Towards Proactive News Grounded Conversation

Anonymous

Keywords: Dialog System, Knowledge Grounded Generation, Text Generation

Abstract: Hot news is one of the most popular topics in daily conversations.However, news grounded conversation has long been stymied by the lack of well-designed task definition and scarce data. In this paper, we propose a novel task, Proactive News Grounded Conversation, in which a dialogue system can proactively lead the conversation based on some key topics of the news.In addition, both information-seeking and chit-chat scenarios are included realistically, where the user may ask a series of questions about the news details or express their opinions and be eager to chat.% With this proactive and informative dialogue system, the conversation becomes more interactive and engaging.To further develop this novel task, we collect a human-to-human Chinese dialogue dataset NewsDialogues, which includes 1K conversations with an average of 14.6 turns and careful annotations for proactive topic transition and grounded knowledge.Furthermore, we introduce two classic methods based on the pre-trained language models to solve this problem, which are the end-to-end method and the read-then-generate method.We conduct extensive experiments to analyze the performance of current models and further present several key findings and challenges to prompt future research. All our code and data will be available after acceptance.

Paper Type: long

Research Area: Dialogue and Interactive Systems