<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2023-retromae">
    <titleInfo>
        <title>RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zheng</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shitao</namePart>
        <namePart type="family">Xiao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yingxia</namePart>
        <namePart type="family">Shao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhao</namePart>
        <namePart type="family">Cao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Anna</namePart>
            <namePart type="family">Rogers</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jordan</namePart>
            <namePart type="family">Boyd-Graber</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Naoaki</namePart>
            <namePart type="family">Okazaki</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Toronto, Canada</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>To better support information retrieval tasks such as web search and open-domain question answering, growing effort is made to develop retrieval-oriented language models, e.g., RetroMAE and many others. Most of the existing works focus on improving the semantic representation capability for the contextualized embedding of the [CLS] token. However, recent study shows that the ordinary tokens besides [CLS] may provide extra information, which help to produce a better representation effect. As such, it’s necessary to extend the current methods where all contextualized embeddings can be jointly pre-trained for the retrieval tasks. In this work, we propose a novel pre-training method called Duplex Masked Auto-Encoder, a.k.a. DupMAE. It is designed to improve the quality of semantic representation where all contextualized embeddings of the pre-trained model can be leveraged. It takes advantage of two complementary auto-encoding tasks: one reconstructs the input sentence on top of the [CLS] embedding; the other one predicts the bag-of-words feature of the input sentence based on the ordinary tokens’ embeddings. The two tasks are jointly conducted to train a unified encoder, where the whole contextualized embeddings are aggregated in a compact way to produce the final semantic representation. DupMAE is simple but empirically competitive: it substantially improves the pre-trained model’s representation capability and transferability, where superior retrieval performances can be achieved on popular benchmarks, like MS MARCO and BEIR. We make our code publicly available at https://github.com/staoxiao/RetroMAE.</abstract>
    <identifier type="citekey">liu-etal-2023-retromae</identifier>
    <identifier type="doi">10.18653/v1/2023.acl-long.148</identifier>
    <location>
        <url>https://aclanthology.org/2023.acl-long.148</url>
    </location>
    <part>
        <date>2023-07</date>
        <extent unit="page">
            <start>2635</start>
            <end>2648</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models
%A Liu, Zheng
%A Xiao, Shitao
%A Shao, Yingxia
%A Cao, Zhao
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F liu-etal-2023-retromae
%X To better support information retrieval tasks such as web search and open-domain question answering, growing effort is made to develop retrieval-oriented language models, e.g., RetroMAE and many others. Most of the existing works focus on improving the semantic representation capability for the contextualized embedding of the [CLS] token. However, recent study shows that the ordinary tokens besides [CLS] may provide extra information, which help to produce a better representation effect. As such, it’s necessary to extend the current methods where all contextualized embeddings can be jointly pre-trained for the retrieval tasks. In this work, we propose a novel pre-training method called Duplex Masked Auto-Encoder, a.k.a. DupMAE. It is designed to improve the quality of semantic representation where all contextualized embeddings of the pre-trained model can be leveraged. It takes advantage of two complementary auto-encoding tasks: one reconstructs the input sentence on top of the [CLS] embedding; the other one predicts the bag-of-words feature of the input sentence based on the ordinary tokens’ embeddings. The two tasks are jointly conducted to train a unified encoder, where the whole contextualized embeddings are aggregated in a compact way to produce the final semantic representation. DupMAE is simple but empirically competitive: it substantially improves the pre-trained model’s representation capability and transferability, where superior retrieval performances can be achieved on popular benchmarks, like MS MARCO and BEIR. We make our code publicly available at https://github.com/staoxiao/RetroMAE.
%R 10.18653/v1/2023.acl-long.148
%U https://aclanthology.org/2023.acl-long.148
%U https://doi.org/10.18653/v1/2023.acl-long.148
%P 2635-2648

Markdown (Informal)

[RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models](https://aclanthology.org/2023.acl-long.148) (Liu et al., ACL 2023)

• RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models (Liu et al., ACL 2023)

ACL

• Zheng Liu, Shitao Xiao, Yingxia Shao, and Zhao Cao. 2023. RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2635–2648, Toronto, Canada. Association for Computational Linguistics.