One paper accepted by ICLR 2025 Workshop World Model; two papers accepted at CVPR 2025; a new paper on arXiv titled 'Memory Helps, but Confabulation Misleads: Understanding Streaming Events in Videos with MLLMs'; one new paper accepted by WACV 2025.
Research Experience
Starting an internship at Amazon London; previous research involved video understanding and multimodal queries.
Education
Bachelor's degree (2018) from Zhejiang University, China; Master's degree (2021) from Technical University of Munich, Germany; Currently pursuing a PhD at Ludwig-Maximilian University (LMU Munich/University of Munich), supervised by Prof. Volker Tresp.
Background
Research interests include Video Understanding and Multimodal Reasoning, at the intersection of Computer Vision and Natural Language Processing. Originally from Hunan, China.
Miscellany
Hobbies include plants, Crusader Kings III, traveling, cooking; has a cute dachshund; open to any collaboration and full-time job opportunities.