Moo Jin Kim
moojink at cs dot stanford dot edu

I am a Ph.D. student in Computer Science at Stanford University, advised by Chelsea Finn and Percy Liang . I conduct research in AI and robotics.

Previously, I graduated summa cum laude from UCLA with a bachelor's degree in Computer Science and received a master's degree in Computer Science from Stanford. At Stanford, I was thankfully one of five students in my graduating class generously supported by the Siebel Scholarship.

I am passionate about teaching and mentoring students and have served as the Head TA for CS231N in Spring 2022 and as a research mentor for LINXS (a Stanford CS diversity outreach program) in Summer 2023.

I also founded Deep Learning Portal, which is an AI mentorship and outreach program that provides high-quality, personalized guidance to disadvantaged college students so that they can learn deep learning fundamentals quickly.

Google Scholar  /  Twitter  /  LinkedIn  /  GitHub

Research

I am currently interested in leveraging multimodal foundation models to learn high-performing, generalizable policies in robotics.

We introduce OpenVLA, a 7B-parameter open-source vision-language-action model (VLA), pretrained on 970k robot episodes from the Open X-Embodiment dataset. OpenVLA sets a new state of the art for generalist robot manipulation policies. It supports controlling multiple robots out of the box and can be quickly adapted to new robot setups via parameter-efficient fine-tuning. OpenVLA models, code, and training data are fully open-source.

We introduce the Open X-Embodiment Dataset, the largest robot learning dataset to date with 1M+ real robot trajectories, spanning 22 robot embodiments. We train large, Transformer-based policies on the dataset (RT-1-X, RT-2-X) and show that co-training with our diverse dataset substantially improves performance.

We introduce BridgeData V2, a large and diverse dataset of robotic manipulation behaviors designed to facilitate research on scalable robot learning. BridgeData V2 contains 60,096 trajectories collected across 24 environments on a publicly available low-cost robot.

We leverage neural radiance fields (NeRFs) to generate perturbed end-effector wrist camera viewpoints while simultaneously calculating corrective actions in order to improve absolute success rates of 6-DoF robotic grasping policies by 22.5% on average.

We augment narrow robotic imitation datasets with broad unlabeled human video demonstrations to greatly enhance the generalization of eye-in-hand visuomotor policies. Although a clear visual domain gap exists between human and robot data, our framework does not need to employ any explicit domain adaptation method and enables robots to generalize to both new environment configurations and new tasks that are unseen in the robot demonstration data.

We conduct extensive experiments to show that utilizing a hand-centric (eye-in-hand) visual perspective consistently improves training efficiency and out-of-distribution generalization in vision-based robotic manipulation. These benefits hold across a variety of learning algorithms, experimental settings, and distribution shifts. When hand-centric observability is not sufficient, we propose a simple yet effective approach to incorporate a third-person information stream while regularizing it via a variational information bottleneck to prevent overfitting.