>
        <url>https://aclanthology.org/2021.emnlp-main.406</url>
    </location>
    <part>
        <date>2021-11</date>
        <extent unit="page">
            <start>4962</start>
            <end>4979</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Consistent Accelerated Inference via Confident Adaptive Transformers
%A Schuster, Tal
%A Fisch, Adam
%A Jaakkola, Tommi
%A Barzilay, Regina
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
%D 2021
%8 November
%I Association for Computational Linguistics
%C Online and Punta Cana, Dominican Republic
%F schuster-etal-2021-consistent
%X We develop a novel approach for confidently accelerating inference in the large and expensive multilayer Transformers that are now ubiquitous in natural language processing (NLP). Amortized or approximate computational methods increase efficiency, but can come with unpredictable performance costs. In this work, we present CATs – Confident Adaptive Transformers – in which we simultaneously increase computational efficiency, while guaranteeing a specifiable degree of consistency with the original model with high confidence. Our method trains additional prediction heads on top of intermediate layers, and dynamically decides when to stop allocating computational effort to each input using a meta consistency classifier. To calibrate our early prediction stopping rule, we formulate a unique extension of conformal prediction. We demonstrate the effectiveness of this approach on four classification and regression tasks.
%R 10.18653/v1/2021.emnlp-main.406
%U https://aclanthology.org/2021.emnlp-main.406
%U https://doi.org/10.18653/v1/2021.emnlp-main.406
%P 4962-4979

Markdown (Informal)

[Consistent Accelerated Inference via Confident Adaptive Transformers](https://aclanthology.org/2021.emnlp-main.406) (Schuster et al., EMNLP 2021)

• Consistent Accelerated Inference via Confident Adaptive Transformers (Schuster et al., EMNLP 2021)

ACL

• Tal Schuster, Adam Fisch, Tommi Jaakkola, and Regina Barzilay. 2021. Consistent Accelerated Inference via Confident Adaptive Transformers. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 4962–4979, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.