Granular-ball based robust representation learning for social recommendation
About
Shiyan Wu is a Master's student in Computer Technology at Chongqing University of Technology (ranked in the top 5% of the major), focusing on the intersection of Large Language Model (LLM) applications, Natural Language Processing (NLP), and Granular Computing. Academic research specializes in robust recommendation algorithms based on Granular-Ball Computing (GBC), specifically utilizing topological homogeneity within granular-balls for pruning and denoising, with core findings accepted by the top-tier CAS Q1 journal EAAI. Engineering expertise encompasses mature AI implementation capabilities, including proficiency in vLLM inference deployment, Agent development frameworks, and the frontline development of an enterprise-level RAG intelligent assistant system featuring a dual-path hybrid retrieval architecture. Furthermore, a solid foundation in mathematical modeling is demonstrated by multiple national awards as a team captain in prestigious competitions such as the "Huawei Cup" and MathorCup, highlighting a strong integration of cutting-edge academic vision and practical coding skills.
Publications
Competitions
-
Dec. 2025
The 2025 "Huawei Cup" 22nd China Post-Graduate Mathematical Modeling Competition