Learning to Plan and Realize Separately for Open-Ended Dialogue Systems

by   Sashank Santhanam, et al.

Achieving true human-like ability to conduct a conversation remains an elusive goal for open-ended dialogue systems. We posit this is because extant approaches towards natural language generation (NLG) are typically construed as end-to-end architectures that do not adequately model human generation processes. To investigate, we decouple generation into two separate phases: planning and realization. In the planning phase, we train two planners to generate plans for response utterances. The realization phase uses response plans to produce an appropriate response. Through rigorous evaluations, both automated and human, we demonstrate that decoupling the process into planning and realization performs better than an end-to-end approach.


page 1

page 7

page 14


Step-by-Step: Separating Planning from Realization in Neural Data-to-Text Generation

Data-to-text generation can be conceptually divided into two parts: orde...

Policy-Driven Neural Response Generation for Knowledge-Grounded Dialogue Systems

Open-domain dialogue systems aim to generate relevant, informative and e...

A Taxonomy of Empathetic Response Intents in Human Social Conversations

Open-domain conversational agents or chatbots are becoming increasingly ...

Designing a Symbolic Intermediate Representation for Neural Surface Realization

Generated output from neural NLG systems often contain errors such as ha...

Requirements Elicitation in Cognitive Service for Recommendation

Nowadays, cognitive service provides more interactive way to understand ...

Deal or No Deal? End-to-End Learning for Negotiation Dialogues

Much of human dialogue occurs in semi-cooperative settings, where agents...

On deploying the Artificial Sport Trainer into practice

Computational Intelligence methods for automatic generation of sport tra...