@inproceedings{chiesurin-etal-2023-dangers,
title = "The Dangers of trusting Stochastic Parrots: Faithfulness and Trust in Open-domain Conversational Question Answering",
author = "Chiesurin, Sabrina and
Dimakopoulos, Dimitris and
Sobrevilla Cabezudo, Marco Antonio and
Eshghi, Arash and
Papaioannou, Ioannis and
Rieser, Verena and
Konstas, Ioannis",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.60",
doi = "10.18653/v1/2023.findings-acl.60",
pages = "947--959",
abstract = "Large language models are known to produce output which sounds fluent and convincing, but is also often wrong, e.g. {``}unfaithful{''} with respect to a rationale as retrieved from a knowledge base. In this paper, we show that task-based systems which exhibit certain advanced linguistic dialog behaviors, such as lexical alignment (repeating what the user said), are in fact preferred and trusted more, whereas other phenomena, such as pronouns and ellipsis are dis-preferred. We use open-domain question answering systems as our test-bed for task based dialog generation and compare several open- and closed-book models. Our results highlight the danger of systems that appear to be trustworthy by parroting user input while providing an unfaithful response.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chiesurin-etal-2023-dangers">
<titleInfo>
<title>The Dangers of trusting Stochastic Parrots: Faithfulness and Trust in Open-domain Conversational Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sabrina</namePart>
<namePart type="family">Chiesurin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dimitris</namePart>
<namePart type="family">Dimakopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="given">Antonio</namePart>
<namePart type="family">Sobrevilla Cabezudo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arash</namePart>
<namePart type="family">Eshghi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ioannis</namePart>
<namePart type="family">Papaioannou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Verena</namePart>
<namePart type="family">Rieser</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ioannis</namePart>
<namePart type="family">Konstas</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>Findings of the Association for Computational Linguistics: ACL 2023</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>Large language models are known to produce output which sounds fluent and convincing, but is also often wrong, e.g. “unfaithful” with respect to a rationale as retrieved from a knowledge base. In this paper, we show that task-based systems which exhibit certain advanced linguistic dialog behaviors, such as lexical alignment (repeating what the user said), are in fact preferred and trusted more, whereas other phenomena, such as pronouns and ellipsis are dis-preferred. We use open-domain question answering systems as our test-bed for task based dialog generation and compare several open- and closed-book models. Our results highlight the danger of systems that appear to be trustworthy by parroting user input while providing an unfaithful response.</abstract>
<identifier type="citekey">chiesurin-etal-2023-dangers</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.60</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.60</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>947</start>
<end>959</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Dangers of trusting Stochastic Parrots: Faithfulness and Trust in Open-domain Conversational Question Answering
%A Chiesurin, Sabrina
%A Dimakopoulos, Dimitris
%A Sobrevilla Cabezudo, Marco Antonio
%A Eshghi, Arash
%A Papaioannou, Ioannis
%A Rieser, Verena
%A Konstas, Ioannis
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F chiesurin-etal-2023-dangers
%X Large language models are known to produce output which sounds fluent and convincing, but is also often wrong, e.g. “unfaithful” with respect to a rationale as retrieved from a knowledge base. In this paper, we show that task-based systems which exhibit certain advanced linguistic dialog behaviors, such as lexical alignment (repeating what the user said), are in fact preferred and trusted more, whereas other phenomena, such as pronouns and ellipsis are dis-preferred. We use open-domain question answering systems as our test-bed for task based dialog generation and compare several open- and closed-book models. Our results highlight the danger of systems that appear to be trustworthy by parroting user input while providing an unfaithful response.
%R 10.18653/v1/2023.findings-acl.60
%U https://aclanthology.org/2023.findings-acl.60
%U https://doi.org/10.18653/v1/2023.findings-acl.60
%P 947-959
Markdown (Informal)
[The Dangers of trusting Stochastic Parrots: Faithfulness and Trust in Open-domain Conversational Question Answering](https://aclanthology.org/2023.findings-acl.60) (Chiesurin et al., Findings 2023)
ACL