Jasmine Shone
Papers
1
Total Citations
6
H-Index
1
About
Jasmine Shone is a rising robotics researcher whose work centers on imitation learning, object-relative manipulation, and keypoint-based representations for robot generalization. Her most-cited paper, "KALM: Keypoint Abstraction Using Large Models for Object-Relative Imitation Learning" (2025, 6 citations), tackles a fundamental challenge in robotics: enabling robots to generalize across novel object configurations and instances in diverse tasks and environments. Shone’s key contribution lies in leveraging large models to abstract task-relevant keypoints, creating a succinct yet powerful representation that captures essential object features and establishes a stable reference frame for imitation. This approach allows robots to transfer learned skills to unseen scenarios without exhaustive retraining, bridging the gap between demonstration and real-world deployment. Though early in her career, Shone’s work has already garnered attention for its innovative fusion of large language/vision models with classical keypoint methods, offering a scalable path toward more adaptable and intelligent robotic systems. Her research promises to advance autonomous manipulation in unstructured settings, from household chores to industrial assembly.
Research Focus
Key Achievements
Top Papers
- 1