An Adversarial Approach to High-Quality, Sentiment-Controlled Neural Dialogue Generation

TL;DR

This paper introduces an adversarial training framework for neural dialogue generation that enables explicit sentiment control while maintaining response relevance and fluency, combining conditional adversarial learning with flexible generator architectures.

Contribution

The work presents a novel adversarial approach for sentiment-controlled dialogue generation that can incorporate various generator models, improving response quality and sentiment accuracy.

Findings

Generated responses are semantically relevant and sentiment-controlled.

The framework outperforms baseline models in automatic and human evaluations.

Flexible architecture allows integration with different sequence-to-sequence models.

Abstract

In this work, we propose a method for neural dialogue response generation that allows not only generating semantically reasonable responses according to the dialogue history, but also explicitly controlling the sentiment of the response via sentiment labels. Our proposed model is based on the paradigm of conditional adversarial learning; the training of a sentiment-controlled dialogue generator is assisted by an adversarial discriminator which assesses the fluency and feasibility of the response generating from the dialogue history and a given sentiment label. Because of the flexibility of our framework, the generator could be a standard sequence-to-sequence (SEQ2SEQ) model or a more complicated one such as a conditional variational autoencoder-based SEQ2SEQ model. Experimental results using automatic and human evaluation both demonstrate that our proposed framework is able to generate both semantically reasonable and sentiment-controlled dialogue responses.