Yunchao Zhang
Like the wind.
I am a third-year PhD student in the HKU Musketeers Foundation Institute of Data Science (HKU-IDS) at The University of Hong Kong, supervised by Prof.Yanchao Yang and co-advised by Prof.Yi Ma. I also work closely with Prof.Leonidas Guibas. Undergraduate studies from Zhejiang University, majoring in Computer Science.
My research interest lies in building embodied agents by learning from natural signals. I aim to create human-like intelligence while enhancing our understanding of human brains. I am also interested in neuroscience and cognitive science.
My recent research focus is 3D representation learning related to the correlation information, and utilize it for more efficient scene manipulation / human interaction.
news
| Jul 07, 2026 | 🎉 One paper is accepted to ICML 2026: GeoMoLa: Geometry-Aware Motion Latents for Learning Robust Manipulation Policies |
|---|---|
| Jan 23, 2025 | 🎉 One paper is accepted to ICLR 2025: Structure-Aware Gaussians through Lightweight Information Shaping |
| Mar 01, 2024 | 🎉 One paper is accepted to NAACL 2024: Navigation as Attackers Wish? Towards Building Byzantine-Robust Embodied Agents under Federated Learning |
| Nov 01, 2023 | 🎉 One paper is accepted to NeurIPS 2023: Universal Prompt Tuning for Graph Neural Networks |
| Apr 01, 2023 | I’ll be joining HKU IDS as a PhD student   |
selected publications
-
GeoMoLa: Geometry-Aware Motion Latents for Learning Robust Manipulation PoliciesInternational Conference on Machine Learning (ICML 2026), 2026 -
InfoGaussian: Mutual Information Shaping for Consistent Dynamics in 3D GaussianInternational Conference on Learning Representations (ICLR 2025), 2024