Long Le

I am a PhD student at the University of Pennsylvania's GRASP Laboratory, advised by Prof. Eric Eaton, and closely working with Prof. Dinesh Jayaraman and Prof. Lingjie Liu. I will join Google DeepMind as a student researcher in Summer 2026.

My research focuses on robot learning and computer vision, including simulation-based learning, tactile sensing, dexterous manipulation, navigation, and post-training vision-language-action models.

Previously, I studied Mathematics and Computer Science at UMass Amherst and worked at Google, CMU Robotics Institute, and Instagram.

Email · Scholar · Twitter · GitHub

Please feel free to send me an email if you are interested in research discussions, collaborations, or working on research projects. I am also always looking to mentor highly motivated students.

Selected Work

I am interested in scalable robot learning that enables robots to perform diverse tasks robustly. Below are some of my works. (* indicates equal contribution.)

Preprint
2026

OmniGuide: Universal Guidance Fields for Enhancing Generalist Robot Policies

Yunzhou Song*, Long Le*, Yong-Hyun Park, Jie Wang, Junyao Shi, Lingjie Liu, Jiatao Gu, Eric Eaton, Dinesh Jayaraman, Kostas Daniilidis

website / twitter / paper

Guidance fields use reconstruction, VLM, and hand-tracking foundation models to steer generalist robot policies with low latency.

Preprint
2025

Maestro: Orchestrating Robotics Modules with Vision-Language Models for Zero-Shot Generalist Robots

Junyao Shi*, Rujia Yang*, Kaitian Chao*, Selina Wan, Yifei Shao, Jiahui Lei, Jianing Qian, Long Le, Pratik Chaudhari, Kostas Daniilidis, Chuan Wen, Dinesh Jayaraman

website

A coding VLM agent composes perception, geometry, control, policy, and image-editing modules into interpretable robot programs.

CVPR
2026

UniPixie: Unified and Probabilistic 3D Physics Learning via Flow Matching

Qilin Huang*, Quynh Anh Huynh*, Long Le*, Chen Wang, Chuhao Chen, Ryan Lucas, Eric Eaton, Lingjie Liu

website coming soon

A unified architecture predicts simulation-ready parameters for MPM, LBS, and spring-mass systems with controllable generation.

Preprint
2025

Pixie: Physics from Pixels

Long Le, Ryan Lucas, Chen Wang, Chuhao Chen, Dinesh Jayaraman, Eric Eaton, Lingjie Liu

website / twitter / paper / code

Distilled CLIP features predict fast 3D physics that generalize across scenes.

ICLR
2025

Articulate Anything

Long Le, Jason Xie, William Liang, Hung-Ju Wang, Yue Yang, Jason Ma, Kyle Vedder, Arjun Krishna, Dinesh Jayaraman, Eric Eaton

website / twitter / paper / code

VLMs automatically generate interactable articulated 3D models from text, images, or videos.

Preprint
2024

Distributed Continual Learning

Long Le, Marcel Hussing, Eric Eaton

paper

Continual and federated learning for independent agents facing unique tasks under network constraints.

Nature MI
2024

A collective AI via lifelong learning and sharing at the edge

A Soltoggio et al.

paper

A survey of approaches and challenges for collaborative lifelong AI on edge devices.