Dynamic Knowledge Fusion for Multi-Domain Dialogue State Tracking

2026-03-11

AI Summary

This work addresses the challenges of multi-domain dialogue state tracking, particularly the difficulty of modeling dialogue history and the scarcity of annotated data. To this end, the authors propose a dynamic knowledge fusion framework that first employs a contrastive learning–trained encoder to select candidate slots relevant to the dialogue history. The structured information of these selected slots is then incorporated as dynamic contextual prompts to precisely integrate domain knowledge, thereby enhancing both the accuracy and consistency of state predictions. Built upon a pure encoder architecture without requiring an additional decoder, the proposed method achieves significant improvements in tracking accuracy and generalization across multiple dialogue benchmarks, demonstrating robust performance in complex conversational scenarios.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

The performance of task-oriented dialogue models is strongly tied to how well they track dialogue states, which records and updates user information across multi-turn interactions. However, current multi-domain DST encounters two key challenges: the difficulty of effectively modeling dialogue history and the limited availability of annotated data, both of which hinder model performance. To tackle the aforementioned problems, we develop a dynamic knowledge fusion framework applicable to multi-domain DST. The model operates in two stages: first, an encoder-only network trained with contrastive learning encodes dialogue history and candidate slots, selecting relevant slots based on correlation scores; second, dynamic knowledge fusion leverages the structured information of selected slots as contextual prompts to enhance the accuracy and consistency of dialogue state tracking. This design enables more accurate integration of dialogue context and domain knowledge. Results obtained from multi-domain dialogue benchmarks indicate that our method notably improves both tracking accuracy and generalization, validating its capability in handling complex dialogue scenarios.

Problem

• dialogue state tracking
• multi-domain
• dialogue history modeling
• annotated data scarcity

Innovation

• Dynamic Knowledge Fusion
• Multi-Domain Dialogue State Tracking
• Contrastive Learning
• Contextual Prompting
• Slot Selection