Yu-Xiao Liu
Papers
1
Total Citations
7
H-Index
1
About
Yu-Xiao Liu is a pioneering researcher at the forefront of robot manipulation and embodied AI, with a focus on developing generalist agents that bridge video, language, and action. Their most notable contribution is the GR-2 model, a generative video-language-action framework that leverages web-scale knowledge—pre-trained on 38 million Internet video clips and over 50 billion tokens—to achieve versatile, real-world robot control. This work, published in 2024 and already garnering 7 citations, represents a paradigm shift in robotics by enabling agents to learn world dynamics from vast, unlabeled video data, dramatically improving generalization across tasks. Liu’s research sits at the intersection of computer vision, natural language processing, and robotics, advancing the goal of creating robots that can understand and act in unstructured environments. Their approach, which combines large-scale pre-training with fine-tuned manipulation skills, has set a new benchmark for generalist robot agents, inspiring follow-up work in embodied AI. With a growing citation impact and a focus on scalable, data-driven methods, Liu is shaping the future of intelligent, adaptable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1