MTSS: Learn from Multiple Domain Teachers and Become a Multi-domain Dialogue Expert

TL;DR

This paper introduces a novel multi-domain dialogue system training approach inspired by school teaching, where domain-specific teachers teach a universal student, resulting in a model that excels across multiple dialogue domains.

Contribution

The paper proposes a multi-teacher, single-student framework that effectively learns multi-domain dialogue policies by leveraging domain-specific teachers, improving over existing methods.

Findings

Achieves competitive results with state-of-the-art methods in multi-domain settings.

Effectively learns domain-specific knowledge and policies through teacher-student training.

Demonstrates robustness in both multi-domain and single domain dialogue tasks.

Abstract

How to build a high-quality multi-domain dialogue system is a challenging work due to its complicated and entangled dialogue state space among each domain, which seriously limits the quality of dialogue policy, and further affects the generated response. In this paper, we propose a novel method to acquire a satisfying policy and subtly circumvent the knotty dialogue state representation problem in the multi-domain setting. Inspired by real school teaching scenarios, our method is composed of multiple domain-specific teachers and a universal student. Each individual teacher only focuses on one specific domain and learns its corresponding domain knowledge and dialogue policy based on a precisely extracted single domain dialogue state representation. Then, these domain-specific teachers impart their domain knowledge and policies to a universal student model and collectively make this student model a multi-domain dialogue expert. Experiment results show that our method reaches competitive results with SOTAs in both multi-domain and single domain setting.