Xiwen Liang

Papers

Total Citations

120

H-Index

About

Xiwen Liang is a leading researcher in embodied AI and visual navigation, with a focus on enabling intelligent robots to navigate complex 3D environments using natural language instructions. Their most notable contribution is the SOON framework (Scenario Oriented Object Navigation with Graph-based Exploration), which tackles one of the field's "holy grail" challenges: allowing robots to navigate toward a language-guided target from any starting point, rather than a fixed location. This work, published in 2021, has garnered 115 citations, reflecting its significant impact on advancing autonomous navigation systems. By integrating graph-based exploration with scenario-oriented reasoning, Liang's research bridges the gap between static navigation benchmarks and real-world, dynamic environments. Their work is highly regarded for pushing the boundaries of how robots understand and interact with their surroundings through language, making it a cornerstone for students and researchers interested in embodied AI, human-robot interaction, and intelligent navigation systems. Liang's contributions continue to inspire new approaches to building more adaptive and human-like robotic agents.