Search and recommendation systems are everywhere to help tackle the information overload problem and have profound impact on people’s life and our society. However, they suffer from the basic information asymmetry problem between the user and system. Conversational systems offer a natural way to bridge the two, and the development of intelligent conversational systems ranks high on the agenda of current AI research.
In this talk, the speaker will first emphasize a new information seeking paradigm–multimodal conversational search & recommendation (MCSR) and introduce the challenges in realizing it. It is then followed by a research overview of the speaker and technical details in tackling some of these challenges. As a new emerging topic, the speaker will also show some identified research gaps that call for significant attention and effort in the near future.
About the Speaker
Dr Lizi Liao is a Research Fellow working with Prof. Tat-Seng Chua and leads the dialogue research group in the NExT++ Lab at the National University of Singapore (NUS), where she joined in July 2019. She received a PhD degree from NUS Graduate School for Integrative Sciences and Engineering (NGS) in November 2019, advised by Prof. Tat-Seng Chua. Her research interest covers conversational systems and multimedia analysis, specifically in developing models that can understand human inputs in various forms (e.g. textual & visual) and communicate with human to perform certain tasks. She publishes regularly in top AI and Data Science conferences and journals such as SIGIR, WWW, ACM MM, AAAI, IJCAI, ICDE, IEEE TKDE, TACL, etc. One of her work was nominated in the ACM MM Best Paper Final List in 2018. She also serves as (senior) PC member of these prestigious conferences and reviewer of the top journals. She served as one of the organizing committee member of ACM MM and chaired sessions in the SIGIR and KDD conference.
She is a tenure-track faculty candidate for the Human-Machine Collaborative Systems, Multimedia cluster.
