Prediction, Selection, and Generation: Exploration of Knowledge-Driven Conversation System

TL;DR

This paper presents a knowledge-driven conversation system that integrates background knowledge with pre-trained models, improving dialogue controllability and diversity, and analyzes key factors affecting its performance.

Contribution

It introduces a novel system combining knowledge bases and pre-training models, and studies the impact of various factors on knowledge-driven dialogue generation.

Findings

Achieved state-of-the-art performance.

Identified key factors influencing dialogue quality.

Provided insights for future research in knowledge-based conversations.

Abstract

In open-domain conversational systems, it is important but challenging to leverage background knowledge. We can use the incorporation of knowledge to make the generation of dialogue controllable, and can generate more diverse sentences that contain real knowledge. In this paper, we combine the knowledge bases and pre-training model to propose a knowledge-driven conversation system. The system includes modules such as dialogue topic prediction, knowledge matching and dialogue generation. Based on this system, we study the performance factors that maybe affect the generation of knowledge-driven dialogue: topic coarse recall algorithm, number of knowledge choices, generation model choices, etc., and finally made the system reach state-of-the-art. These experimental results will provide some guiding significance for the future research of this task. As far as we know, this is the first work to study and analyze the effects of the related factors.