The role of Rapport in iHuman - Conversational Agent Interaction: Modeling conversation to improve task performance in human- ag

Laboratoire: Inria Paris & PRAIRIE Institute for Interdisciplinary Research on AI
Adresse: Inria Paris, 2 rue Simone Iff, 75012 Paris
Équipe: COML (Communication & Machine Learning), ENS & Inria
 
This project exists at the intersection of Cognitive Science and AI and therefore has dual goals: to better understand the role of rapport (connivence) in human-human interaction, and to build better embodied conversational agents (ECAs) that have the ability to engage their users in both social and task talk, where the social talk serves to improve task performance. In order to achieve this objective, we collect and annotate natural conversational data, model human-human conversation, integrate the models into ECAs, and then evaluate their performance. We have internships available for CogMaster students from all filières (no programming experience necessary). 

Specifically, we currently have positions for interns to answer scientific questions by annotating conversational data for  conversational phenomena and analyzing the resultant annotations, or to build data science models of conversation, or to implement modules for the ECA dialogue system. For more information on the project, potential candidates may look at the SARA (Socially-Aware Robot Assistant) website at http://articulab.hcii.cs.cmu.edu/projects/sara/ and read some of the publications associated with the project, here http://articulab.hcii.cs.cmu.edu/publications/.

Contact: justine.cassell@inria.fr