Lehong Wu
Lehong Wu is a first-year Ph.D. student in the Machine Learning Department at Carnegie Mellon University, advised by Prof. Aviral Kumar. He received his Bachelor's degree in EECS from Peking University in July 2026. His research focuses on robot learning, particularly robotic reinforcement learning and manipulation, with the goal of building autonomous and scalable robot learning systems in the real world. He is known for publications including SimLauncher (IROS 2025 oral), UniTac2Pose (CoRL 2025), and MacDiff (ECCV 2024).
“The paper's core contribution is demonstrating that a VLM can be trained to produce free-form natural language reasoning before issuing high-level instructions to a frozen low-level robot policy”
Source→“The work is supported by ONR, Schmidt Sciences AI2050 Early Career Fellowship, and a TRI U3.0 project”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.