@inproceedings{daheim-etal-2022-controllable,
title = "Controllable Factuality in Document-Grounded Dialog Systems Using a Noisy Channel Model",
author = "Daheim, Nico and
Thulke, David and
Dugast, Christian and
Ney, Hermann",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.98",
doi = "10.18653/v1/2022.findings-emnlp.98",
pages = "1365--1381",
abstract = "In this work, we present a model for document-grounded response generation in dialog that is decomposed into two components according to Bayes{'} theorem. One component is a traditional ungrounded response generation model and the other component models the reconstruction of the grounding document based on the dialog context and generated response. We propose different approximate decoding schemes and evaluate our approach on multiple open-domain and task-oriented document-grounded dialog datasets. Our experiments show that the model is more factual in terms of automatic factuality metrics than the baseline model. Furthermore, we outline how introducing scaling factors between the components allows for controlling the tradeoff between factuality and fluency in the model output. Finally, we compare our approach to a recently proposed method to control factuality in grounded dialog, CTRL (Rashkin et al., 2021), and show that both approaches can be combined to achieve additional improvements.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="daheim-etal-2022-controllable">
<titleInfo>
<title>Controllable Factuality in Document-Grounded Dialog Systems Using a Noisy Channel Model</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nico</namePart>
<namePart type="family">Daheim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Thulke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christian</namePart>
<namePart type="family">Dugast</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hermann</namePart>
<namePart type="family">Ney</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>Findings of the Association for Computational Linguistics: EMNLP 2022</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>In this work, we present a model for document-grounded response generation in dialog that is decomposed into two components according to Bayes’ theorem. One component is a traditional ungrounded response generation model and the other component models the reconstruction of the grounding document based on the dialog context and generated response. We propose different approximate decoding schemes and evaluate our approach on multiple open-domain and task-oriented document-grounded dialog datasets. Our experiments show that the model is more factual in terms of automatic factuality metrics than the baseline model. Furthermore, we outline how introducing scaling factors between the components allows for controlling the tradeoff between factuality and fluency in the model output. Finally, we compare our approach to a recently proposed method to control factuality in grounded dialog, CTRL (Rashkin et al., 2021), and show that both approaches can be combined to achieve additional improvements.</abstract>
<identifier type="citekey">daheim-etal-2022-controllable</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.98</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.98</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>1365</start>
<end>1381</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Controllable Factuality in Document-Grounded Dialog Systems Using a Noisy Channel Model
%A Daheim, Nico
%A Thulke, David
%A Dugast, Christian
%A Ney, Hermann
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F daheim-etal-2022-controllable
%X In this work, we present a model for document-grounded response generation in dialog that is decomposed into two components according to Bayes’ theorem. One component is a traditional ungrounded response generation model and the other component models the reconstruction of the grounding document based on the dialog context and generated response. We propose different approximate decoding schemes and evaluate our approach on multiple open-domain and task-oriented document-grounded dialog datasets. Our experiments show that the model is more factual in terms of automatic factuality metrics than the baseline model. Furthermore, we outline how introducing scaling factors between the components allows for controlling the tradeoff between factuality and fluency in the model output. Finally, we compare our approach to a recently proposed method to control factuality in grounded dialog, CTRL (Rashkin et al., 2021), and show that both approaches can be combined to achieve additional improvements.
%R 10.18653/v1/2022.findings-emnlp.98
%U https://aclanthology.org/2022.findings-emnlp.98
%U https://doi.org/10.18653/v1/2022.findings-emnlp.98
%P 1365-1381
Markdown (Informal)
[Controllable Factuality in Document-Grounded Dialog Systems Using a Noisy Channel Model](https://aclanthology.org/2022.findings-emnlp.98) (Daheim et al., Findings 2022)
ACL