@inproceedings{sreedhar-parisien-2022-prompt,
title = "Prompt Learning for Domain Adaptation in Task-Oriented Dialogue",
author = "Sreedhar, Makesh Narsimhan and
Parisien, Christopher",
editor = "Ou, Zhijian and
Feng, Junlan and
Li, Juanzi",
booktitle = "Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)",
month = dec,
year = "2022",
address = "Abu Dhabi, Beijing (Hybrid)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.seretod-1.4",
doi = "10.18653/v1/2022.seretod-1.4",
pages = "24--30",
abstract = "Conversation designers continue to face significant obstacles when creating productionquality task-oriented dialogue systems. The complexity and cost involved in schema development and data collection is often a major barrier for such designers, limiting their ability to create natural, user-friendly experiences. We frame the classification of user intent as the generation of a canonical form, a lightweight semantic representation using natural language. We show that canonical forms offer a promising alternative to traditional methods for intent classification. By tuning soft prompts for a frozen large language model, we show that canonical forms generalize very well to new, unseen domains in a zero- or few-shot setting. The method is also sample-efficient, reducing the complexity and effort of developing new task-oriented dialogue domains.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sreedhar-parisien-2022-prompt">
<titleInfo>
<title>Prompt Learning for Domain Adaptation in Task-Oriented Dialogue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Makesh</namePart>
<namePart type="given">Narsimhan</namePart>
<namePart type="family">Sreedhar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Parisien</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 Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhijian</namePart>
<namePart type="family">Ou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junlan</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juanzi</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, Beijing (Hybrid)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Conversation designers continue to face significant obstacles when creating productionquality task-oriented dialogue systems. The complexity and cost involved in schema development and data collection is often a major barrier for such designers, limiting their ability to create natural, user-friendly experiences. We frame the classification of user intent as the generation of a canonical form, a lightweight semantic representation using natural language. We show that canonical forms offer a promising alternative to traditional methods for intent classification. By tuning soft prompts for a frozen large language model, we show that canonical forms generalize very well to new, unseen domains in a zero- or few-shot setting. The method is also sample-efficient, reducing the complexity and effort of developing new task-oriented dialogue domains.</abstract>
<identifier type="citekey">sreedhar-parisien-2022-prompt</identifier>
<identifier type="doi">10.18653/v1/2022.seretod-1.4</identifier>
<location>
<url>https://aclanthology.org/2022.seretod-1.4</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>24</start>
<end>30</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Prompt Learning for Domain Adaptation in Task-Oriented Dialogue
%A Sreedhar, Makesh Narsimhan
%A Parisien, Christopher
%Y Ou, Zhijian
%Y Feng, Junlan
%Y Li, Juanzi
%S Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, Beijing (Hybrid)
%F sreedhar-parisien-2022-prompt
%X Conversation designers continue to face significant obstacles when creating productionquality task-oriented dialogue systems. The complexity and cost involved in schema development and data collection is often a major barrier for such designers, limiting their ability to create natural, user-friendly experiences. We frame the classification of user intent as the generation of a canonical form, a lightweight semantic representation using natural language. We show that canonical forms offer a promising alternative to traditional methods for intent classification. By tuning soft prompts for a frozen large language model, we show that canonical forms generalize very well to new, unseen domains in a zero- or few-shot setting. The method is also sample-efficient, reducing the complexity and effort of developing new task-oriented dialogue domains.
%R 10.18653/v1/2022.seretod-1.4
%U https://aclanthology.org/2022.seretod-1.4
%U https://doi.org/10.18653/v1/2022.seretod-1.4
%P 24-30
Markdown (Informal)
[Prompt Learning for Domain Adaptation in Task-Oriented Dialogue](https://aclanthology.org/2022.seretod-1.4) (Sreedhar & Parisien, SereTOD 2022)
ACL