Abstract
Multi-role dialogue is challenging in natural language processing (NLP), which needs not only to understand sentences but also to simulate interaction among roles. However, the existing methods assume that only two speakers are present in a conversation. In real life, this assumption is not always valid. More often, there are multiple speakers involved. To address this issue, we propose a multi-role interposition dialogue system (MIDS) that generates reasonable responses based on the dialogue context and next speaker prediction. The MIDS employs multiply role-defined encoders to understand each speaker and an independent sequence model to predict the next speaker. The independent sequence model also works as a controller to integrate encoders with weights. Then, an attention-enhanced decoder generates responses based on the dialogue context, speaker prediction, and integrated encoders. Moreover, with the help of unique speaker prediction, the MIDS is able to generate diverse responses and allow itself to join (interpose) the conversation when appropriate. Furthermore, a novel reward function and an updating policy of reinforcement learning (RL) are applied to the MIDS, which further enable MIDS the ability to write drama scripts. The experimental results demonstrate that the MIDS offers a significant improvement to the accuracy of speaker prediction and the reduction of response generation perplexity. It is also able to interact with users without cues during real-life online conversations and avoid meaningless conversation loops while generating scripts. This paper marks the first step toward multi-role humorous dialogue generation.
Highlights
Dialogue systems are responsible for interaction between humans and machines, making them vital components of chatbots and personal assistants [1]
Inspired by the idea that roles with different backgrounds should have stable response patterns and similar speaking orders in conversation groups, we propose a multi-role interposition dialogue system (MIDS) in this paper
We propose MIDS, a novel dynamic fusion framework for multi-role response generation, which utilizes speaker prediction to control the combination of role-defined encoders and the contextual attention based decoder
Summary
Dialogue systems are responsible for interaction between humans and machines, making them vital components of chatbots and personal assistants [1]. A few role-related dialogue systems have been proposed recently, such as multi-party conversations [2], [3], rolebased contextual models [4], and persona-based conversation agents [5] These works show that role-sensitive models are able to enhance the generation and avoid mundane responses. RELATED WORK Exiting systems solve speaker prediction or response generation separately, and to our knowledge, no existing model is capable of handling multi-role dialogues directly, e.g. Seq2Seq [8] or RNN/CNN [9], [10]. The proposed MIDS realizes multi-role dialogue generation by integrating three sub-tasks: dialogue generation, multi-role interaction (speaker prediction) and RL framework. Recent progress towards these subtasks is discussed as follows
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
