@inproceedings{wen-etal-2025-guideline,
title = "Guideline Compliance in Task-Oriented Dialogue: The Chained Prior Approach",
author = "Wen, Xiangyu and
Zhong, Jianyuan and
Xu, Zhijian and
Xu, Qiang",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-naacl.377/",
doi = "10.18653/v1/2025.findings-naacl.377",
pages = "6750--6776",
ISBN = "979-8-89176-195-7",
abstract = "Task-oriented dialogue (TOD) systems are widely used across various domains, including customer service, appointment scheduling, and technical support. In real-world scenarios, such systems must adhere to given operational guidelines. However, existing solutions based on large language models often cannot achieve strict guideline compliance, even when fine-tuned with domain knowledge. To address this issue, we introduce a novel TOD system named GuidedTOD, which explicitly considers domain-specific guidelines by integrating a policy module. This module employs a Markov Chain, termed Chained Prior, to efficiently encode and dynamically update guideline knowledge. During inference, the Chained Prior re-ranks outputs from the domain-expert language model using beam search, ensuring guideline adherence. Experimental results show that GuidedTOD significantly improves guideline compliance, achieving approximately 20{\%} better action prediction accuracy than state-of-the-art solutions. Code is available here: https://github.com/cure-lab/GuidedTOD."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wen-etal-2025-guideline">
<titleInfo>
<title>Guideline Compliance in Task-Oriented Dialogue: The Chained Prior Approach</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiangyu</namePart>
<namePart type="family">Wen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianyuan</namePart>
<namePart type="family">Zhong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhijian</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qiang</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-195-7</identifier>
</relatedItem>
<abstract>Task-oriented dialogue (TOD) systems are widely used across various domains, including customer service, appointment scheduling, and technical support. In real-world scenarios, such systems must adhere to given operational guidelines. However, existing solutions based on large language models often cannot achieve strict guideline compliance, even when fine-tuned with domain knowledge. To address this issue, we introduce a novel TOD system named GuidedTOD, which explicitly considers domain-specific guidelines by integrating a policy module. This module employs a Markov Chain, termed Chained Prior, to efficiently encode and dynamically update guideline knowledge. During inference, the Chained Prior re-ranks outputs from the domain-expert language model using beam search, ensuring guideline adherence. Experimental results show that GuidedTOD significantly improves guideline compliance, achieving approximately 20% better action prediction accuracy than state-of-the-art solutions. Code is available here: https://github.com/cure-lab/GuidedTOD.</abstract>
<identifier type="citekey">wen-etal-2025-guideline</identifier>
<identifier type="doi">10.18653/v1/2025.findings-naacl.377</identifier>
<location>
<url>https://aclanthology.org/2025.findings-naacl.377/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>6750</start>
<end>6776</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Guideline Compliance in Task-Oriented Dialogue: The Chained Prior Approach
%A Wen, Xiangyu
%A Zhong, Jianyuan
%A Xu, Zhijian
%A Xu, Qiang
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F wen-etal-2025-guideline
%X Task-oriented dialogue (TOD) systems are widely used across various domains, including customer service, appointment scheduling, and technical support. In real-world scenarios, such systems must adhere to given operational guidelines. However, existing solutions based on large language models often cannot achieve strict guideline compliance, even when fine-tuned with domain knowledge. To address this issue, we introduce a novel TOD system named GuidedTOD, which explicitly considers domain-specific guidelines by integrating a policy module. This module employs a Markov Chain, termed Chained Prior, to efficiently encode and dynamically update guideline knowledge. During inference, the Chained Prior re-ranks outputs from the domain-expert language model using beam search, ensuring guideline adherence. Experimental results show that GuidedTOD significantly improves guideline compliance, achieving approximately 20% better action prediction accuracy than state-of-the-art solutions. Code is available here: https://github.com/cure-lab/GuidedTOD.
%R 10.18653/v1/2025.findings-naacl.377
%U https://aclanthology.org/2025.findings-naacl.377/
%U https://doi.org/10.18653/v1/2025.findings-naacl.377
%P 6750-6776
Markdown (Informal)
[Guideline Compliance in Task-Oriented Dialogue: The Chained Prior Approach](https://aclanthology.org/2025.findings-naacl.377/) (Wen et al., Findings 2025)
ACL