CATCH: A Framework for Controllable Theme Detection in Dialogue Systems
Published:Dec 25, 2025 15:33
•1 min read
•ArXiv
Analysis
This paper addresses the challenge of theme detection in user-centric dialogue systems, a crucial task for understanding user intent without predefined schemas. It highlights the limitations of existing methods in handling sparse utterances and user-specific preferences. The proposed CATCH framework offers a novel approach by integrating context-aware topic representation, preference-guided topic clustering, and hierarchical theme generation. The use of an 8B LLM and evaluation on a multi-domain benchmark (DSTC-12) suggests a practical and potentially impactful contribution to the field.
Key Takeaways
- •Proposes CATCH, a novel framework for theme detection in dialogue systems.
- •Addresses limitations of existing methods in handling sparse utterances and user preferences.
- •Integrates context-aware topic representation, preference-guided clustering, and hierarchical generation.
- •Evaluated on a multi-domain customer dialogue benchmark (DSTC-12) with an 8B LLM.
Reference
“CATCH integrates three core components: (1) context-aware topic representation, (2) preference-guided topic clustering, and (3) a hierarchical theme generation mechanism.”