Bowen Xu
Assistant ProfessorSoftMax Lab
Software Engineering Group at NCSU
Department of Computer Science
College of Engineering
Office: Engineering Building II (EB2), Room 3228
Email: bxu22@ncsu.edu
My team has developed a framework named Evaloop to fairly assess LLMs' robustness for programming. Based on this, we are actively maintaining a leaderboard with the results on more than 100 LLMs. Check it out at https://evalooop.github.io!
Research Focus
My research interests span software engineering, machine learning, and program analysis. My primary research interests are:- AI for Code (Model layer)
- Privacy Risk [AIWare-25] [ICSE-26]
- Backdoor Attack and Defense on Large Code Models [TSE-24]
- Code Data Interpretation and Quality [ASE-23] [ICPC-22]
- Code Representation Learning [TOSEM-23] [ICSME-23] [SANER-22] [FSE-22] [TSE-21]
- Model Compression [ASE-22] [ICSE-SEIS-24] [ICSME-25]
- Software Quality Management (Application layer)
- Bug Management and Analysis [TOSEM-23] [TSE-25]
- Vulnerability Detection [ASE-23]
- Vulnerability Repair [ICSE-24] [TSE-23] [AIWare-24]
- Safety of AI-enabled Software System (System layer)
- Violation Identification of AI System [TOSEM-25]
- Backdoor Attack in Reinforcement Learning System[S&P-24]
- Backporting in Robot Operating System [Forge 2026]
- Understand and Assist Developers' Activities (Human-centered layer)
- Developer-AI Interaction/Collaboration [ICPC-26]
- Code Reuse and Re-implementation [EMSE-19]
- Automatic Code Review [ICPC-23] [PAKDD-19]
- Software Knowledge Interpretation [EMSE-24] [ASE-23] [ESEM-18] [ASE-16]
- Developer-oriented Chatbot [ICSE-23] [ASE-22] [FSE-19] [ASE-17][FSE-17] [EMSE-17] [MSR-16]
- Semantic Code Search [FSE-22]
News
January, 2026: One work titled Backporting in Robot Operating System: Identifying Commit Purpose and Propagation Need with Large Language Models is accepted by the 3rd ACM International Conference on AI Foundation Models and Software Engineering (FORGE 2026). Congrats Pankaj and Kyle!December, 2025: One work titled Understanding Codebase like a Professional! Human-AI Collaboration for Code Comprehension is accepted by the 34st IEEE/ACM International Conference on Program Comprehension (ICPC 2026).December, 2025: One work titled How Do Semantically Equivalent Code Transformations Impact Membership Inference on LLMs for Code? is accepted by Research Track of International Conference on Software Engineering (ICSE 2026). Congrats Hua and Nazmul!September, 2025: One work titled How Quantization Impacts Privacy Risk on LLMs for Code? is accepted by Main Track of 2nd ACM International Conference on AI-powered Software (AIware 2025). Congrats Nazmul and Hua!August, 2025: One work titled Stealthy Backdoor Attack for Code Models was accepted by Transactions on Software Engineering wins the Best Paper Award of the year 2024. Yay! AnnouncementJuly, 2025: One work titled Learning From the Best: What Makes Popular Hugging Face Models? A Registered Report is accepted by Registered Report Track of International Conference on Software Maintenance and Evolution (ICSME). Congrats Yinan!June, 2025: Two works titled Quantization Is Not a Dealbreaker: Empirical Insights from Large Code Models and APIDocBooster: An Extract-Then-Abstract Framework Leveraging Large Language Models for Augmenting API Documentation are accepted by International Conference on Software Maintenance and Evolution (ICSME).May, 2025: I gave a talk titled How Program Analysis and AI Can Better Support Domain-Specific Software Development and Maintenance? at SAS Institute. It's fun!April, 2025: One work titled A Comprehensive Study of OOP-Related Bugs in C++ Compilers is accepted by IEEE Transactions on Software Engineering (TSE).March, 2025: I start to serve as a member of Editorial Board for one of my favorite journals Empirical Software Engineering Journal (aka. EMSE). Please consider submitting your work to EMSE!Jan, 2025: My application on Safe Use of AI for Coding to OpenAI Researcher Access Program is awarded.Jan, 2025: One work titled Finding Safety Violations of AI-Enabled Control Systems through the Lens of Synthesized Proxy Programs is accepted by ACM Transactions on Software Engineering and Methodology (TOSEM).November, 2024: One work titled PTM4Tag+: Tag recommendation of stack overflow posts with pre-trained models is accepted by Empirical Software Engineering (EMSE).October, 2024: One work titled Prioritizing Speech Test Cases is accepted by ACM Transactions on Software Engineering and Methodology (TOSEM).October, 2024: I will serve Forge 2025 as a co-chair of Benchmarking Track. Please consider submitting your work!October, 2024: One work titled An empirical study on the effectiveness of large language models for SATD identification and classification is accepted by Empirical Software Engineering (EMSE).August, 2024: I gave a talk titled Systematic Code Migration at Amazon AWS.August, 2024: One work titled Leveraging Large Language Model for Automatic Patch Correctness Assessment is accepted by Transactions on Software Engineering (TSE).March, 2024: One work titled BAFFLE: Backdoor Attack in Offline Reinforcement Learning is accepted by IEEE Symposium on Security and Privacy (S&P 2024).Feb, 2024: One work titled Stealthy Backdoor Attack for Code Models is accepted by IEEE Transactions on Software Engineering (TSE).December, 2023: Two works titled Out of Sight, Out of Mind: Better Automatic