@inproceedings{ke-etal-2025-catch,
title = "{CATCH}: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation",
author = "Ke, Rui and
Xu, Jiahui and
Wang, Kuang and
Yang, Shenghao and
Jiang, Feng and
Li, Haizhou",
editor = "Hedayatnia, Behnam and
Chen, Vivian and
Chen, Zhang and
Gupta, Raghav and
Galley, Michel",
booktitle = "Proceedings of the Twelfth Dialog System Technology Challenge",
month = aug,
year = "2025",
address = "Avignon, France",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.dstc-1.2/",
pages = "17--26",
ISBN = "979-8-89176-330-2",
abstract = "Theme detection is a fundamental task in user-centric dialogue systems, aiming to identify the latent topic of each utterance without relying on predefined schemas. Unlike intent induction, which operates within fixed label spaces, theme detection requires cross-dialogue consistency and alignment with personalized user preferences, posing significant challenges. Existing methods often struggle with sparse, short utterances and fail to capture user-level thematic preferences across dialogues. To address these challenges, we propose CATCH (Controllable Theme Detection with Contextualized Clustering and Hierarchical Generation), a unified framework that integrates three core components: (1) context-aware topic representation, which enriches utterance-level semantics using surrounding topic segments; (2) preference-guided topic clustering, which jointly models semantic proximity and personalized feedback to align themes across conversations; and (3) a hierarchical theme generation mechanism designed to suppress noise and produce robust, coherent topic labels. Experiments on a multi-domain customer dialogue benchmark demonstrate that CATCH achieves state-of-the-art performance in both theme classification and topic distribution quality. Notably, it ranked second in the official blind evaluation of the DSTC-12 Controllable Theme Detection Track, showcasing its effectiveness and generalizability in real-world dialogue systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ke-etal-2025-catch">
<titleInfo>
<title>CATCH: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Ke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiahui</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kuang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shenghao</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Feng</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haizhou</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Twelfth Dialog System Technology Challenge</title>
</titleInfo>
<name type="personal">
<namePart type="given">Behnam</namePart>
<namePart type="family">Hedayatnia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivian</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhang</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raghav</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michel</namePart>
<namePart type="family">Galley</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Avignon, France</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-330-2</identifier>
</relatedItem>
<abstract>Theme detection is a fundamental task in user-centric dialogue systems, aiming to identify the latent topic of each utterance without relying on predefined schemas. Unlike intent induction, which operates within fixed label spaces, theme detection requires cross-dialogue consistency and alignment with personalized user preferences, posing significant challenges. Existing methods often struggle with sparse, short utterances and fail to capture user-level thematic preferences across dialogues. To address these challenges, we propose CATCH (Controllable Theme Detection with Contextualized Clustering and Hierarchical Generation), a unified framework that integrates three core components: (1) context-aware topic representation, which enriches utterance-level semantics using surrounding topic segments; (2) preference-guided topic clustering, which jointly models semantic proximity and personalized feedback to align themes across conversations; and (3) a hierarchical theme generation mechanism designed to suppress noise and produce robust, coherent topic labels. Experiments on a multi-domain customer dialogue benchmark demonstrate that CATCH achieves state-of-the-art performance in both theme classification and topic distribution quality. Notably, it ranked second in the official blind evaluation of the DSTC-12 Controllable Theme Detection Track, showcasing its effectiveness and generalizability in real-world dialogue systems.</abstract>
<identifier type="citekey">ke-etal-2025-catch</identifier>
<location>
<url>https://aclanthology.org/2025.dstc-1.2/</url>
</location>
<part>
<date>2025-08</date>
<extent unit="page">
<start>17</start>
<end>26</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CATCH: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation
%A Ke, Rui
%A Xu, Jiahui
%A Wang, Kuang
%A Yang, Shenghao
%A Jiang, Feng
%A Li, Haizhou
%Y Hedayatnia, Behnam
%Y Chen, Vivian
%Y Chen, Zhang
%Y Gupta, Raghav
%Y Galley, Michel
%S Proceedings of the Twelfth Dialog System Technology Challenge
%D 2025
%8 August
%I Association for Computational Linguistics
%C Avignon, France
%@ 979-8-89176-330-2
%F ke-etal-2025-catch
%X Theme detection is a fundamental task in user-centric dialogue systems, aiming to identify the latent topic of each utterance without relying on predefined schemas. Unlike intent induction, which operates within fixed label spaces, theme detection requires cross-dialogue consistency and alignment with personalized user preferences, posing significant challenges. Existing methods often struggle with sparse, short utterances and fail to capture user-level thematic preferences across dialogues. To address these challenges, we propose CATCH (Controllable Theme Detection with Contextualized Clustering and Hierarchical Generation), a unified framework that integrates three core components: (1) context-aware topic representation, which enriches utterance-level semantics using surrounding topic segments; (2) preference-guided topic clustering, which jointly models semantic proximity and personalized feedback to align themes across conversations; and (3) a hierarchical theme generation mechanism designed to suppress noise and produce robust, coherent topic labels. Experiments on a multi-domain customer dialogue benchmark demonstrate that CATCH achieves state-of-the-art performance in both theme classification and topic distribution quality. Notably, it ranked second in the official blind evaluation of the DSTC-12 Controllable Theme Detection Track, showcasing its effectiveness and generalizability in real-world dialogue systems.
%U https://aclanthology.org/2025.dstc-1.2/
%P 17-26
Markdown (Informal)
[CATCH: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation](https://aclanthology.org/2025.dstc-1.2/) (Ke et al., DSTC 2025)
ACL