Learning to Express in Knowledge-Grounded Conversation

TL;DR

This paper introduces a novel model for knowledge-grounded dialogue generation that captures response structure and style using latent variables, enabling more diverse and style-consistent responses.

Contribution

It proposes a segmentation-based generation model with latent variables to learn and control response structure and style in knowledge-grounded conversations.

Findings

Model effectively learns response structure and style from few examples.

Generated responses match desired content style and structure.

Outperforms baseline models on benchmark datasets.

Abstract

Grounding dialogue generation by extra knowledge has shown great potentials towards building a system capable of replying with knowledgeable and engaging responses. Existing studies focus on how to synthesize a response with proper knowledge, yet neglect that the same knowledge could be expressed differently by speakers even under the same context. In this work, we mainly consider two aspects of knowledge expression, namely the structure of the response and style of the content in each part. We therefore introduce two sequential latent variables to represent the structure and the content style respectively. We propose a segmentation-based generation model and optimize the model by a variational approach to discover the underlying pattern of knowledge expression in a response. Evaluation results on two benchmarks indicate that our model can learn the structure style defined by a few examples and generate responses in desired content style.