Modeling Multi-Action Policy for Task-Oriented Dialogues

TL;DR

This paper introduces a novel recurrent policy model called gCAS for task-oriented dialogue systems that predicts multiple actions per turn, enhancing expressiveness and robustness in conversations.

Contribution

The paper proposes the gCAS model that allows multi-action prediction per turn, improving dialogue management over existing single-action approaches.

Findings

gCAS outperforms existing models in multi-action prediction

Multi-action modeling improves dialogue robustness

Experimental results validate the effectiveness of gCAS

Abstract

Dialogue management (DM) plays a key role in the quality of the interaction with the user in a task-oriented dialogue system. In most existing approaches, the agent predicts only one DM policy action per turn. This significantly limits the expressive power of the conversational agent and introduces unwanted turns of interactions that may challenge users' patience. Longer conversations also lead to more errors and the system needs to be more robust to handle them. In this paper, we compare the performance of several models on the task of predicting multiple acts for each turn. A novel policy model is proposed based on a recurrent cell called gated Continue-Act-Slots (gCAS) that overcomes the limitations of the existing models. Experimental results show that gCAS outperforms other approaches. The code is available at https://leishu02.github.io/