I am currently a first-year PhD student in computer science at the University of Massachusetts Amherst, advised by Prof. Chuang Gan. My primary research focus centers around Robotics and Embodied AI, with a particular emphasis on robot skill acquisition. My work primarily revolves around equipping robots with the capability to acquire versatile skills and improve their generalization abilities. The practical applications of my research span across various domains, including domestic robots and autonomous manipulation tasks. Previous to my graduate study, I obtained my bachelor degree in Peking University when I worked as an undergraduate member in Center on Frontiers of Computing Studies (CFCS), under the guidance of Prof. Hao Dong.
- Sep 2023: Welcome to our CoRL Workshop: TGR
- Sep 2023: First-year Phd in Umass
- The purpose of this document is to share the excitement of the authors with the community and highlight a promising research direction in robotics and AI. The authors believe the proposed paradigm is a feasible path towards accomplishing the longstanding goal of robotics research: deploying robots, or embodied AI agents more broadly, in various non-factory real-world settings to perform diverse tasks.
- RoboGen leverages the latest advancements in foundation and generative models. Instead of directly using or adapting these models to produce policies or low-level actions, we advocate for a generative scheme, which uses these models to automatically generate diversified tasks, scenes, and training supervisions, thereby scaling up robotic skill learning with minimal human supervision.
- We apply soft prompt-based learning to molecular dynamics simulations, achieving strong generalization across various conditions with limited training data. The framework includes pre-training with data mixing and prompts, followed by meta-learning-based fine-tuning.
- We propose a novel framework that learns to perform very few test-time interactions for quickly adapting the affordance priors to more accurate instance-specific posteriors by eliminating the dynamic and kinematic uncertainties.
- We propose an interaction-for-perception framework to predict dense geometry-aware, interaction-aware, and task-aware visual action affordance for 3D articulated objects.
🎖 Honors and Awards
- SenseTime Scholarship 2021-2022
- Pexpertmaker to Merit Student in Peking University 2021-2022
- Yanhong Li Scholarship in Peking University 2021-2022
- Benz Scholarship in Peking University 2020-2021
- Lee Wai Wing Scholarship in Peking University 2019-2020
- Merit Student in Peking University 2019-2020 and 2020-2021
- Second class prize in PKU-CPC 2021
- Silver medal in National Olympiad in Informatics (NOI), China Computer Federation, 2018
- Sep 2023 - now, University of Massachusetts Amherst.
- Aug 2019 - July 2023, undergraduate in Peking University.
- Mar 2023 - Aug 2023, Taichi Graphics, where I was lucky to work with Tiantian Liu and Yuanming Hu
- Apr 2022 - Oct 2022, with Dr. Jie Fu in Mila & BAAI.