Likun Tang
Papers
1
Total Citations
2
H-Index
1
About
Likun Tang is a rising researcher at the forefront of embodied AI and vision-language navigation, with a focus on enabling home-assistance robots to perform complex, long-horizon tasks. Their key contributions center on integrating Large Language Models (LLMs) with visual perception to improve robotic exploration and object-searching capabilities. Tang’s notable work, "ImagineNav: Prompting Vision-Language Models as Embodied Navigator through Scene Imagination" (2024), introduces a novel paradigm where LLMs are prompted to imagine scene layouts, allowing robots to reason about unseen environments and plan more efficient paths. This approach bridges commonsense inference with spatial reasoning, advancing the field of visual navigation. While still early in their career, with 2 citations on this pioneering paper, Tang’s research has already demonstrated significant potential to impact real-world robotics, offering a scalable solution for household automation. Their work stands out for its creative use of scene imagination, pushing the boundaries of how vision-language models can be deployed as embodied navigators.
Research Focus
Key Achievements
Top Papers
- 1