
CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining
I-Chun Arthur Liu, Krzysztof Choromanski, Sandy Huang, Connor Schenck
Under Review

PhD Candidate · University of Southern California
I'm a final-year PhD candidate at the University of Southern California, advised by Prof. Gaurav Sukhatme and working closely with Prof. Daniel Seita. My research focuses on 3D representations that help robots learn bimanual manipulation efficiently, enabling two-armed robots to handle complex real-world tasks with less training data. I've done research internships at Google DeepMind and Amazon Robotics, and I hold an M.S. in Computer Science from UCLA and a B.S. in Computer Science from UW–Madison.
One paper accepted to ICRA 2026!
One paper accepted to CoRL 2025!
Started as a Student Researcher at Google DeepMind for Summer/Fall 2025.
Passed my PhD qualification exam and was promoted to PhD Candidate!
One paper accepted to CoRL 2024!
* denotes equal contribution. † denotes equal advising.

I-Chun Arthur Liu, Krzysztof Choromanski, Sandy Huang, Connor Schenck
Under Review
Google DeepMind
Jun 2025 – Dec 2025
Mountain View, CA
Developed 3D pre‑training framework for robotic manipulation.
USC Robotic Embedded Systems Laboratory (RESL)
Aug 2023 – Present
Los Angeles, CA
See Selected Publications for details.
Amazon Robotics
May 2023 – Aug 2023
North Reading, MA
Conducted research on targeted bin picking.
USC RESL
Jan 2021 – May 2023
Los Angeles, CA
Published three first-author papers on robot manipulation, imitation learning, and reinforcement learning.
Academia Sinica
Jan 2019 – Dec 2020
Taipei, Taiwan
Researched computer vision topics including object detection and instance segmentation.
USC Viterbi School of Engineering
Aug 2023
USC Viterbi School of Engineering
May 2023