Juan Zhai is an Assistant Professor in the Manning College of Information & Computer Sciences (CICS) at University of Massachusetts, Amherst (UMass). She co-directs the Laboratory for Advanced Software Engineering Research (LASER) lab. She is also a member of UMass NLP group. She is broadly interested in Software-AI Synergy, Software Dependability, Software Reasoning, ML/NLP for Software Engineering, and Trustworthy AI.
I am always looking for students to work with. Drop me an email if you are interested in working with me!
Research Areas
Formal Specification Synthesis
The fundamental challenge in developing high-quality software is ensuring that its behavior aligns with the intended specifications. At the heart of this challenge is the persistent absence of formal specifications which are precise, unambiguous definitions of expected behavior. Formal specifications are essential not only for traditional activities like debugging, testing, and verification, but also for enabling large language model agents to generate correct and reliable code. Despite their importance, formal specifications remain rare in practice due to the complexity of authoring and maintaining them, especially in fast-evolving, real-world systems.
My research addresses this gap by developing techniques to automatically synthesize formal specifications. Our tool C2S, successfully translates natural language comments into formal specifications, improving test oracle generation, reducing false positives in automated testing, and enhancing static taint analysis. We have also generate LTL formulas for IoT systems from natural language commands, and are now exploring LLM-driven specification synthesis.
Comment Generation and Maintenance
Comments often convey high-level semantic intent, but they are frequently outdated, incomplete, or inconsistent with code. Our research focuses on inferring and maintaining behaviorally accurate comments to support understanding, maintenance, and reasoning.
We developed CPC, a novel software reasoning method that enables bidirectional analysis across comments and code implementation. To keep comments updated as software evolves, we developed LLMCup, a framework that automatically updates comments using large language models with a ranking-based refinement strategy.
Trustworthy AI Analysis and Improvement
As AI-driven software systems increasingly impact critical aspects of society, ensuring their trustworthiness has become both a technical and moral imperative. Our research spans the AI stack, with a focus on improving correctness, robustness, and fairness through practical, scalable tools and techniques.
Deep Learning Frameworks Testing
We develop tools to test and enhance the reliability of core AI infrastructure, including ModelMeta, DevMuT, DLJSFuzzer, and Citadel, addressing bugs and inefficiencies in popular deep learning frameworks.
Training Diagnosis and Repair
To prevent wasteful or faulty training runs, we create tools such as AutoTrainer and Dream that proactively detect and repair training issues before models are fully trained.
Bias Detection and Mitigation in ML Systems
We design automated methods to detect, explain, and mitigate bias across the ML lifecycle, from training and pruning to fine-tuning. Our work addresses biases in linguistic style, prompt language and provider preferences.
Current PhD Students
Join Our Research Group
For Prospective PhD Students
I am actively seeking motivated PhD students interested in Software Engineering, AI Safety, and Trustworthy AI research. Strong background in programming and interest in research is essential.
Contact: Please email me with your CV, transcript, and research interests.
Application: Apply through the UMass CS PhD program and mention my name in your application.
For Undergraduate and Master's Students
I welcome undergraduate and Master's students interested in research opportunities, independent studies, thesis projects, visiting opportunities, or internships throughout the year.
Contact: Please email me with your CV, transcript, and research interests.
Research Areas Available
- • Formal Specification Synthesis
- • Comment Generation and Maintenance
- • AI/ML System Testing and Debugging
- • Bias Detection and Mitigation
- • Software-AI Synergy
💡 I am always looking for students to work with!
Drop me an email at juanzhai at umass dot edu if you are interested in working with me!
Publication Overview
Course History
Graduate Level Courses
Hot Topics in Software Engineering Research
Fall 2024Theory and Practice of Software Engineering
Fall 2025, Fall 2024, Spring 2024Undergraduate Level Courses
Software Engineering
Spring 2023, Spring 2020Data Structure
Multiple SemestersIntroduction to Computer Science
Multiple SemestersComplete Teaching Timeline
Note: ×2 indicates I taught two separate sections of the course that term
Conference Program Committee & Reviewing
Software Engineering Conferences
International Conference on Software Engineering
2026, 2025Symposium on the Foundations of Software Engineering
2025, 2024International Symposium on Software Testing and Analysis
2026, 2025, 2024International Conference on Automated Software Engineering
2025, 2024, 2023New Ideas and Emerging Results Track
2025Predictive Models and Data Analytics in Software Engineering
2025International Workshop on Large Language Models for Code
2026International Conference on Automation of Software Test
2023ICSE Demo Track
2022AI/ML Conferences
Conference on Computer Vision and Pattern Recognition
2024, 2023International Conference on Learning Representations
2025, 2024Conference on Neural Information Processing Systems
2024, 2023International Conference on Machine Learning
2022Journal Editorial & Reviewing
Transactions on Software Engineering and Methodology
2024, 2023, 2022, 2021Transactions on Software Engineering
2020Journal of Systems and Software
2024, 2023, 2019Journal of Software
2020Empirical Software Engineering
2019