Learning to Select Knowledge for Response Generation in Dialog Systems

TL;DR

This paper introduces an end-to-end neural dialogue model with a novel knowledge selection mechanism that improves response informativeness by effectively choosing relevant external knowledge during training and inference.

Contribution

It proposes a new knowledge selection method using both prior and posterior distributions, enhancing the model's ability to incorporate relevant knowledge in responses.

Findings

Outperforms previous models in automatic evaluations

Achieves higher human-rated response quality

Effectively selects appropriate knowledge during training and inference

Abstract

End-to-end neural models for intelligent dialogue systems suffer from the problem of generating uninformative responses. Various methods were proposed to generate more informative responses by leveraging external knowledge. However, few previous work has focused on selecting appropriate knowledge in the learning process. The inappropriate selection of knowledge could prohibit the model from learning to make full use of the knowledge. Motivated by this, we propose an end-to-end neural model which employs a novel knowledge selection mechanism where both prior and posterior distributions over knowledge are used to facilitate knowledge selection. Specifically, a posterior distribution over knowledge is inferred from both utterances and responses, and it ensures the appropriate selection of knowledge during the training process. Meanwhile, a prior distribution, which is inferred from utterances only, is used to approximate the posterior distribution so that appropriate knowledge can be selected even without responses during the inference process. Compared with the previous work, our model can better incorporate appropriate knowledge in response generation. Experiments on both automatic and human evaluation verify the superiority of our model over previous baselines.