<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="berchansky-etal-2023-optimizing">
    <titleInfo>
        <title>Optimizing Retrieval-augmented Reader Models via Token Elimination</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Moshe</namePart>
        <namePart type="family">Berchansky</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Peter</namePart>
        <namePart type="family">Izsak</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Avi</namePart>
        <namePart type="family">Caciularu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ido</namePart>
        <namePart type="family">Dagan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Moshe</namePart>
        <namePart type="family">Wasserblat</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Houda</namePart>
            <namePart type="family">Bouamor</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Juan</namePart>
            <namePart type="family">Pino</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kalika</namePart>
            <namePart type="family">Bali</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Singapore</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Fusion-in-Decoder (FiD) is an effective retrieval-augmented language model applied across a variety of open-domain tasks, such as question answering, fact checking, etc. In FiD, supporting passages are first retrieved and then processed using a generative model (Reader), which can cause a significant bottleneck in decoding time, particularly with long outputs. In this work, we analyze the contribution and necessity of all the retrieved passages to the performance of reader models, and propose eliminating some of the retrieved information, at the token level, that might not contribute essential information to the answer generation process. We demonstrate that our method can reduce run-time by up to 62.2%, with only a 2% reduction in performance, and in some cases, even improve the performance results.</abstract>
    <identifier type="citekey">berchansky-etal-2023-optimizing</identifier>
    <identifier type="doi">10.18653/v1/2023.emnlp-main.93</identifier>
    <location>
        <url>https://aclanthology.org/2023.emnlp-main.93</url>
    </location>
    <part>
        <date>2023-12</date>
        <extent unit="page">
            <start>1506</start>
            <end>1524</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Optimizing Retrieval-augmented Reader Models via Token Elimination
%A Berchansky, Moshe
%A Izsak, Peter
%A Caciularu, Avi
%A Dagan, Ido
%A Wasserblat, Moshe
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F berchansky-etal-2023-optimizing
%X Fusion-in-Decoder (FiD) is an effective retrieval-augmented language model applied across a variety of open-domain tasks, such as question answering, fact checking, etc. In FiD, supporting passages are first retrieved and then processed using a generative model (Reader), which can cause a significant bottleneck in decoding time, particularly with long outputs. In this work, we analyze the contribution and necessity of all the retrieved passages to the performance of reader models, and propose eliminating some of the retrieved information, at the token level, that might not contribute essential information to the answer generation process. We demonstrate that our method can reduce run-time by up to 62.2%, with only a 2% reduction in performance, and in some cases, even improve the performance results.
%R 10.18653/v1/2023.emnlp-main.93
%U https://aclanthology.org/2023.emnlp-main.93
%U https://doi.org/10.18653/v1/2023.emnlp-main.93
%P 1506-1524

Markdown (Informal)

[Optimizing Retrieval-augmented Reader Models via Token Elimination](https://aclanthology.org/2023.emnlp-main.93) (Berchansky et al., EMNLP 2023)

• Optimizing Retrieval-augmented Reader Models via Token Elimination (Berchansky et al., EMNLP 2023)

ACL

• Moshe Berchansky, Peter Izsak, Avi Caciularu, Ido Dagan, and Moshe Wasserblat. 2023. Optimizing Retrieval-augmented Reader Models via Token Elimination. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 1506–1524, Singapore. Association for Computational Linguistics.