Inference is All You Need: Self Example Retriever for Cross-domain Dialogue State Tracking with ChatGPT

TL;DR

This work tackles cross-domain dialogue state tracking by removing the need for parameter updates. It introduces SERI-DST, an inference-only pipeline that uses in-context learning with ChatGPT and a self-example retriever to fetch demonstrations from non-target domains, guiding DST without training. The approach achieves competitive DomainJGA on MultiWOZ2.1, and the authors analyze error types, reasoning patterns, and domain-transfer effects to understand when retrieval helps. Overall, the framework offers a scalable, explainable path for domain transfer in DST and motivates future exploration of LLM-based retrieval in in-context learning settings.

Abstract

Traditional dialogue state tracking approaches heavily rely on extensive training data and handcrafted features, limiting their scalability and adaptability to new domains. In this paper, we propose a novel method that leverages inference and in-context learning with ChatGPT for domain transfer in dialogue state tracking, without any parameter updates. By guiding ChatGPT's chain of thought, we enable it to retrieve relevant examples and generalize knowledge to accurately infer dialogue states, solely through inference. Experimental results on the MultiWOZ dataset demonstrate competitive performance and promising generalization across domains. Our parameter-free approach offers a scalable and adaptable solution, opening new research directions in domain transfer learning.

Figures (3)

• Figure 1: Overview of SERI-DST.
• Figure 2: Error analysis of SERI-DST and other fine-tuned DST models.
• Figure 3: Impact of different domains on transfer learning.