About Me

Good morning, afternoon and night! This is Xinyuan Xia (Charlotte, 夏欣媛).

I am a senior student majoring in Artificial Intelligence at Shanghai Jiao Tong University. Recently, I’m taking an Internship at OpenRobotLab, advised by Prof. Jiangmiao Pang and Prof. Hanqing Wang, aiming to close the real to sim to real pipeline in Visual Language Navigation. Before that, I’ve joined summer school at NUS SOC and developed a cloud-native collaboration website, supervised by Prof. Richard T.B. Ma. I’ve also joined MedIA Lab, advised by Prof. Yi Hong and delve into llm-assisted medical image analysis (specifically in brain MRI).

If you are interested in any aspect of me, I am always open to discussions and collaborations. Feel free to reach out to me at char-lotte [at] sjtu.edu.cn

Research Interests

Machine Learning
Embodied AI
Computer Vision

My current research focuses on enhancing the spatial intelligence of embodied agents. In the long term, I aim to enable robots to exceed human capabilities in both operational precision and task variety, while exploring the learning and evolutionary pathways of machines. From an individual perspective, my goal is to alleviate repetitive labor, granting people more leisure time to pursue creativity and personal growth. From a societal standpoint, I seek to combat inequalities and contribute to social well-being by harnessing AI and robotics to drive positive change.

News

June 2025: Our paper ‘Rethinking the Embodied Gap in Vision-and-Language Navigation: A Holistic Study of Physical and Visual Disparities’ has been accepted to ICCV 2025. Much thanks to Liuyi Wang and Hanqing Wang for their guidance and support.
March 2025: Our paper ‘Brain MRI Segmentation with Language-Driven Detection and Context-Aware Descriptions’ has been accepted to ICASSP 2025. Much thanks to Qiang Fu and Prof. Yi Hong for their guidance and support.

Awards

2023：Shanghai Scholarship (Top 0.2% in Shanghai)
2023, 2024：Academic Excellence Scholarship of SJTU (Top 10% in SJTU)
2023：Second Prize in China Undergraduate Mathematical Contest in Modelling (Top 3% among 54000+ teams)