Yuxing Long
Papers
6
Total Citations
159
H-Index
4
About
Yuxing Long is an emerging researcher at the forefront of embodied AI, specializing in robotic manipulation, visual language navigation, and zero-shot object navigation. His work focuses on bridging the gap between powerful foundation models and real-world robot capabilities, enabling machines to understand, reason, and act within complex environments. Long's most influential contribution, ManipLLM (65 citations), introduces an embodied multimodal large language model that dramatically improves the generalizability of robotic manipulation by accurately predicting contact points and end-effector directions across diverse object categories. His work on visual language navigation, "Discuss Before Moving" (48 citations), pioneers a multi-expert discussion framework that enhances robot planning and perception beyond single-model reasoning approaches. His research on pixel-guided navigation skills (36 citations) elegantly connects foundation models with locomotion abilities for zero-shot object navigation in home-assistance scenarios, while InstructNav further extends this vision to generic instruction-following in unexplored environments. With over 150 citations accumulated primarily within 2024, Long's rapid rise reflects the timeliness and significance of his research. His contributions are particularly valuable for students exploring how large language models can be grounded in physical robotic systems to achieve generalizable, real-world intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6