Yunshuang Nie
Papers
1
Total Citations
32
H-Index
1
About
Yunshuang Nie is a rising researcher in Embodied AI, with a primary focus on Vision-and-Language Navigation (VLN)—a field that challenges agents to interpret natural language instructions and navigate complex 3D environments. Her most cited work, "NavCoT: Boosting LLM-Based Vision-and-Language Navigation via Learning Disentangled Reasoning" (2025, 32 citations), makes a significant contribution by addressing a critical bottleneck in LLM-driven navigation: the entanglement of perception and reasoning. Nie’s approach introduces a disentangled reasoning framework that separates visual grounding from sequential decision-making, enabling large language models to more effectively parse spatial cues and linguistic commands. This innovation not only improves navigation accuracy but also enhances model interpretability, a key step toward trustworthy embodied agents. Her work has quickly garnered attention, accumulating citations within its first year, signaling its impact on the VLN community. By bridging the gap between high-level language understanding and low-level robotic control, Nie is helping to define the next generation of intelligent, instruction-following agents—research that holds promise for applications in assistive robotics, autonomous exploration, and human-robot interaction.
Research Focus
Key Achievements
Top Papers
- 1