Integrating User and Agent Models: A Deep Task-Oriented Dialogue System

TL;DR

This paper introduces SAMIA, a novel framework that models users as Seq2Seq and integrates user and agent models to improve task-oriented dialogue systems, reducing reliance on handcrafted rules and enhancing robustness.

Contribution

The paper proposes a new asymmetric integration framework for user and agent models, modeling users as Seq2Seq and leveraging this for training and filtering in dialogue systems.

Findings

SAMIA improves dialogue system stability and robustness.

Experiments on coffee ordering data validate the effectiveness of SAMIA.

User modeling as Seq2Seq enhances system flexibility.

Abstract

Task-oriented dialogue systems can efficiently serve a large number of customers and relieve people from tedious works. However, existing task-oriented dialogue systems depend on handcrafted actions and states or extra semantic labels, which sometimes degrades user experience despite the intensive human intervention. Moreover, current user simulators have limited expressive ability so that deep reinforcement Seq2Seq models have to rely on selfplay and only work in some special cases. To address those problems, we propose a uSer and Agent Model IntegrAtion (SAMIA) framework inspired by an observation that the roles of the user and agent models are asymmetric. Firstly, this SAMIA framework model the user model as a Seq2Seq learning problem instead of ranking or designing rules. Then the built user model is used as a leverage to train the agent model by deep reinforcement learning. In the test phase, the output of the agent model is filtered by the user model to enhance the stability and robustness. Experiments on a real-world coffee ordering dataset verify the effectiveness of the proposed SAMIA framework.