@inproceedings{lee-etal-2022-generative,
title = "Generative Multi-hop Retrieval",
author = "Lee, Hyunji and
Yang, Sohee and
Oh, Hanseok and
Seo, Minjoon",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.92",
doi = "10.18653/v1/2022.emnlp-main.92",
pages = "1417--1436",
abstract = "A common practice for text retrieval is to use an encoder to map the documents and the query to a common vector space and perform a nearest neighbor search (NNS); multi-hop retrieval also often adopts the same paradigm, usually with a modification of iteratively reformulating the query vector so that it can retrieve different documents at each hop. However, such a bi-encoder approach has limitations in multi-hop settings; (1) the reformulated query gets longer as the number of hops increases, which further tightens the embedding bottleneck of the query vector, and (2) it is prone to error propagation. In this paper, we focus on alleviating these limitations in multi-hop settings by formulating the problem in a fully generative way. We propose an encoder-decoder model that performs multi-hop retrieval by simply generating the entire text sequences of the retrieval targets, which means the query and the documents interact in the language model{'}s parametric space rather than L2 or inner product space as in the bi-encoder approach. Our approach, Generative Multi-hop Retrieval (GMR), consistently achieves comparable or higher performance than bi-encoder models in five datasets while demonstrating superior GPU memory and storage footprint.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2022-generative">
<titleInfo>
<title>Generative Multi-hop Retrieval</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hyunji</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sohee</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanseok</namePart>
<namePart type="family">Oh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minjoon</namePart>
<namePart type="family">Seo</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>A common practice for text retrieval is to use an encoder to map the documents and the query to a common vector space and perform a nearest neighbor search (NNS); multi-hop retrieval also often adopts the same paradigm, usually with a modification of iteratively reformulating the query vector so that it can retrieve different documents at each hop. However, such a bi-encoder approach has limitations in multi-hop settings; (1) the reformulated query gets longer as the number of hops increases, which further tightens the embedding bottleneck of the query vector, and (2) it is prone to error propagation. In this paper, we focus on alleviating these limitations in multi-hop settings by formulating the problem in a fully generative way. We propose an encoder-decoder model that performs multi-hop retrieval by simply generating the entire text sequences of the retrieval targets, which means the query and the documents interact in the language model’s parametric space rather than L2 or inner product space as in the bi-encoder approach. Our approach, Generative Multi-hop Retrieval (GMR), consistently achieves comparable or higher performance than bi-encoder models in five datasets while demonstrating superior GPU memory and storage footprint.</abstract>
<identifier type="citekey">lee-etal-2022-generative</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.92</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.92</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>1417</start>
<end>1436</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Generative Multi-hop Retrieval
%A Lee, Hyunji
%A Yang, Sohee
%A Oh, Hanseok
%A Seo, Minjoon
%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 lee-etal-2022-generative
%X A common practice for text retrieval is to use an encoder to map the documents and the query to a common vector space and perform a nearest neighbor search (NNS); multi-hop retrieval also often adopts the same paradigm, usually with a modification of iteratively reformulating the query vector so that it can retrieve different documents at each hop. However, such a bi-encoder approach has limitations in multi-hop settings; (1) the reformulated query gets longer as the number of hops increases, which further tightens the embedding bottleneck of the query vector, and (2) it is prone to error propagation. In this paper, we focus on alleviating these limitations in multi-hop settings by formulating the problem in a fully generative way. We propose an encoder-decoder model that performs multi-hop retrieval by simply generating the entire text sequences of the retrieval targets, which means the query and the documents interact in the language model’s parametric space rather than L2 or inner product space as in the bi-encoder approach. Our approach, Generative Multi-hop Retrieval (GMR), consistently achieves comparable or higher performance than bi-encoder models in five datasets while demonstrating superior GPU memory and storage footprint.
%R 10.18653/v1/2022.emnlp-main.92
%U https://aclanthology.org/2022.emnlp-main.92
%U https://doi.org/10.18653/v1/2022.emnlp-main.92
%P 1417-1436
Markdown (Informal)
[Generative Multi-hop Retrieval](https://aclanthology.org/2022.emnlp-main.92) (Lee et al., EMNLP 2022)
ACL
- Hyunji Lee, Sohee Yang, Hanseok Oh, and Minjoon Seo. 2022. Generative Multi-hop Retrieval. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 1417–1436, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.