Yunshuang Li

Hello world!

I am a second-year Ph.D. student in Computer Science at the University of Southern California, where I am co-advised by Prof. Daniel Seita and Prof. Gaurav Sukhatme.

I received an M.S. degree in Robotics at the University of Pennsylvania in May 2024. I was a member of the Perception, Action, & Learning (PAL) Research Group, advised by Prof. Dinesh Jayaraman. Previously, I worked with Prof. Qi Dou at the Chinese University of Hong Kong as an undergrad researcher. I received my Honorable B.S. degree in Automatic Control from Chu Kochen Honors College, Zhejiang University in 2022.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Twitter

Research

My research focuses on robot learning for dexterous manipulation. I aim to enhance robotic capabilities under resource constraints by effectively leveraging diverse data sources and available physical resources. A key emphasis of my work is extending task horizons, enabling robots to perform long-horizon, multi-stage tasks and multitasking, as well as more delicate behaviors such as nonprehensile manipulation. Ultimately, my goal is to advance robotic manipulation toward human-level dexterity, enabling robots to learn, adapt, and interact with the physical world more effectively.

(* indicates equal contribution, † indicates equal advising)

We propose Geometry-aware Dexterous Pushing and Pulling, a method for nonprehensile manipulation which frames the problem as synthesizing and learning pre-contact dexterous hand poses that lead to effective pushing and pulling.

We propose to train a large combinatorial family of representations organized by scene entities. This hierarchical object decomposition for task-oriented representations permits selectively assembling different representations specific to each task while scaling in representational capacity with the complexity of the scene and the task.

We propose a challenging vision-based extrinsic contact localization task: with only a single RGB-D camera view of a robot workspace, identify when and where an object held by the robot contacts the rest of the environment.

This is the report paper on Workflow Recognition Challenge in MICCAI 2021. I lead the MedAIR team and rank the first over all the 5 rank method in one sub-challenge on multi-modal (videos and kinematics) workflow recognition of robotic surgery videos.

Service

Mentorship:
Conference reviewer: CoRL
Workshop reviewer: NeurIPS, ICRA, RSS
Teaching Assistant at USC: CSCI 103: Introduction to Programming, Fa25.
Teaching Assistant at UPenn: MEAM 5200: Introduction to Robotics, Sp23; CIS 5200: Machine Learning, Fa23; CIS 5190, Applied Machine Learning, Sp24.
: Instructor in Spring 23 at G.W. Carver High School.

Awards

----PhD----
USC Viterbi School of Engineering Fellowship, 2024
USC WiSE Graduate Top-Off Award(10k$), 2024
CoRL 2024 Travel Grant, 2024
----Master----
Best Conference Paper Award ICRA, 2024
UPenn MS Outstanding Research Award, 2024
UPenn President Gutmann Leadership Award, 2023
UPenn Summer Internship Award, 2023
Best Paper Award at CoRL LEAP workshop, 2023
CoRL 2023 Travel Grant, 2023
----Undergrad----
Chiang Chen Overseas Graduate Fellowship (10 students each year in mainland China, 50k$), 2022
Outstanding Graduate of Zhejiang Province issued by Department of Education of Zhejiang Province, 2022
National Scholarship issued by the Ministry of Education of PRC, 2021
First Prize Scholarship issued by Zhejiang University, 2018-2021

Media

Inspired by the template here.