Smart Education in the Era of Human-Machine Collaboration: Opportunities and Challenges Brought by Generative AI
August 23 (Saturday) 3:45 PM-5:30 PM
Location: Grand Ballroom C, 3rd Floor

Introduction:
        Peng Xin is Vice Dean of the School of Computer Science and Technology at Fudan University and a Distinguished Professor of the Ministry of Education's Changjiang Scholars Program. He is a Distinguished Member of the China Computer Federation (CCF), Vice Director of the Software Engineering Committee, Standing Member of the Open Source Development Committee, and Vice Director of the Automotive Basic Software Division of the Society of Automotive Engineers of China. He is Co-Editor-in-Chief of the Journal of Software: Evolution and Process and serves on the Editorial Board of ACM Transactions on Software Engineering and Methodology, Empirical Software Engineering, Automated Software Engineering, and the Journal of Software. In 2016, he received the NASAC Young Software Innovation Award, was selected for the Shanghai Oriental Talent Program in 2023, and received the Zhongchuang Software Talent Award in 2024. His research focuses on intelligent software development, cloud native and intelligent operations and maintenance, ubiquitous computing software systems, and basic software for intelligent connected vehicles. His research has won numerous awards, including the IEEE Transactions on Software Engineering Best Paper Award, the ICSM Best Paper Award, the ACM SIGSOFT Outstanding Paper Award, and the IEEE TCSE Outstanding Paper Award. He served as the organizing committee chair and program co-chair for the 2022 and 2023 CCF ChinaSoft conferences, and as a member of the program committees for ICSE, FSE, ASE, ISSTA, ICSME, and SANER.

Report Title: Software Talent Cultivation in the AI Era

Report Introduction:

        The rapid development of large language models and multimodal large models is driving the advent of a new era of intelligence, characterized by the integration of man, machine, and objects and the intelligent interconnection of all things. Software-defined thinking and related technologies support the on-demand integration of human, machine, and physical resources, and intelligence is becoming a core feature of all types of integrated human-machine-physical systems. With the wave of intelligent development, will AI models, represented by neural networks, gradually replace software? Conversely, will generative AI spell the end of programming? This report, building on the background, explores these two questions and explains what kind of software talent is needed in the intelligent era.

Introduction:
        Jiang Hua is a professor at the Institute of Higher Education, Dalian University of Technology. He holds a doctorate in management and is a doctoral supervisor. He is the director of the Discipline Evaluation Center of Dalian University of Technology and the Education Evaluation and Development Strategy Research Center, a key think tank in Liaoning Province. He is also a standing member of the Education Evaluation Branch of the China Higher Education Society and the National Quality Monitoring Society. He is also a consultant for Liaoning Province's "Double First-Class" initiative. His primary research areas are higher education governance and education measurement and evaluation. He has written monographs such as "Resources and Efficiency: A Study of Performance Evaluation in Foreign Higher Education," "A Study of Performance Evaluation Plans and Implementation Strategies for Provincial Universities," and "Organizational Change in China's Private Higher Education: An Organizational Sociology Perspective."

Report Title: A Competency Structure Model and Evolutionary Patterns of University Teachers' Use of Generative Artificial Intelligence

Report Introduction: 

       The rapid rise of generative AI is driving profound changes in university teaching models, faculty roles, and educational governance. The role of faculty is gradually shifting from traditional classroom lecturers to instructional designers and collaborators, and generative AI tools are evolving from auxiliary backend tools to collaborative partners in the classroom. Against this backdrop, human-computer collaborative work methods are becoming increasingly important for university faculty in their daily teaching and research. This report will focus on the competency structure model and its evolutionary patterns for university teachers using generative AI, helping to clarify the underlying logic of teacher competency development and providing a reference for future professional development.

Introduction:
        Senior Internet Product Director, Physics Teacher, Master of Education from Beijing Normal University. With eight years of experience in the internet education industry, he has overseen the innovative design and implementation of numerous AI+ education products from scratch. Starting in 2023, we will focus on the big model field for education, overseeing the Jiuzhang big model product for Good Future and launching the Jiuzhang Aixue AI Math Teacher, dedicated to providing users with an interactive learning experience comparable to that of a real teacher.

