This speech explores the practical implementation of Science Education Agents to enhance scientific literacy, addressing China's strategic goal of AI-driven educational transformation. Building upon our previously established cognitive framework, this presentation introduces “Qizhi” (启智), a newly deployed intelligent tutoring agent system dedicated to primary science education. Driven by an underlying science education large language model (SE-LLM) fine-tuned on a curriculum-aligned dataset (which previously demonstrated 89.1% accuracy in scientific reasoning), “Qizhi” translates cutting-edge AI capabilities into autonomous pedagogical practice through three core functionalities: accurate open Q&A for precise scientific explanations; heuristic scaffolding utilizing Socratic dialogue to dynamically guide student inquiry rather than providing direct answers; and real-time scientific literacy assessment to evaluate conceptual mastery and critical thinking. By transitioning from theoretical model training to an applied, closed-loop agentic learning environment, the system effectively fosters students’ curiosity and logical reasoning. The presentation concludes by sharing empirical insights from Qizhi’s deployment and outlining future directions—such as VR-integrated experiential learning and multi-agent collaborative inquiry—solidifying intelligent educational agents as transformative tools for next-generation science education.

Zhi Liu is a Professor, PhD supervisor, and Deputy Head of the Department of Data Science within the Faculty of Artificial Intelligence in Education at Central China Normal University (CCNU). He has maintained a long-term appointment as a Guest Researcher at the Humboldt University of Berlin since 2017 and is currently a Visiting Scholar at the German Research Center for Artificial Intelligence (DFKI) for the 2025-2026 term. Specializing in text mining, educational data mining, and intelligent tutoring systems, Professor Liu has published over 60 SCI/SSCI-indexed papers in top-tier journals such as Knowledge-Based Systems, Computers & Education, and IEEE Transactions on Learning Technologies, including six ESI Highly Cited Papers (Top 1%). His scholarly impact is further underscored by his recognition as a Top 1% Highly Cited Scholar by CNKI (2024-2025). As a leading researcher, he serves as the Principal Investigator for the National Key R&D Program of China (2030 Major Projects) and several projects under the National Natural Science Foundation of China. Professor Liu holds various academic leadership roles, including Chair of the Organizing Committee for ICET, Guest Associate Editor for Frontiers in Artificial Intelligence, and Editorial Board member for Discover Education. His distinguished contributions have earned him numerous prestigious accolades, most notably the First Prize of the Hubei Social Science Outstanding Achievement Award (2025), the First Prize of the Science and Technology Progress Award of Hubei Province (2024), and the First Prize of the Hubei Province Teaching Achievement Award (2022).