Automated repairing of software bugs with LLM-agent(s)

Organisatsiooni nimi
Software Engineering Analytics
Kokkuvõte
Recent advancements in Large Language Models (LLMs) have greatly enhanced the automation of software development tasks, including code synthesis, test generation, and program repair [1]. This thesis aims to evaluate an agent-based workflow [2] built on the LLM-modulo [3] architecture, which utilizes LLM-driven agents to understand code, identify faults, and apply patches through continuous collaboration. These agents will rely on external tools as verifiers to ensure both the syntactic and functional correctness of the generated patches. For further details, please contact.

[1] Jin, M., Shahriar, S., Tufano, M., Shi, X., Lu, S., Sundaresan, N., & Svyatkovskiy, A. (2023, November). Inferfix: End-to-end program repair with llms. In Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (pp. 1646-1656).

[2] Xia, C. S., Deng, Y., Dunn, S., & Zhang, L. (2024). Agentless: Demystifying llm-based software engineering agents. arXiv preprint arXiv:2407.01489.

[3] Kambhampati, S., Valmeekam, K., Guan, L., Stechly, K., Verma, M., Bhambri, S., ... & Murthy, A. (2024). LLMs Can't Plan, But Can Help Planning in LLM-Modulo Frameworks. arXiv preprint arXiv:2402.01817.
Lõputöö kaitsmise aasta
2024-2025
Juhendaja
Faiz Ali Shah
Suhtlemiskeel(ed)
inglise keel
Nõuded kandideerijale
Tase
Bakalaureus, Magister
Märksõnad
#SEA, #PROGRAM_REPAIR #LLM_AGENTS

Kandideerimise kontakt

 
Nimi
Faiz Ali Shah
Tel
E-mail
faiz.ali.shah@ut.ee