About Me
Here is Shitou Zhang (Patrick, 张石头).
I am a researcher specializing in BCI-based AI, working under the mentorship of Prof. TP Jung, Prof. John K. Zao, Dr. Zuchao Li, and Prof. Ping Wang. Previously, my focus was on training large language models before shifting to the intersection of BCI and AI. I am currently developing infrastructure to accelerate research and development of BCI-based AI models. If you are interested in this field, please feel free to contact me via email to join our efforts.
Education Background
- Sep. 2022 - Jun. 2023: Department of Information Engineering, the Chinese University of Hong Kong (MSc)
- Sep. 2018 - Jun. 2022: School of Information Management, Wuhan University (BSc)
Research Interests
- Foundation Model Pretraining
- Model Personalization
“Trainging language models is to achieve scaling and avoid bottlenecks.” – Jared Kaplan
My research interests center around training foundational models and adapting them for practical applications. BCI sensors and the volume of BCI data are expected to grow exponentially in the next decade. The intrinsic characteristics, generation, processing, and application of BCI data are vastly different from stationary modalities such as text and image. Through my research in training and adapting foundational models, I aim to contribute to building a closed-loop ecosystem for the R&D of BCI-based AI models.
News and Updates
- Dec 2023: One paper is accepted by IPM.
- Nov 2023: ArcMMLU paper is available on arXiv, check out the introduction video made by PaperWeekly.
- Nov 2023: Our ArcMMLU benchmark is open-sourced, available on Github and HuggingFace
- Nov 2023: Our [Archives Meet GPT] paper is accepted by iConference 2024.
- Oct 2023: Our LingoWhale-8B is open-sourced, available on Github and HuggingFace
- Aug 2023: BATGPT new version is updated on arXiv
- Jul 2023: Our BATGPT-15B-sirius is open-sourced, available on GitHub and HuggingFace.
- Jul 2023: ArcGPT paper is available on arXiv.
- Jul 2023: BATGPT paper is available on arXiv.
- May 2023: One paper is accepted by ACL SRW 2023.