The Research Institute for Publishing at Wuhan University (WHU) showcased three AI-powered publishing innovations at the 15th China International Digital Publishing Expo, which was held in Zhengzhou, Henan province, on Aug 28-31.

On Aug 29, at the technology and publishing innovation seminar, Fang Qing, director of the Research Institute for Publishing, presented a keynote speech on “Building Discipline Large Models Based on Artificial Intelligence”.

Together with representatives from Tongfang Knowledge Network Digital Technology Co, Fang officially unveiled the “Publishing Discipline Large Model”, attracting widespread attention from academia and industry.

This model, the first vertical large model in the publishing field, was jointly developed by WHU, the Tongfang Knowledge Network, and the Semantic Publishing & Knowledge Service Laboratory of the National Press and Publication Administration.

It integrates a publishing discipline corpus with general large-model capabilities from Huazhi, using AI to empower knowledge production. Through human–machine collaboration, it constructs a knowledge graph and “knowledge brain map”, creating the industry’s first semantic vector database for publishing. The model currently provides five key services — general question and answer, professional courses, knowledge graphs, publishing entries, and discipline applications — setting a new standard for AI-driven publishing scholarship.

Founded in 2005, the China International Digital Publishing Expo is the nation’s top platform for promoting digital publishing. This year, over 600 domestic and international enterprises showcased innovations in AI publishing, digital media, online literature, cultural products, and immersive mixed reality experiences.