Yinfei Yang

Apple (United Kingdom), Nanjing University of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Yinfei Yang is a researcher at the forefront of embodied AI and 3D scene understanding, with a focus on bridging vision, language, and robotics. His most impactful work tackles the challenge of Vision-and-Language Navigation (VLN), where he pioneered a new path to scaling agents that follow natural-language instructions in photorealistic environments. By leveraging synthetic instructions and imitation learning, his 2023 paper (31 citations) addresses the critical scarcity of human instruction data, advancing the development of robots capable of real-world navigation. In parallel, Yang has made significant contributions to 3D scene synthesis, introducing simple yet effective methods for generating immersive indoor scenes from just one or a few images. His 2023 work (19 citations) enables high-resolution novel-view synthesis, including far extrapolations, while maintaining 3D consistency—a key step for virtual reality and robotics. With a career spanning both foundational theory and applied systems, Yang’s research is shaping how machines perceive, navigate, and interact with complex environments. His achievements reflect a deep commitment to solving data scarcity and scalability challenges, making him a rising voice in the intersection of computer vision, natural language processing, and robotics.