Yang, Shu
Zhou, Fengtao
Mayer, Leon
Huang, Fuxiang
Chen, Yiliang
Wang, Yihui
He, Sunan
Nie, Yuxiang
Wang, Xi
Jin, Yueming
Sun, Huihui
Xu, Shuchang
Liu, Alex Qinyang
Li, Zheng
Qin, Jing
Teoh, Jeremy YuenChun
Maier-Hein, Lena
Chen, Hao
Funding for this research was provided by:
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Research Grants Council of Hong Kong (No. G-HKUST605/24)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Innovation and Technology Commission (No. GHP/006/22GD and ITCPD/17-9)
Article History
Received: 21 August 2025
Accepted: 22 January 2026
First Online: 4 February 2026
Competing interests
: S.Y. and H.C. are inventors on a patent application related to this work that is currently being prepared for filing via the Patent Cooperation Treaty (PCT) route, with The Hong Kong University of Science and Technology as the applicant. The application will cover the pre-training framework, model architecture, and pre-trained parameters presented in this manuscript. All other authors declare no competing interests.