CATCH: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation

TL;DR

CATCH is a novel framework that enhances theme detection in dialogues by integrating contextualized clustering and hierarchical generation, effectively capturing user preferences and improving topic coherence.

Contribution

It introduces a unified approach combining context-aware representation, preference-guided clustering, and hierarchical generation for better theme detection in dialogues.

Findings

Outperforms existing methods on DSTC-12 benchmark.

Effectively captures user preferences across dialogues.

Produces coherent and robust topic labels.

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 for accurate topic representation 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 dialogue; 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 (DSTC-12) demonstrate the effectiveness of CATCH with 8B LLM in both theme clustering and topic generation quality.