TWIZ: A conversational Task Wizard with multimodal curiosity-exploration

This paper describes the vision, scientific contributions, and technical details of the Task Wizard (TWIZ) team’s participation in the Alexa TaskBot Challenge 2021. Our bot design envisions the support of an engaging experience, where users are guided through multimodal conversations, towards the successful completion of the selected task. This is achieved through four key principles: a) robust dialog interaction, by making core dialog components (utterance processing, dialog manager, and response generator) tailored for an in-the-wild setting, b) effective and conversational task grounding, c) delivery of an immersive multimodal task guiding experience, and d) engagement maximization and user cognitive stimuli. In the following sections, we present the proposed methods to tackle each one of these principles, leading to a novel multimodal curiosity-exploration task guiding conversational assistant, that manages to balance user engagement and cognitive load, while guiding users through complex tasks.