Loo Yi

Yi Loo

I am 5th PhD student at the Singapore University of Technology and Design, advised by Malika Meghjani. I am broadly intersted in multi-agent and robot cooperation, focusing on Human-AI Collaboration and Human-Robot Collaboration. I am interested in developing agents and robots that adapts and aligns with human prefences. Recently, I have been looking into developing methods to leverage World Models (WM) for colllaborative decision making and also methods that enable Visual Language Action (VLA) models to collaborate and coordination.

I was born and raised in Penang, Malaysia. My family name "Loo" comes from the Penang Hokkien pronunciation for "吕" and my given name "Yi" comes from the Mandarin pronunciation for "奕".

I just defended my thesis and am on job market for any research positions in AI/Robot Collaboration!

Email / CV / Google Scholar / Github

Selected Publications

For a full list of my publications, please check out my Google Scholar page.
	Co²Gen: Collaboratively Generating Task Plans for Decentralized Multi-Robot and Human Collaboration Yi Loo, Siddhant Shrivastava, Minh-Long Pham, Malika Meghjani Under Review paper TL;DR: We develop a method for generating decentralized multi-robot demontration data for decentralized VLA model finetuning.
	Efficient Generation of Diverse Cooperative Agents with World Models Yi Loo, Akshunn Trivedi, Malika Meghjani RLDM, 2025 paper TL;DR: We develop a method with World Models to efficiently generate diverse collaborative agents for Zero-Shot Coordiantion training.
	A Hierarchical Approach to Population Training for Human-AI Collaboration Yi Loo, Gong Chen, Malika Meghjani IJCAI, 2023 paper / code TL;DR: We propose a hierarchichal RL agent for Human-AI Coordination, The hierarchichal structure allows the agent to switch between best response convention that adapts to novel partenrs.

Website template from here.