Unsupervised Extraction of Dialogue Policies from Conversations
This provides a productivity tool for conversation designers to develop dialogue policies more efficiently, though it is incremental as it builds on existing LLM and graph techniques.
The paper tackles the challenge of extracting dialogue policies from conversational data by proposing a graph-based method that converts conversations into canonical forms and uses graph traversal to generate dialogue flows, resulting in a more effective representation than LLM prompting alone.
Dialogue policies play a crucial role in developing task-oriented dialogue systems, yet their development and maintenance are challenging and typically require substantial effort from experts in dialogue modeling. While in many situations, large amounts of conversational data are available for the task at hand, people lack an effective solution able to extract dialogue policies from this data. In this paper, we address this gap by first illustrating how Large Language Models (LLMs) can be instrumental in extracting dialogue policies from datasets, through the conversion of conversations into a unified intermediate representation consisting of canonical forms. We then propose a novel method for generating dialogue policies utilizing a controllable and interpretable graph-based methodology. By combining canonical forms across conversations into a flow network, we find that running graph traversal algorithms helps in extracting dialogue flows. These flows are a better representation of the underlying interactions than flows extracted by prompting LLMs. Our technique focuses on giving conversation designers greater control, offering a productivity tool to improve the process of developing dialogue policies.