Vulnerability Repair by Broadening Input Ranges and Sources and Curiosity-Driven Testing for Sequential Decision-Making Process are accepted by 46th International Conference on Software Engineering (ICSE 2024).December, 2023: One work titled Greening Large Language Models of Code is accepted by The 46th International Conference on Software Engineering (ICSE 2024) -- Software Engineering in Society Track (ICSE 2024 SEIS).November, 2023: I will serve ICSE 2025 as a Review Process Co-chair of Research Track. Please consider submitting your work!November, 2023: I will serve ASE 2024 as a PC of Research Track. Please consider submitting your work!November, 2023: I will serve Forge 2024 (co-located with ICSE 2024) as a Local Chair. Please consider submitting your work!October, 2023: One work titled Representation Learning for Stack Overflow Posts: How Far are We? is accepted by ACM Transactions on Software Engineering and Methodology (TOSEM).September, 2023: I will serve CAIN 2024 as a PC of Technical Track. Please consider submitting your work!July, 2023: I will serve FSE 2024 as a PC of Technical Track. Please consider submitting your work!July, 2023: One work titled The Devil is in the Tails: An Exploratory Study on Long-tailed (Technical track) is accepted by ASE'23.July, 2023: One work titled Are We Ready to Embrace Generative AI for Software Q&A? (NIER track) is accepted by ASE'23.June, 2023: One work titled Self-Supervised Code Change Representation Learning (Technical track) is accepted by ICSME'23.June, 2023: I contribute to one chapter Supporting Collateral Evolution in Software Ecosystems of the book titled Software Ecosystems: Tooling and Analytics. The book is planned for release end of Augest 2023.May, 2023: One work titled Multi-Granularity Detector for Vulnerability Fixes is accepted by IEEE Transactions on Software Engineering (TSE).April, 2023: I will serve ASE 2023 as a PC of NIER Track. Please consider submitting your work!April, 2023: I will serve APSEC 2023 as a PC of Technical Track. Please consider submitting your work!March, 2023: I will serve PRDC 2023 as a PC of Technical Track. Please consider submitting your work!March, 2023: I will serve ICSE 2024 as a PC of Demonstrations Track. Please consider submitting your work!March, 2023: I delivered a lecture as a guest instructor on code embedding for the graduate students from Dalhousie University.March, 2023: One work titled Generation-based Code Review Automation: How Far Are We? is accepted to appear at the Research track of the 31st IEEE/ACM International Conference on Program Comprehension (ICPC 2023).January, 2023: One work titled TECHSUMBOT: A Stack Overflow Answer Summarization Tool for Technical Query is accepted to appear at the Demostractions track of the 45th IEEE/ACM International Conference on Software Engineering (ICSE 2023).December, 2022: One work titled Curiosity-Driven and Victim-Aware Adversarial Policies is accepted by The Annual Computer Security Applications Conference and won Honorable Mention Award.November, 2022: One work titled Duplicate Bug Report Detection: How Far Are We? is accepted by ACM Transactions on Software Engineering and Methodology.July, 2022: Two works titled Compressing Pre-trained Models of Code into 3 MB and Answer Summarization for Technical Queries: Benchmark and New Approach (Technical track) are accepted by ASE'22.June, 2022: One work titled How to Better Utilize Code Graphs in Semantic Code Search? (Technical track) is accepted by ESEC/FSE'22.June, 2022: [Call For Paper] I am one of the organizers of the ESEC/FSE workshop the 6th edition of the MaLTeSQuE on machine learning techniques for software quality evolution. Please submit your great work(s)!March, 2022: One work titled PTM4Tag: Sharpening Tag Recommendation of Stack Overflow with Pre-trained Models (Technical track) is accepted by ICPC'22.December, 2021: Two works titled Aspect-Based API Review Classification: How Far Can Pre-Trained Transformer Model Go? (Technical track) and Can Identifier Splitting Improve Open-Vocabulary Language Model of Code? (ERA track) are accepted by SANER'22.August, 2021: I graduated from SMU, a beautiful university carries lots of great memories!June, 2021: One work titled Post2Vec: Learning Distributed Representations of Stack Overflow Posts is accepted by IEEE Transactions on Software Engineering (TSE).
Bio
Dr. Bowen Xu is a (tenure-track) Assistant Professor in the Department of Computer Science at North Carolina State University (NC State). He received his PhD degree from School of Computing and Information Systems (SCIS) at Singapore Management University (SMU). His research interests lie primarily in the fields of machine learning and software engineering. Recently, he has focused on securing AI models for various software engineering tasks. His works won several research paper awards, such as 2024 Best Paper Award at IEEE Transactions on Software Engineering (TSE) journal, the Highly Commended Full Paper Award at ESEM 2018, the Honorable Mention Award at ACSAC 2022, nominated for ACM SIGSOFT Distinguished Paper Award at ASE 2022. He co-organized two FSE workshops SEA4DQ in 2024 and MaLTeS in 2022. He served as associate editor for EMSE journal. He is also invited to be a PC and referee for many top-tier conferences and Journals, such as ICSE, FSE, ASE, TSE, TOSEM, EMSE, etc. Contact him at: bxu22@ncsu.edu. More info at: https://www.bowenxu.me.
