Grace Tang
Papers
3
Total Citations
8
H-Index
2
About
Grace Tang is a pioneering researcher whose work bridges the critical gap between medical physics and robotic intelligence. Her early contributions in radiation oncology introduced a novel method for intrafractional 3D localization using kilovoltage digital tomosynthesis, enabling real-time patient motion monitoring during sliding-window intensity modulated radiation therapy (IMRT). This work, published in 2015, laid the groundwork for more precise and adaptive cancer treatments, demonstrating her ability to solve complex clinical challenges through innovative imaging sequences. More recently, Tang has emerged as a leader in robotics and artificial intelligence with her groundbreaking KALIE framework (Keypoint Affordance Learning from Imagined Environments). This approach fine-tunes vision-language models for open-world manipulation without requiring any robot data, a significant leap toward building generalist robotic systems capable of handling novel objects. By leveraging large pre-trained models, KALIE enables robots to understand affordances and execute tasks in unstructured environments, marking a paradigm shift in robotic learning. With over 8 citations across her most-cited works, Tang’s research exemplifies a rare interdisciplinary impact—from improving cancer treatment precision to advancing the frontiers of autonomous manipulation. Her work continues to inspire students and researchers at the intersection of medical physics, computer vision, and robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3