Mengying Lin
Papers
2
Total Citations
5
H-Index
2
About
Mengying Lin is a researcher advancing the field of embodied AI, with a primary focus on object-goal navigation for mobile robots. Her work addresses a critical bottleneck in autonomous navigation: enabling agents to efficiently locate target objects in unfamiliar environments without relying on pre-learned spatial layouts. Lin’s major contribution lies in integrating Large Language Models (LLMs) with object affinities transfer—a novel approach that allows robots to leverage commonsense knowledge about object co-occurrence (e.g., knowing a cup is often near a coffee machine) to generalize beyond their training data. Her most-cited paper, “Advancing Object Goal Navigation Through LLM-enhanced Object Affinities Transfer” (2024, 3 citations), and its 2025 follow-up (2 citations) demonstrate how LLM-enhanced heuristics outperform traditional network-based and graph-based methods in complex, unseen layouts. By moving beyond static, data-driven affinities, Lin’s work offers a scalable path toward more adaptable and intelligent robotic systems. Her research is particularly impactful for students and engineers working on embodied perception, semantic navigation, and human-robot interaction, providing a clear bridge between large-scale language models and real-world robotic decision-making.
Research Focus
Key Achievements
Top Papers
- 1
- 2