Yining Hong

Papers

Total Citations

H-Index

About

Yining Hong is a rising star in artificial intelligence, whose research bridges the critical gap between 2D vision-language models and the rich, three-dimensional physical world. Her most notable contribution is the introduction of **3D-VLA: A 3D Vision-Language-Action Generative World Model**, a pioneering framework that moves beyond traditional flat inputs to integrate 3D spatial understanding with language and action. This work addresses a fundamental limitation of existing models, which often ignore the dynamics of the physical world and the complex relations between objects. By proposing a generative world model that learns the underlying physics and interactions of 3D scenes, Hong enables more robust and context-aware action prediction. Though her paper "3D-VLA" is recent (2024) and has already garnered 14 citations, signaling its immediate impact, her broader portfolio demonstrates a commitment to embodied AI and reasoning. Her work is essential for advancing robotics and autonomous systems, offering a path toward agents that truly understand and interact with the world as humans do.