Tsinghua University, in collaboration with Donghua University, has released the first fiber knowledge large model in China. - Tsinghua University
**Abstract:** On October 20, 2023, a significant milestone in the field of fiber science and technology was achieved with the launch of AlphaFiber 1.0, China's first knowledge model specifically designed for the fiber industry. The event took place at the Songjiang campus of Donghua University and was attended by nearly a hundred experts and scholars from both domestic and international fiber and artificial intelligence (AI) communities. **Key Events and Participants:** - **Launch of AlphaFiber 1.0:** The model, a collaborative effort between Professor Li Shi's team from the Department of Automation at Tsinghua University, the College of Materials Science and Engineering at Donghua University (which includes the State Key Laboratory of Modified Fibers), and the Jiangsu Jicui Advanced Fiber Materials Research Institute Co., Ltd., was officially unveiled. - **Attendees:** Notable attendees included Professor Meifang Zhu, the Dean of the College of Materials Science and Engineering at Donghua University and a member of the Chinese Academy of Sciences, Professor Li Shi from Tsinghua University, Yang Lin, Deputy Director of the Science Popularization and Science Communication Division at the Shanghai Science and Technology Commission, Professor Yan Lu from the University of Potsdam and Helmholtz Center, Vice President Xiaoping Duan from the China National Textile and Apparel Council, Executive Vice President and Chief Research Institute Director Songlin Wang from Hengyi Petrochemical Co., Ltd., Co-founder Yu Hu from iFLYTEK Co., Ltd., and Deputy Director Jialiang Zhou from the Jiangsu Jicui Advanced Fiber Materials Research Institute. **Technical and Functional Highlights:** - **AlphaFiber 1.0:** This model leverages the capabilities of large language models in natural language processing. It has been trained on an extensive dataset, including literature from books, journals, and patents, as well as experimental data such as raw material formulas, experimental procedures, and physical and chemical properties. Additionally, it incorporates knowledge systems, expert experiences, and industrial processes. - **Precision and Automation:** AlphaFiber 1.0 is designed to provide precise information and automate content generation in the fiber industry, thereby enhancing the dissemination and application of fiber science knowledge. - **Educational Impact:** Professor Li Shi highlighted that AlphaFiber 1.0 can assist in explaining complex fiber science concepts and expand teaching methods, offering personalized guidance to educators and students. This is expected to revolutionize educational practices in the field of fiber materials. **Release Ceremony and Interactive Session:** - **Ceremony:** The official release was marked by a lighting ceremony involving Professor Meifang Zhu, Professor Li Shi, Yang Lin, Professor Yan Lu, Vice President Xiaoping Duan, and Director Songlin Wang. - **Interactive Q&A:** Following the release, the attendees engaged in an interactive session, asking questions and discussing the potential applications and implications of AlphaFiber 1.0. **Future Prospects:** - **Continuous Improvement:** The developers emphasized that AlphaFiber 1.0 will undergo continuous upgrades and optimizations to better serve the fiber materials community. - **Support for Science Popularization:** The model is expected to play a crucial role in popularizing fiber materials science, making it more accessible and understandable to a broader audience. **Conclusion:** The launch of AlphaFiber 1.0 represents a significant advancement in the integration of AI and fiber science. It promises to enhance research, production, and education in the fiber industry by providing a powerful tool for knowledge dissemination and content generation. The collaboration between leading institutions and companies in China and Germany underscores the global interest and potential impact of this innovative technology.
