Published in FSE 2026 (Industry Paper), 2026
An industry paper on using AutoCodeRover, an agentic program repair system, to automatically resolve SonarQube-flagged issues.
Recommended citation: M. Mirchev, R. Shariffdeen, H. Ruan, Y. Zhang, A. Roychoudhury. (2026). "AutoCodeRover: Agentic Program Repair for SonarQube Issues." Industry Track, Symposium on the Foundations of Software Engineering (FSE).
Published in Communications of the ACM, 69(6), 2026
A survey/perspective on the role of large language models in software security analysis.
Recommended citation: D. Wolff, M. Mirchev, A. Roychoudhury. (2026). "Large Language Models in Software Security Analysis." Communications of the ACM, 69(6).
Published in ICSE 2026 (48th International Conference on Software Engineering), 2026
Evaluates how agentic AI systems can automatically fix security vulnerabilities discovered by OSS-Fuzz.
Recommended citation: Y. Zhang, J. Wang, D. Berzin, M. Mirchev, A. Roychoudhury. (2026). "Fixing Security Vulnerabilities with Agentic AI in OSS-Fuzz." Proceedings of the 48th IEEE/ACM International Conference on Software Engineering (ICSE).
Published in IEEE Transactions on Software Engineering, 2025
Proposes using Computation Tree Logic (CTL) specifications to guide automated program repair.
Recommended citation: Y. Liu, Y. Song, M. Mirchev, A. Roychoudhury. (2025). "Computation Tree Logic Guided Program Repair." IEEE Transactions on Software Engineering.
Published in ICSE 2025 (47th International Conference on Software Engineering), 2025
Presents a program-analysis-based approach for detecting malicious Python packages in the software supply chain, including faults that could poison LLM training data.
Recommended citation: R. Shariffdeen, B. Hassanshahi, M. Mirchev, A. El Husseini, et al. (2025). "Detecting Python Malware in the Software Supply Chain with Program Analysis." Proceedings of the 47th IEEE/ACM International Conference on Software Engineering (ICSE).
Published in arXiv preprint, 2024
Proposes an approach for combining large language model code generation with formal assurance, so generated programs come with guarantees.
Recommended citation: M. Mirchev, A. Costea, A. K. Singh, A. Roychoudhury. (2024). "Assured Automatic Programming via Large Language Models." arXiv preprint.
Published in 5th ACM/IEEE International Workshop on Automated Program Repair (APR), 2024
A report on the 2024 Automated Program Repair (APR) competition.
Recommended citation: R. Shariffdeen, Y. Noller, M. Mirchev, H. Ruan, G. Xiang, A. Costa, G. J. Duck, et al. (2024). "APR Competition 2024." Proceedings of the 5th ACM/IEEE International Workshop on Automated Program Repair.
Published in NDSS 2024, 2024
Uses large language models to guide protocol fuzzing, improving the discovery of bugs in protocol implementations.
Recommended citation: R. Meng, M. Mirchev, M. Böhme, A. Roychoudhury. (2024). "Large Language Model guided Protocol Fuzzing." Network and Distributed System Security Symposium (NDSS).
Published in 2023 IEEE/ACM International Workshop on Automated Program Repair (APR), 2023
A report on the 2023 Automated Program Repair (APR) competition.
Recommended citation: R. Shariffdeen, M. Mirchev, A. Roychoudhury. (2023). "Program Repair Competition." 2023 IEEE/ACM International Workshop on Automated Program Repair (APR).
Published in ICSE 2023 (Tool Demonstrations Track), 2023
A tool-track paper presenting Cerberus, a framework for building and evaluating automated program repair tools.
Recommended citation: R. Shariffdeen, M. Mirchev, Y. Noller, A. Roychoudhury. (2023). "Cerberus: A Program Repair Framework." Proceedings of the 45th International Conference on Software Engineering (ICSE), Tool Demonstrations Track.
Published in ICSE 2023 (Research Track), 2023
Studies how large language model output can be automatically repaired, combining LLM-based code generation with automated program repair.
Recommended citation: Z. Fan, X. Gao, M. Mirchev, A. Roychoudhury, S. H. Tan. (2023). "Automated Repair of Programs from Large Language Models." Proceedings of the 45th International Conference on Software Engineering (ICSE), Research Track.