The recent launch of large model technology represented by GPT-4 has had a profound impact on software development. What is the program generation ability of large models? What are the problems in program analysis and generation? What changes will happen in the future software development? The Peking University Program Understanding and Generation Research Team is an early pioneer and continuous contributor to deep learning-based program understanding and generation. Based on their own research experience, the speaker briefly elaborates on the research process and development status of large-model-based program understanding and generation methods, focusing on exploring the problems currently existing in large-model-based program design automation.
Ge Li | Professor and doctoral supervisor of the School of Computer Science at Peking University Chang Jiang Scholar appointed by the Ministry of Education.
Roundtable: National Capital Supports the Development of China's Open Source Ecology, Industry and Enterprises.
Xuemei Lin | General Manager of National Dispatching Science and Technology Reform Fund
Kexiang Yu | Assistant General Manager of TEDA Industrial President of Capital Business Division.