@inproceedings{safa-sahin-2025-zero,
title = "A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding",
author = {Safa, Abdulfattah and
{\c{S}}ahin, G{\"o}zde G{\"u}l},
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.387/",
doi = "10.18653/v1/2025.naacl-long.387",
pages = "7562--7579",
ISBN = "979-8-89176-189-6",
abstract = "Dialogue State Tracking (DST) is crucial for understanding user needs and executing appropriate system actions in task-oriented dialogues. Majority of existing DST methods are designed to work within predefined ontologies and assume the availability of gold domain labels, struggling with adapting to new slots values. While Large Language Models (LLMs)-based systems show promising zero-shot DST performance, they either require extensive computational resources or they underperform existing fully-trained systems, limiting their practicality. To address these limitations, we propose a zero-shot, open-vocabulary system that integrates domain classification and DST in a single pipeline. Our approach includes reformulating DST as a question-answering task for less capable models and employing self-refining prompts for more adaptable ones. Our system does not rely on fixed slot values defined in the ontology allowing the system to adapt dynamically. We compare our approach with existing SOTA, and show that it provides up to 20{\%} better Joint Goal Accuracy (JGA) over previous methods on datasets like MultiWOZ 2.1, with up to 90{\%} fewer requests to the LLM API."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="safa-sahin-2025-zero">
<titleInfo>
<title>A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Abdulfattah</namePart>
<namePart type="family">Safa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gözde</namePart>
<namePart type="given">Gül</namePart>
<namePart type="family">Şahin</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>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</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">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>Dialogue State Tracking (DST) is crucial for understanding user needs and executing appropriate system actions in task-oriented dialogues. Majority of existing DST methods are designed to work within predefined ontologies and assume the availability of gold domain labels, struggling with adapting to new slots values. While Large Language Models (LLMs)-based systems show promising zero-shot DST performance, they either require extensive computational resources or they underperform existing fully-trained systems, limiting their practicality. To address these limitations, we propose a zero-shot, open-vocabulary system that integrates domain classification and DST in a single pipeline. Our approach includes reformulating DST as a question-answering task for less capable models and employing self-refining prompts for more adaptable ones. Our system does not rely on fixed slot values defined in the ontology allowing the system to adapt dynamically. We compare our approach with existing SOTA, and show that it provides up to 20% better Joint Goal Accuracy (JGA) over previous methods on datasets like MultiWOZ 2.1, with up to 90% fewer requests to the LLM API.</abstract>
<identifier type="citekey">safa-sahin-2025-zero</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.387</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.387/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>7562</start>
<end>7579</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding
%A Safa, Abdulfattah
%A Şahin, Gözde Gül
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F safa-sahin-2025-zero
%X Dialogue State Tracking (DST) is crucial for understanding user needs and executing appropriate system actions in task-oriented dialogues. Majority of existing DST methods are designed to work within predefined ontologies and assume the availability of gold domain labels, struggling with adapting to new slots values. While Large Language Models (LLMs)-based systems show promising zero-shot DST performance, they either require extensive computational resources or they underperform existing fully-trained systems, limiting their practicality. To address these limitations, we propose a zero-shot, open-vocabulary system that integrates domain classification and DST in a single pipeline. Our approach includes reformulating DST as a question-answering task for less capable models and employing self-refining prompts for more adaptable ones. Our system does not rely on fixed slot values defined in the ontology allowing the system to adapt dynamically. We compare our approach with existing SOTA, and show that it provides up to 20% better Joint Goal Accuracy (JGA) over previous methods on datasets like MultiWOZ 2.1, with up to 90% fewer requests to the LLM API.
%R 10.18653/v1/2025.naacl-long.387
%U https://aclanthology.org/2025.naacl-long.387/
%U https://doi.org/10.18653/v1/2025.naacl-long.387
%P 7562-7579
Markdown (Informal)
[A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding](https://aclanthology.org/2025.naacl-long.387/) (Safa & Şahin, NAACL 2025)
ACL
- Abdulfattah Safa and Gözde Gül Şahin. 2025. A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 7562–7579, Albuquerque, New Mexico. Association for Computational Linguistics.