As an Artificial Intelligence Researcher, my focus is on embodied intelligence, vla/world model, ensemble learning, multi-modality. I specialize in developing math-intensive proofs and strategies aimed at robust and private AI applications. My expertise also covers Multi-Modal Fusion, Deep Learning Interpretability, and Reliability. I am committed to designing AI systems that seamlessly integrate diverse data forms such as images, point clouds, text, and speech, ensuring their reliability and interpretability across various implementations.

I am driven by a passion to push AI technology to new frontiers, tackling complex challenges in real-world settings. My current projects emphasize Robotics AI Safety, AI for Science (AI4Sci), AI-generated content (AIGC), and Embodied Intelligence. These areas represent the forefront of applying advanced AI methods to ensure effective, safe, and ethical integration into everyday applications and complex scientific domains.

See more in my Blog.