Projects

(back to main)

Zhiyu's research interest has been focused on the intersections of automated program repair, automatic programming, and computer science education. His research work on program repair for intelligent tutoring (TOSEM22, ISSTA23, program repair of AI programs (ICSE23), explorations on AI (i.e. LLM) programming (ISSTA24a, ISSTA24b) have been published in top-tier software engineering venues. His has led the development of Intelligent Tutoring System to boost Computer Science education at scale via SE and LLM techniques, which has been deployed in NUS.

Explorations on Program Repair and AI Programming

Program Repair for Intellient Tutoring of Programming

  1. Providing correctness gurantee for automated feedback in CS-1 scenario. (Verifix, TOSEM 2022)
  2. Automated conceputal-based grading for CS-1 programming assignment. (ConceptGrader, ISSTA 2023)
  3. Build realistic Intelligent Tutoring System via long-term Software Engineering course. (Experience on ITS, arXiv 2023)

In addition to the research projects, we also pursuit realistic impact in Computer Education and Teaching. We have built an Intelligent Tutor System (ITS) that integrates all above research outcomes throughout a new and novel Software Engineering course (CS3213 Foundations of Software Engineering) and deployed it in NUS to facilitate programming teaching (Experience on ITS, arXiv 2023). We are still actively evolving the ITS for next generation Computer Education!

Techniques on Trustworthy Automated Programming

  1. Evaulation of automated repair for LLM-generated programs. (LMDefects, ICSE 2023)

We are working to link the LLM-powered automated programming and ITS to reshape the programing teaching for future demand.