<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gao-callan-2021-condenser">
    <titleInfo>
        <title>Condenser: a Pre-training Architecture for Dense Retrieval</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Luyu</namePart>
        <namePart type="family">Gao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jamie</namePart>
        <namePart type="family">Callan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021-nov</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online and Punta Cana, Dominican Republic</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Pre-trained Transformer language models (LM) have become go-to text representation encoders. Prior research fine-tunes deep LMs to encode text sequences such as sentences and passages into single dense vector representations for efficient text comparison and retrieval. However, dense encoders require a lot of data and sophisticated techniques to effectively train and suffer in low data situations. This paper finds a key reason is that standard LMs’ internal attention structure is not ready-to-use for dense encoders, which needs to aggregate text information into the dense representation. We propose to pre-train towards dense encoder with a novel Transformer architecture, Condenser, where LM prediction CONditions on DENSE Representation. Our experiments show Condenser improves over standard LM by large margins on various text retrieval and similarity tasks.</abstract>
    <identifier type="citekey">gao-callan-2021-condenser</identifier>
    <identifier type="doi">10.18653/v1/2021.emnlp-main.75</identifier>
    <location>
        <url>https://aclanthology.org/2021.emnlp-main.75</url>
    </location>
    <part>
        <date>2021-nov</date>
        <extent unit="page">
            <start>981</start>
            <end>993</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Condenser: a Pre-training Architecture for Dense Retrieval
%A Gao, Luyu
%A Callan, Jamie
%S Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
%D 2021
%8 nov
%I Association for Computational Linguistics
%C Online and Punta Cana, Dominican Republic
%F gao-callan-2021-condenser
%X Pre-trained Transformer language models (LM) have become go-to text representation encoders. Prior research fine-tunes deep LMs to encode text sequences such as sentences and passages into single dense vector representations for efficient text comparison and retrieval. However, dense encoders require a lot of data and sophisticated techniques to effectively train and suffer in low data situations. This paper finds a key reason is that standard LMs’ internal attention structure is not ready-to-use for dense encoders, which needs to aggregate text information into the dense representation. We propose to pre-train towards dense encoder with a novel Transformer architecture, Condenser, where LM prediction CONditions on DENSE Representation. Our experiments show Condenser improves over standard LM by large margins on various text retrieval and similarity tasks.
%R 10.18653/v1/2021.emnlp-main.75
%U https://aclanthology.org/2021.emnlp-main.75
%U https://doi.org/10.18653/v1/2021.emnlp-main.75
%P 981-993

Markdown (Informal)

[Condenser: a Pre-training Architecture for Dense Retrieval](https://aclanthology.org/2021.emnlp-main.75) (Gao & Callan, EMNLP 2021)

• Condenser: a Pre-training Architecture for Dense Retrieval (Gao & Callan, EMNLP 2021)

ACL

• Luyu Gao and Jamie Callan. 2021. Condenser: a Pre-training Architecture for Dense Retrieval. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 981–993, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.