<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ferrando-etal-2022-measuring">
    <titleInfo>
        <title>Measuring the Mixing of Contextual Information in the Transformer</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Javier</namePart>
        <namePart type="family">Ferrando</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Gerard</namePart>
        <namePart type="given">I</namePart>
        <namePart type="family">Gállego</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Marta</namePart>
        <namePart type="given">R</namePart>
        <namePart type="family">Costa-jussà</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yoav</namePart>
            <namePart type="family">Goldberg</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Zornitsa</namePart>
            <namePart type="family">Kozareva</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The Transformer architecture aggregates input information through the self-attention mechanism, but there is no clear understanding of how this information is mixed across the entire model. Additionally, recent works have demonstrated that attention weights alone are not enough to describe the flow of information. In this paper, we consider the whole attention block –multi-head attention, residual connection, and layer normalization– and define a metric to measure token-to-token interactions within each layer. Then, we aggregate layer-wise interpretations to provide input attribution scores for model predictions. Experimentally, we show that our method, ALTI (Aggregation of Layer-wise Token-to-token Interactions), provides more faithful explanations and increased robustness than gradient-based methods.</abstract>
    <identifier type="citekey">ferrando-etal-2022-measuring</identifier>
    <identifier type="doi">10.18653/v1/2022.emnlp-main.595</identifier>
    <location>
        <url>https://aclanthology.org/2022.emnlp-main.595</url>
    </location>
    <part>
        <date>2022-12</date>
        <extent unit="page">
            <start>8698</start>
            <end>8714</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Measuring the Mixing of Contextual Information in the Transformer
%A Ferrando, Javier
%A Gállego, Gerard I.
%A Costa-jussà, Marta R.
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F ferrando-etal-2022-measuring
%X The Transformer architecture aggregates input information through the self-attention mechanism, but there is no clear understanding of how this information is mixed across the entire model. Additionally, recent works have demonstrated that attention weights alone are not enough to describe the flow of information. In this paper, we consider the whole attention block –multi-head attention, residual connection, and layer normalization– and define a metric to measure token-to-token interactions within each layer. Then, we aggregate layer-wise interpretations to provide input attribution scores for model predictions. Experimentally, we show that our method, ALTI (Aggregation of Layer-wise Token-to-token Interactions), provides more faithful explanations and increased robustness than gradient-based methods.
%R 10.18653/v1/2022.emnlp-main.595
%U https://aclanthology.org/2022.emnlp-main.595
%U https://doi.org/10.18653/v1/2022.emnlp-main.595
%P 8698-8714

Markdown (Informal)

[Measuring the Mixing of Contextual Information in the Transformer](https://aclanthology.org/2022.emnlp-main.595) (Ferrando et al., EMNLP 2022)

• Measuring the Mixing of Contextual Information in the Transformer (Ferrando et al., EMNLP 2022)

ACL

• Javier Ferrando, Gerard I. Gállego, and Marta R. Costa-jussà. 2022. Measuring the Mixing of Contextual Information in the Transformer. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 8698–8714, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.