Xinpeng Zhang
Algorithm Engineer of ZHIPU AI-CodeGeeX
Xinpeng Zhang, Algorithm Engineer of Smart Spectrum AI-CodeGeeX, graduated from the University of Melbourne with a Master's degree in Artificial Intelligence. After graduation, he has been working on the research and application of natural language processing and data mining, focusing on the collection and cleaning of large-scale datasets as well as the alignment techniques of large-scale models, and his research results have been published in several international academic conferences. Currently, he is in charge of algorithm development for code project-level content understanding and generation in CodeGeeX team, aiming to promote the innovation and application of the technology.
Topic
CodeGeeX: Exploring and Applying RAG and Long-Text Models at the Code Project Level
CodeGeeX is an intelligent programming assistant meticulously crafted by the IntellectAI team, equipped with various functionalities such as code completion, code querying, code interpretation, and code retrieval, effectively enhancing programming efficiency. This sharing session will review the valuable experiences gained from the evolution of CodeGeeX from an open-source code model to an intelligent programming assistant. It will provide a detailed overview of CodeGeeX's application at the project level and explore the exploration and implementation of RAG and long-text capabilities in this process. Through this sharing, we hope to inspire thoughts on the application and development of intelligent programming assistants in the future programming field.