Current research focuses on the trustworthiness (robustnesss, safety, and security) assessment of AI assistants/systems using statistical, data science, and systems/software engineering techiques and on improving the productivity of knowledge professionals (data, software engineers, etc.) using robust and secure AI assistants.
Current and past research contributed to the following areas:
Engineering Trustworthy AI Assistants
- Hallucination Detection in Large Language Models with Metamorphic Relations. ACM International Conference on the Foundations of Software Engineering (FSE), 2025 ACM SIGSOFT Distinguished Paper award
- ChatGPT Incorrectness Detection in Software Reviews. 46th International Conference on Software Engineering (ICSE), 12 Pages. 2024
- Applications and Challenges of Fairness APIs in Machine Learning Software. ACM Transactions on Software Engineering and Methodology (TOSEM), 2025.
- PCS: Perceived Confidence Scoring of Black Box LLMs with Metamorphic Relations [Preprint]
- Bias Testing and Mitigation in Black Box LLMs using Metamorphic Relations [Preprint]
AI for Software Issue Management
- ChatGPT Inaccuracy Mitigation during Technical Report Understanding: Are We There Yet?47th International Conference on Software Engineering (ICSE), 12 Pages. 2025 Best CAS Project of the Year Award at the IBM TechXChange 2024
- Stack Trace-Based Crash Deduplication with Transformer Adaptation [Preprint]
- TriagerX: Dual Transformers for Bug Triaging Tasks with Content and Interaction Based Rankings [Preprint] Best CAS Project of the Year Award at the IBM TechXChange 2024
AI for Coding and Security
- An Empirical Study of Reasoning Steps in Thinking Code LLMs [Preprint]
- LLM For Loop Invariant Generation and Fixing: How Far Are We? [Preprint]
- LLM Assisted Coding with Metamorphic Specification Mutation Agent. [Preprint]
- PAGENT: Learning to Patch Software Engineering Issue Agents [Preprint]
Quality and Security of Crowd Technical Knowledge
- Automatic API Usage Scenario Documentation from Technical Q&A Sites. ACM Transactions on Software Engineering and Methodology (TOSEM). 2020.
- Automatic Mining of Opinions Expressed about APIs in Stack Overflow. IEEE Transactions on Software Engineering (TSE). 2019. [Stack Overflow Blog]
- Automatic Summarization of API Reviews. 32nd IEEE/ACM International Conference on Automated Software Engineering (ASE). Pages 159-170. [Stack Overflow Blog] Nominated for ACM SIGSOFT Distinguished Paper Award
- A Large-Scale Study of IoT Security Weaknesses in Crowd-Shared C/C++ IoT Code Examples. ACM Transactions on Software Engineering and Methodology (TOSEM), 2024.
- Reputation Gaming in Stack Overflow. ACM Transactions on Software Engineering and Methodology (TOSEM), 2024.
- An Empirical Study of C++ Vulnerabilities in Crowd-Sourced Code Examples. IEEE Transactions on Software Engineering (TSE). 2020. [BBC News]