科学研究
学术报告
当前位置: 主页 > 科学研究 > 学术报告 > 正文

Catalyzing Quantum Communication: From Embezzling to Natural Language Processing

发布时间:2025-06-10 作者: 浏览次数:
Speaker: 肖运龙 DateTime: 2025年6月11日(周三)下午16:00-17:00
Brief Introduction to Speaker:

Yunlong Xiao holds a Ph.D. in Mathematical Physics from the Max Planck Institute for Mathematics in the Sciences (MiS) in Leipzig, Germany, where he conducted his research under the guidance of Prof. Jürgen Jost and Prof. Naihuan Jing, successfully completing his degree in February 2017. He also obtained a second Ph.D. in Pure Mathematics from South China University of Technology, working under the supervision of Prof. Naihuan Jing, and graduated in June 2017. Following his academic pursuits, he served as a Postdoctoral Fellow at the Institute for Quantum Science and Technology at the University of Calgary, Canada, working alongside Prof. Barry C. Sanders and Prof. Gilad Gour until August 2019. Continuing his research career, he worked as a Research Fellow at the Quantum Hub in the School of Physical and Mathematical Sciences at Nanyang Technological University, Singapore, under the mentorship of Prof. Mile Gu. Currently, Yunlong holds the position of Senior Scientist at the Quantum Innovation Centre (Q.InC), The Agency for Science, Technology, and Research (A*STAR), Singapore. His pursuits converge on quantum foundations, quantum causal inference, quantum resource theory, and quantum communication, and his contributions have found a home in journals such as Phys. Rev. Lett., Phys. Rev. Research, Phys. Rev. D, Light Sci. Appl., npj Quantum Information, Commun. Phys., Quantum Sci. Technol., New J. Phys. His influence extends beyond publications, as evidenced by his contributed talks at conferences including APS March Meeting, AQIS, QPL, and QIP.

Place: 新文科楼四楼402会议室
Abstract:We report three recent advances in quantum communication: first, we derive fundamental limits via dynamical quantum resource theory; second, we introduce a catalyst-driven framework that amplifies teleportation and channel capacities—achieving arbitrarily high fidelity with negligible disturbance to the catalyst; and third, we use transformer-based language models (BERT) to enhance superdense coding for large-text transmission over noisy channels without conventional error correction. Together, these results inaugurate a hybrid quantum–classical paradigm that unites catalytic protocols and NLP techniques to enable scalable, high-fidelity quantum networks.