Wu Chaoran

The Digital Innovation Competition at New York University Shanghai centers on the intersection of Artificial Intelligence and Education, investigating the application of large language models (LLMs) in the educational domain to enhance teaching effectiveness A collaborative teaching platform—primarily developed and maintained by computer science students under faculty supervision—has been established to collect student assignments and compare them against reference solutions By analyzing the similarity between students’ current code Abstract Syntax Trees (ASTs) and correct answers (with a matching threshold of over 20%), the system delivers customized hints to learners who have demonstrated genuine effort, without disclosing the full solution This methodology cultivates a more authentic coding environment for computer science students and lays the foundation for an LLM-driven heuristic guidance system rooted in the Problem-Based Learning (PBL) framework Through collaboration with the LLM, students engage in solving real-world problems, thereby reinforcing their practical understanding and application of theoretical concepts