Hello, my name is Zhuoran Li and your can call me Ran for short:) I am always driven by interesting and challenging problems (not specific methods) about developing generalist robots that learn and plan to help people. I am currently a research assistant working with Prof. Jiajun Wu, Prof. David Hsu, and Dr. Weiyu Liu, working on automatically learning effective abstractions—state abstractions (predicates) and action abstractions (skills)—from everyday human videos for robot planning, using neural-symbolic methods and foundation models.

Previously, I received my Bachelor of Science with the highest distinction from the National University of Singapore (NUS). I had a wonderful research experience at the Stanford Vision and Learning Lab (SVL), working on designing ideal scene representations for robotic manipulationby leveraging foundation models, as an undergraduate visiting research intern, where I was grateful to be advised by Prof. Jiajun Wu and Prof. Yunzhu Li.

At the intersection of automated planning and machine learning, my long-term research goal is to build generalist, human-centered robots that perceive, model, and interact with the real world, executing long-horizon tasks from simple language commands like "make the coffee". I pursue robust generalization through a structured representation of world knowledge that leverages the rich knowledge embedded in human language and everyday activity and grounds it through interaction, connecting abstract concepts to sensorimotor skills. Concretely, I develop neural-symbolic methods and foundation model pipelines that automatically learn effective abstractions—state abstractions (predicates) and action abstractions (skills)—from everyday human videos for robot planning. These structures make learning more data-efficient, more compositionally generalizable, and accelerate inference and planning. Before formalizing this long-term research direction, I worked on designing scene representations for robotic manipulation. That experience gives me a strong foundation in robotic perception and manipulation.