Muyuan Lin
Papers
1
Total Citations
2
H-Index
1
About
Muyuan Lin is a rising researcher in embodied AI and robot learning, with a focus on bridging language, vision, and physical action. Their most cited work, "LOC-ZSON: Language-driven Object-Centric Zero-Shot Object Retrieval and Navigation" (2024), introduces a novel object-centric image representation that enables robots to navigate complex scenes and retrieve objects based on natural language commands—without prior training on specific objects. This contribution advances zero-shot generalization in robotic manipulation and navigation, a critical step toward adaptable, real-world AI systems. By fine-tuning vision-language models with object-centric losses, Lin’s approach tackles the challenge of handling cluttered, dynamic environments where traditional methods fail. Though early in their career, with 2 citations to date, the work has already garnered attention for its practical elegance and potential to scale. Lin’s research sits at the intersection of computer vision, natural language processing, and robotics, promising to make human-robot interaction more intuitive and robust. Their commitment to open-source benchmarks and reproducible methods further underscores a dedication to advancing the field transparently.
Research Focus
Key Achievements
Top Papers
- 1