<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lin-byrne-2022-retrieval">
    <titleInfo>
        <title>Retrieval Augmented Visual Question Answering with Outside Knowledge</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Weizhe</namePart>
        <namePart type="family">Lin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Bill</namePart>
        <namePart type="family">Byrne</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>Outside-Knowledge Visual Question Answering (OK-VQA) is a challenging VQA task that requires retrieval of external knowledge to answer questions about images. Recent OK-VQA systems use Dense Passage Retrieval (DPR) to retrieve documents from external knowledge bases, such as Wikipedia, but with DPR trained separately from answer generation, introducing a potential limit on the overall system performance. Instead, we propose a joint training scheme which includes differentiable DPR integrated with answer generation so that the system can be trained in an end-to-end fashion. Our experiments show that our scheme outperforms recent OK-VQA systems with strong DPR for retrieval. We also introduce new diagnostic metrics to analyze how retrieval and generation interact. The strong retrieval ability of our model significantly reduces the number of retrieved documents needed in training, yielding significant benefits in answer quality and computation required for training.</abstract>
    <identifier type="citekey">lin-byrne-2022-retrieval</identifier>
    <identifier type="doi">10.18653/v1/2022.emnlp-main.772</identifier>
    <location>
        <url>https://aclanthology.org/2022.emnlp-main.772</url>
    </location>
    <part>
        <date>2022-12</date>
        <extent unit="page">
            <start>11238</start>
            <end>11254</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Retrieval Augmented Visual Question Answering with Outside Knowledge
%A Lin, Weizhe
%A Byrne, Bill
%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 lin-byrne-2022-retrieval
%X Outside-Knowledge Visual Question Answering (OK-VQA) is a challenging VQA task that requires retrieval of external knowledge to answer questions about images. Recent OK-VQA systems use Dense Passage Retrieval (DPR) to retrieve documents from external knowledge bases, such as Wikipedia, but with DPR trained separately from answer generation, introducing a potential limit on the overall system performance. Instead, we propose a joint training scheme which includes differentiable DPR integrated with answer generation so that the system can be trained in an end-to-end fashion. Our experiments show that our scheme outperforms recent OK-VQA systems with strong DPR for retrieval. We also introduce new diagnostic metrics to analyze how retrieval and generation interact. The strong retrieval ability of our model significantly reduces the number of retrieved documents needed in training, yielding significant benefits in answer quality and computation required for training.
%R 10.18653/v1/2022.emnlp-main.772
%U https://aclanthology.org/2022.emnlp-main.772
%U https://doi.org/10.18653/v1/2022.emnlp-main.772
%P 11238-11254

Markdown (Informal)

[Retrieval Augmented Visual Question Answering with Outside Knowledge](https://aclanthology.org/2022.emnlp-main.772) (Lin & Byrne, EMNLP 2022)

• Retrieval Augmented Visual Question Answering with Outside Knowledge (Lin & Byrne, EMNLP 2022)

ACL

• Weizhe Lin and Bill Byrne. 2022. Retrieval Augmented Visual Question Answering with Outside Knowledge. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 11238–11254, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.