Xinran Jiang
Papers
1
Total Citations
4
H-Index
1
About
Xinran Jiang is a rising researcher at the forefront of robotics and machine learning, specializing in learning from demonstration and generative modeling. Her work addresses a critical bottleneck in robotics: the scarcity of expensive, action-labeled robot data. Jiang’s key contribution, exemplified by her highly cited 2025 paper "GraphMimic: Graph-to-Graphs Generative Modeling from Videos for Policy Learning," introduces a novel framework that leverages abundant, unlabeled video data. By modeling visual demonstrations as structured graphs and generating diverse, actionable policies, her approach enables robots to learn complex manipulation skills without direct action supervision. This work, already garnering 4 citations shortly after publication, promises to dramatically lower the cost of robotic skill acquisition. Jiang’s research sits at the intersection of computer vision, graph neural networks, and reinforcement learning, offering a scalable pathway toward more adaptable and intelligent autonomous systems. Her innovative use of video as a rich, untapped data source marks her as a notable emerging voice in the field, with potential for significant impact on real-world robotic applications.
Research Focus
Key Achievements
Top Papers
- 1