Zhenyu Zhao
Zhenyu Zhao is a PhD researcher at the University of Southern California, where he works in the Physical Superintelligence Lab under Prof. Yue Wang, focusing on general-purpose intelligence for humanoid robots, including teleoperation, policy learning, and policy evaluation. He is best known as co-first author, alongside Hongyi Jing, of Humanoid Everyday, a large-scale multimodal dataset of 10,300 trajectories across 260 manipulation tasks that also introduces a cloud-based evaluation platform for humanoid policies. The paper received a Best Paper Award at the Embodied AI Workshop at CVPR 2026 and an oral presentation at ICRA 2026. Zhao is committed to open-source research and has also co-authored Psi-Zero, an open foundation model for humanoid loco-manipulation presented at RSS 2026.
“This paper is a dataset and infrastructure paper, not a breakthrough in robot capabilities. But it matters strategically: it establishes the first large-scale, multimodal, task-diverse benchmark for full-body humanoid manipulation.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.