@inproceedings{han-etal-2023-efficient,
title = "Efficient Latent Variable Modeling for Knowledge-Grounded Dialogue Generation",
author = "Han, Gunsoo and
Jo, Daejin and
Nam, Daniel and
Yoon, Eunseop and
Kwon, Taehwan and
Rho, Seungeun and
On, Kyoung-Woon and
Yoo, Chang and
Kim, Sungwoong",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2023",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-emnlp.177",
doi = "10.18653/v1/2023.findings-emnlp.177",
pages = "2683--2702",
abstract = "Knowledge-grounded dialogue generation requires first retrieving appropriate external knowledge based on a conversational context and then generating a response grounded on the retrieved knowledge. In general, these two sequential modules, a knowledge retriever and a response generator, have been separately trained in a supervised manner. However, obtaining intermediate labels of the ground-truth knowledge is expensive, especially in open-domain conversations. Latent variable modeling avoids this need for the labels. In this paper, we propose an efficient algorithm for this latent variable modeling that is able to leverage a large amount of dialogue data. Rather than directly training the complex retriever, we adapt a query generator with an off-the-shelf retriever, and the query generator and response generator are simultaneously trained over the latent variable of query. Moreover, we employ lower bound of the evidence as a training objective and modify it to robustly perform the joint training. Experimental results on diverse knowledge-grounded dialogue datasets show that the proposed algorithm significantly outperforms the supervised learning algorithm even without the use of the annotated knowledge while maintaining efficiency and scalability.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="han-etal-2023-efficient">
<titleInfo>
<title>Efficient Latent Variable Modeling for Knowledge-Grounded Dialogue Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gunsoo</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daejin</namePart>
<namePart type="family">Jo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Nam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eunseop</namePart>
<namePart type="family">Yoon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Taehwan</namePart>
<namePart type="family">Kwon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seungeun</namePart>
<namePart type="family">Rho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kyoung-Woon</namePart>
<namePart type="family">On</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chang</namePart>
<namePart type="family">Yoo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sungwoong</namePart>
<namePart type="family">Kim</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>Findings of the Association for Computational Linguistics: EMNLP 2023</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>Knowledge-grounded dialogue generation requires first retrieving appropriate external knowledge based on a conversational context and then generating a response grounded on the retrieved knowledge. In general, these two sequential modules, a knowledge retriever and a response generator, have been separately trained in a supervised manner. However, obtaining intermediate labels of the ground-truth knowledge is expensive, especially in open-domain conversations. Latent variable modeling avoids this need for the labels. In this paper, we propose an efficient algorithm for this latent variable modeling that is able to leverage a large amount of dialogue data. Rather than directly training the complex retriever, we adapt a query generator with an off-the-shelf retriever, and the query generator and response generator are simultaneously trained over the latent variable of query. Moreover, we employ lower bound of the evidence as a training objective and modify it to robustly perform the joint training. Experimental results on diverse knowledge-grounded dialogue datasets show that the proposed algorithm significantly outperforms the supervised learning algorithm even without the use of the annotated knowledge while maintaining efficiency and scalability.</abstract>
<identifier type="citekey">han-etal-2023-efficient</identifier>
<identifier type="doi">10.18653/v1/2023.findings-emnlp.177</identifier>
<location>
<url>https://aclanthology.org/2023.findings-emnlp.177</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>2683</start>
<end>2702</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Efficient Latent Variable Modeling for Knowledge-Grounded Dialogue Generation
%A Han, Gunsoo
%A Jo, Daejin
%A Nam, Daniel
%A Yoon, Eunseop
%A Kwon, Taehwan
%A Rho, Seungeun
%A On, Kyoung-Woon
%A Yoo, Chang
%A Kim, Sungwoong
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Findings of the Association for Computational Linguistics: EMNLP 2023
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F han-etal-2023-efficient
%X Knowledge-grounded dialogue generation requires first retrieving appropriate external knowledge based on a conversational context and then generating a response grounded on the retrieved knowledge. In general, these two sequential modules, a knowledge retriever and a response generator, have been separately trained in a supervised manner. However, obtaining intermediate labels of the ground-truth knowledge is expensive, especially in open-domain conversations. Latent variable modeling avoids this need for the labels. In this paper, we propose an efficient algorithm for this latent variable modeling that is able to leverage a large amount of dialogue data. Rather than directly training the complex retriever, we adapt a query generator with an off-the-shelf retriever, and the query generator and response generator are simultaneously trained over the latent variable of query. Moreover, we employ lower bound of the evidence as a training objective and modify it to robustly perform the joint training. Experimental results on diverse knowledge-grounded dialogue datasets show that the proposed algorithm significantly outperforms the supervised learning algorithm even without the use of the annotated knowledge while maintaining efficiency and scalability.
%R 10.18653/v1/2023.findings-emnlp.177
%U https://aclanthology.org/2023.findings-emnlp.177
%U https://doi.org/10.18653/v1/2023.findings-emnlp.177
%P 2683-2702
Markdown (Informal)
[Efficient Latent Variable Modeling for Knowledge-Grounded Dialogue Generation](https://aclanthology.org/2023.findings-emnlp.177) (Han et al., Findings 2023)
ACL
- Gunsoo Han, Daejin Jo, Daniel Nam, Eunseop Yoon, Taehwan Kwon, Seungeun Rho, Kyoung-Woon On, Chang Yoo, and Sungwoong Kim. 2023. Efficient Latent Variable Modeling for Knowledge-Grounded Dialogue Generation. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 2683–2702, Singapore. Association for Computational Linguistics.