Sequence-to-Sequence Learning for Task-oriented Dialogue with Dialogue State Representation

TL;DR

This paper introduces a hybrid dialogue system that combines sequence-to-sequence learning with explicit dialogue state representation and knowledge base querying, improving task-oriented dialogue performance.

Contribution

It proposes modeling dialogue states as fixed-size distributed representations to enhance sequence-to-sequence dialogue systems with knowledge base access.

Findings

Significant performance improvement over baseline models

Outperforms existing sequence-to-sequence models in automatic and human evaluations

Effective integration of dialogue state representation with knowledge base querying

Abstract

Classic pipeline models for task-oriented dialogue system require explicit modeling the dialogue states and hand-crafted action spaces to query a domain-specific knowledge base. Conversely, sequence-to-sequence models learn to map dialogue history to the response in current turn without explicit knowledge base querying. In this work, we propose a novel framework that leverages the advantages of classic pipeline and sequence-to-sequence models. Our framework models a dialogue state as a fixed-size distributed representation and use this representation to query a knowledge base via an attention mechanism. Experiment on Stanford Multi-turn Multi-domain Task-oriented Dialogue Dataset shows that our framework significantly outperforms other sequence-to-sequence based baseline models on both automatic and human evaluation.