Report Title: Exploring Personalized Learning in the Big Model Era

Report Introduction:  

        The education industry often refers to the "impossible triangle," which is the difficulty of achieving scale, personalization, and high quality simultaneously. However, with the advent of the big model era, the entire education industry is expected to gradually overcome this "impossible triangle." Specifically in knowledge learning scenarios, how can big models be used to address individual student challenges, such as misunderstanding knowledge and difficulty in completing exercises? The TAL Education Xueersi Big Model team is experimenting with scenarios like question-answering and targeted learning, accurately diagnosing students' weaknesses and providing one-on-one, real-time answers. This has led to a model with greater practical value and potential for personalized and targeted learning.

Introduction:
        Lu Yu is an Associate Professor and doctoral supervisor in the School of Education at Beijing Normal University. He holds a PhD in Computer Engineering from the National University of Singapore and is the Director of the Artificial Intelligence Laboratory at the Beijing Future Education Advanced Innovation Center. He has extensively engaged in research in the field of artificial intelligence and its educational applications at research institutions such as the Agency for Research and Development (A*STAR) in Singapore. He currently serves as an Associate Editor for several SCI/SSCI journals, including IEEE Transactions on Learning Technologies, and co-chairs several important international academic conferences in this field, including AIED. He has published three books in this field, both in Chinese and English, and over 100 academic papers in both Chinese and English. He has presided over one National Key R&D Program project, three National Natural Science Foundation projects, key projects of the Beijing Municipal Education Science Planning, and a Ministry of Education Humanities and Social Sciences Research Project.

Report Title: Constructing Human-Computer Collaborative Educational Agents

Report Introduction:

        With the rapid development of generative AI technologies such as large models, educational agents, as a key vehicle for AI-enabled education, are showing broad application prospects. This research focuses on the design and application of educational agents supported by large models, exploring their system architecture, key technologies, and model construction. On this basis, this paper focuses on typical human-machine collaborative education scenarios, such as classroom teaching and teacher research, analyzing their practical application effects, potential risks, and challenges. The aim is to provide theoretical references and technical solutions for the deep integration of intelligent agents and education.

Introduction:
        Xu Yonghui is a professor and doctoral supervisor at Shandong University. His research focuses on trustworthy artificial intelligence, multimodal large models, and graph neural networks. He has published over 100 papers at leading international AI conferences such as AAAI, IJCAI, and ICDE, and in prestigious journals such as TNNLS, TKDE, and TKDD, and has applied for over 20 core intellectual property rights. He has long served as a program committee member for leading AI and data mining conferences such as IJCAI, AAAI, and KDD, and as a reviewer for leading academic journals such as IEEE TKDE and TNNLS. He has undertaken over 20 national, provincial, and ministerial projects, including projects from the National Key R&D Program, projects from the National Natural Science Foundation of China, sub-projects from major projects of the National Natural Science Foundation of China, and the Alibaba Global Innovation Research Program. He has also been recognized with the Second Prize of the National Teaching Achievement Award, the First Prize of the Shandong Provincial Science and Technology Progress Award, and the Special Prize of the Shandong Provincial Teaching Achievement Award.

Report Title: Learning Science and Education Empowered by Big Models

Report Introduction:

        The groundbreaking development of AI big models has brought about a paradigm shift in the fields of learning science and education. This report systematically explores how big models, through their core capabilities (language understanding, knowledge reasoning, and multimodal processing), reconstruct the theoretical framework of learning science: deepening cognitive process modeling, enabling dynamic personalized learning, expanding context construction and metacognition cultivation, and promoting the transformation of education from "knowledge transfer" to "cognitive partnership." At a practical level, big models are reshaping three levels of scenarios: micro-teaching (intelligent guidance/dynamic content generation), meso-systems (curriculum mapping/adaptive assessment), and macro-ecological systems (resource equity/lifelong learning), significantly improving educational efficiency and accessibility. However, technological empowerment comes with significant challenges: model illusions, algorithmic bias, cognitive inertia risks, and data privacy issues urgently need to be addressed, while also requiring the reconstruction of teacher roles and evaluation systems. Future education requires a new paradigm of "human-machine collaboration"—developing reliable, specialized models, innovating teaching methods, and establishing ethical standards and policy governance frameworks to enable big models to become educational enhancement tools that support human intelligence.