Keynotes

MMM 2027 is pleased to announce the following keynote talk.

The Return of the Always-On Assistant

Angela Yao
Dean's Chair Associate Professor, School of Computing, National University of Singapore

Abstract

The idea of an always-on personal assistant is decades old. Wearable computing, lifelogging, and mixed-initiative interfaces each imagined a system that sees what we see, remembers what we forget, and offers help at the right moment. Each ran into the same limit: machines could record, but not understand, and the vision receded. Multimodal large language models are now reviving this vision, and with it, the old questions in a new form. An assistant running continuously over a video stream must decide what to attend to, what to remember, and when to speak. It must also recognize when a request is ambiguous enough to ask rather than guess. This talk revisits these questions through recent work on streaming egocentric video. It covers models that learn what evidence to gather and when it is enough to respond, user-centric spatial and personal memory, and interactive clarification. The talk ends by looking ahead to assistants that not only answer but act in the physical world, and to what it will take for people to trust a system that never switches off.

Biography

Angela Yao is a Dean's Chair Associate Professor in the School of Computing at the National University of Singapore, where she leads research on human-centric computer vision and video understanding. She has received Singapore's NRF Investigatorship (2026), NRF Fellowship for AI (2019), and the German Pattern Recognition Award (2018). She completed her PhD at ETH Zurich and previously held a faculty position at the University of Bonn. Angela will serve as Program Chair of CVPR 2027 and ICCV 2029.