<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ramnath-etal-2020-towards">
    <titleInfo>
        <title>Towards Interpreting BERT for Reading Comprehension Based QA</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sahana</namePart>
        <namePart type="family">Ramnath</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Preksha</namePart>
        <namePart type="family">Nema</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Deep</namePart>
        <namePart type="family">Sahni</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mitesh</namePart>
        <namePart type="given">M</namePart>
        <namePart type="family">Khapra</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2020-nov</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>BERT and its variants have achieved state-of-the-art performance in various NLP tasks. Since then, various works have been proposed to analyze the linguistic information being captured in BERT. However, the current works do not provide an insight into how BERT is able to achieve near human-level performance on the task of Reading Comprehension based Question Answering. In this work, we attempt to interpret BERT for RCQA. Since BERT layers do not have predefined roles, we define a layer’s role or functionality using Integrated Gradients. Based on the defined roles, we perform a preliminary analysis across all layers. We observed that the initial layers focus on query-passage interaction, whereas later layers focus more on contextual understanding and enhancing the answer prediction. Specifically for quantifier questions (how much/how many), we notice that BERT focuses on confusing words (i.e., on other numerical quantities in the passage) in the later layers, but still manages to predict the answer correctly. The fine-tuning and analysis scripts will be publicly available at https://github.com/iitmnlp/BERT-Analysis-RCQA.</abstract>
    <identifier type="citekey">ramnath-etal-2020-towards</identifier>
    <identifier type="doi">10.18653/v1/2020.emnlp-main.261</identifier>
    <location>
        <url>https://aclanthology.org/2020.emnlp-main.261</url>
    </location>
    <part>
        <date>2020-nov</date>
        <extent unit="page">
            <start>3236</start>
            <end>3242</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Towards Interpreting BERT for Reading Comprehension Based QA
%A Ramnath, Sahana
%A Nema, Preksha
%A Sahni, Deep
%A Khapra, Mitesh M.
%S Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
%D 2020
%8 nov
%I Association for Computational Linguistics
%C Online
%F ramnath-etal-2020-towards
%X BERT and its variants have achieved state-of-the-art performance in various NLP tasks. Since then, various works have been proposed to analyze the linguistic information being captured in BERT. However, the current works do not provide an insight into how BERT is able to achieve near human-level performance on the task of Reading Comprehension based Question Answering. In this work, we attempt to interpret BERT for RCQA. Since BERT layers do not have predefined roles, we define a layer’s role or functionality using Integrated Gradients. Based on the defined roles, we perform a preliminary analysis across all layers. We observed that the initial layers focus on query-passage interaction, whereas later layers focus more on contextual understanding and enhancing the answer prediction. Specifically for quantifier questions (how much/how many), we notice that BERT focuses on confusing words (i.e., on other numerical quantities in the passage) in the later layers, but still manages to predict the answer correctly. The fine-tuning and analysis scripts will be publicly available at https://github.com/iitmnlp/BERT-Analysis-RCQA.
%R 10.18653/v1/2020.emnlp-main.261
%U https://aclanthology.org/2020.emnlp-main.261
%U https://doi.org/10.18653/v1/2020.emnlp-main.261
%P 3236-3242

Markdown (Informal)

[Towards Interpreting BERT for Reading Comprehension Based QA](https://aclanthology.org/2020.emnlp-main.261) (Ramnath et al., EMNLP 2020)

• Towards Interpreting BERT for Reading Comprehension Based QA (Ramnath et al., EMNLP 2020)

ACL

• Sahana Ramnath, Preksha Nema, Deep Sahni, and Mitesh M. Khapra. 2020. Towards Interpreting BERT for Reading Comprehension Based QA. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 3236–3242, Online. Association for Computational Linguistics.