Page Not Found
Page not found. Your pixels are in another canvas.
Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Junda Zhao is a PhD candidate at the University of Toronto working on reliable testing and verification for AI coding agents. Seeking a postdoctoral position starting in late 2027.
Posts
portfolio
publications
A New Fuzzy Belonging-based Multi-view K-means Clustering Algorithm
IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON 2019)
A multi-view k-means clustering algorithm based on fuzzy belonging.
Transformer Models for Mining Intents and Predicting Activities from Emails in Knowledge-Intensive Processes
Engineering Applications of Artificial Intelligence, 2024
Transformer models that detect user intents in emails, using user actions as weak supervision, and predict the next activities in knowledge-intensive business processes.
Variational Prefix Tuning for Diverse and Accurate Code Summarization Using Pre-trained Language Models
Journal of Systems and Software, 2025
A parameter-efficient method that lets pre-trained language models of code generate diverse yet accurate sets of code summaries, by integrating a conditional variational autoencoder as a modular prefix component.
Evaluating and Mitigating the Misguidance Effect of Buggy Code in LLM-Generated Unit Tests
ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2026)
Buggy code steers LLMs toward tests that confirm the bug instead of exposing it. We quantify this misguidance effect across 11 LLMs on Defects4J and mitigate it with specification-based test generation.
Do Coverage and Mutation Scores of LLM-Generated Test Suites Correlate with Their Effectiveness? (Replicability Study)
ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2026)
ACM SIGSOFT Distinguished Paper Award. A large-scale replication of the relationships among coverage, mutation score, test-suite size, and real-bug detection, using more than 100,000 test cases generated by 11 LLMs.
Accurate Specification Recovery for Effective Test Generation
SPLASH/ISSTA 2026 Doctoral Symposium
Doctoral symposium paper describing my dissertation research.
talks
Evaluating and Mitigating the Misguidance Effect of Buggy Code in LLM-Generated Unit Tests
Published:
teaching
MIE350: Design and Analysis of Information Systems
Undergraduate course, Teaching Assistant, University of Toronto, Department of Mechanical & Industrial Engineering, 2022
Teaching Assistant in Fall 2022. Responsibilities included leading tutorial sessions, helping to prepare teaching materials, and grading projects and tests.
MIE1517: Introduction to Deep Learning
Graduate course, Teaching Assistant, University of Toronto, Department of Mechanical & Industrial Engineering, 2024
Teaching Assistant in Winter 2024, and in Fall and Winter terms since Fall 2024. Responsibilities include leading tutorial sessions, helping to develop and improve course materials, and grading projects and tests.
MIE370: Introduction to Machine Learning
Undergraduate course, Teaching Assistant, University of Toronto, Department of Mechanical & Industrial Engineering, 2025
Teaching Assistant in Fall 2025. Responsibilities included leading tutorial sessions, helping to prepare teaching materials, and grading projects and tests.