@inproceedings{kim-etal-2024-groundial,
title = "{G}roun{D}ial: Human-norm Grounded Safe Dialog Response Generation",
author = "Kim, Siwon and
Dai, Shuyang and
Kachuee, Mohammad and
Ray, Shayan and
Taghavi, Tara and
Yoon, Sungroh",
editor = "Graham, Yvette and
Purver, Matthew",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2024",
month = mar,
year = "2024",
address = "St. Julian{'}s, Malta",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-eacl.109",
pages = "1582--1588",
abstract = "Current conversational AI systems based on large language models (LLMs) are known to generate unsafe responses agreeing to offensive user input or including toxic content. Previous research aimed to alleviate the toxicity by fine-tuning LLM with manually annotated safe dialogue histories. However, the dependency on additional tuning requires substantial costs. To remove the dependency, we propose GrounDial, where response safety is achieved by grounding responses to commonsense social rules without requiring fine-tuning. A hybrid approach of in-context learning and human-norm-guided decoding of GrounDial enables the response to be quantitatively and qualitatively safer even without additional data or tuning.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kim-etal-2024-groundial">
<titleInfo>
<title>GrounDial: Human-norm Grounded Safe Dialog Response Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Siwon</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuyang</namePart>
<namePart type="family">Dai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="family">Kachuee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shayan</namePart>
<namePart type="family">Ray</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tara</namePart>
<namePart type="family">Taghavi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sungroh</namePart>
<namePart type="family">Yoon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yvette</namePart>
<namePart type="family">Graham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Purver</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">St. Julian’s, Malta</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Current conversational AI systems based on large language models (LLMs) are known to generate unsafe responses agreeing to offensive user input or including toxic content. Previous research aimed to alleviate the toxicity by fine-tuning LLM with manually annotated safe dialogue histories. However, the dependency on additional tuning requires substantial costs. To remove the dependency, we propose GrounDial, where response safety is achieved by grounding responses to commonsense social rules without requiring fine-tuning. A hybrid approach of in-context learning and human-norm-guided decoding of GrounDial enables the response to be quantitatively and qualitatively safer even without additional data or tuning.</abstract>
<identifier type="citekey">kim-etal-2024-groundial</identifier>
<location>
<url>https://aclanthology.org/2024.findings-eacl.109</url>
</location>
<part>
<date>2024-03</date>
<extent unit="page">
<start>1582</start>
<end>1588</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T GrounDial: Human-norm Grounded Safe Dialog Response Generation
%A Kim, Siwon
%A Dai, Shuyang
%A Kachuee, Mohammad
%A Ray, Shayan
%A Taghavi, Tara
%A Yoon, Sungroh
%Y Graham, Yvette
%Y Purver, Matthew
%S Findings of the Association for Computational Linguistics: EACL 2024
%D 2024
%8 March
%I Association for Computational Linguistics
%C St. Julian’s, Malta
%F kim-etal-2024-groundial
%X Current conversational AI systems based on large language models (LLMs) are known to generate unsafe responses agreeing to offensive user input or including toxic content. Previous research aimed to alleviate the toxicity by fine-tuning LLM with manually annotated safe dialogue histories. However, the dependency on additional tuning requires substantial costs. To remove the dependency, we propose GrounDial, where response safety is achieved by grounding responses to commonsense social rules without requiring fine-tuning. A hybrid approach of in-context learning and human-norm-guided decoding of GrounDial enables the response to be quantitatively and qualitatively safer even without additional data or tuning.
%U https://aclanthology.org/2024.findings-eacl.109
%P 1582-1588
Markdown (Informal)
[GrounDial: Human-norm Grounded Safe Dialog Response Generation](https://aclanthology.org/2024.findings-eacl.109) (Kim et al., Findings 2024)
ACL