A Survey on Common Ground Building in Conversational Agents to Support Communicative Success

PhD Qualifying Examination


Title: "A Survey on Common Ground Building in Conversational Agents to 
Support Communicative Success"

by

Mr. Dingdong LIU


Abstract:

Communication is a shared journey where everyone actively contributes and 
consistently checks their construal to ensure they are on the same page. 
Establishing and maintaining such shared understanding, or ``common 
ground,'' is crucial for successful communication. With the recent 
surge in AI agents, the question of how to facilitate common ground building 
between AI assistance and human users has gained significant 
importance/attention. This survey explores the techniques and algorithms that 
enable conversational AI agents to establish, maintain, and repair common 
ground with their human interlocutors. To provide a structured analysis, the 
survey papers are categorized based on their reactive or proactive approach. We 
scrutinize the impact of these approaches on the quality of interaction, user 
satisfaction, and the overall effectiveness of the agents. Furthermore, this 
survey underscores the research challenges and opportunities presented by 
common ground building in conversational agents, providing a comprehensive 
overview of the current state of the art and signalling potential avenues for 
future exploration.


Date:                   Friday, 26 April 2024

Time:                   3:30pm - 5:30pm

Venue:                  Room 4475
                        Lifts 25/26

Committee Members:      Dr. Xiaojuan Ma (Supervisor)
                        Prof. Fugee Tsung (Supervisor)
                        Prof. Pedro Sander (Chairperson)
                        Prof. Bertram Emil Shi
                        Dr. Long Chen