Passion and Thinking are always my best teachers

I'm particularly interested in Explainable AI (now about generative foundation models). I'm very obsessed with exploring and understanding everything! In this era, AI is dependent on data-driven deep learning methods, which is inexplicable and uncontrollable. I believe that theory-guided intelligence is the future of AI. My goal is to embrace the AGI via theory4practice methods!🌞The most noteworthy thing is that I'm a devoted advocate of Slow Science, or rather, a faithful of Thinking.

Currently, I'm an incoming CS PhD student at UW-Madison. Before it, I was a Research Assistant advised by Prof. Yang Yuan at IIIS, THU (Category/ML Theory & Explainable AI 2023.05 ~ 2023.05). I start my research working on multimodal learning (2021.11~2023.06).

Note: Since now is the Oppenheimer Time of AI, I become actively seeking like-minded collaborators from both Theory and Practice aspects - If you think we have the same reseach goal and style, please feel free to contact me! :)

⭐ Recent News ⭐

07/2023. One paper has been accepted by ICCV 2023 as poster!

03/2023. One paper has been accepted by ICME 2023 as Oral! 

Selected Papers

* for equal contribution. 

Few shot font generation via transferring similarity guided global style and quantization local style [paper]

• Wei Pan, Anna Zhu, Xinyu Zhou, Brian K. Iwana, Shilin Li.

• International Conference on Computer Vision (ICCV) 2023 

Scene Text Involved "Text"-to-Image Retrieval through Logically Hierarchical Matching [paper]

• Xinyu Zhou, Anna Zhu, Huen Chen, Wei Pan.

• Oral in IEEE International Conference on Multimedia and Expo (ICME) 2023

Selected Awards

• 04/2023. Outstanding Undergraduate Students of the Times.

• 09/2022. National Scholarship.

• 04/2022. Intelligent Base: "Future Star ", Awarded by Ministry of Education.

Academic Service

ACM MM'23-24, PRCV'23-24, NeurIPS'23, ICPR'24, ICME'24.

Hobbies

My interests are quite broad - Thanks to my ability to quickly learn and master the vast majority of things. [Learn More]