AI-to-AI o'zaro ta'sir paradigmasidagi kaskadli CI/CD nosozliklarini oldini oluvchi infratuzilma AI agentlari uchun topologik AST sensori
Qabul qilingan: 2026-05-15
Nashr etilgan: 2026-06-06
Annotatsiya
Ушбу маколада, AI билан AI самарали о'заро тасири учун инфратузилма агентлари макссус "сезги" катламини талаб килади. Biz mavhum sintaksis daraxtlari (AST) asosidagi topologik filtrlash moduliga ega autonom tutqichning arxitekturasini taklif qilamiz, uni to'g'ridan-to'g'ri CI/CD runner qobig'iga integratsiya qilish mumkin. Bizning vositamiz joylashtirish quvur liniyasi Agentlarini almashtirmaydi, balki ular uchun mikrojarrohlik optikasi vazifasini bajaradi. Алгоритм qat'iy ravishda semantik G_{sub} kichik grafikni ajratib oladi va kontekstni O\left(N\right) dan O\left(K\right) tokengacha siqadi.
Kalit so‘zlar
Adabiyotlar ro'yxati
-
Liu, K., et al. (2019). TBar: Revisiting Template-based Automated Program Repair. International Symposium on Software Testing and Analysis (ISSTA). https://dl.acm.org/doi/10.1145/3293882.3330577
-
Long, F., & Rinard, M. (2016). Automatic synthesis of code forces for software repair. Principles of Programming Languages (POPL). https://people.csail.mit.edu/rinard/paper/popl16.pdf
-
Just, R., et al. (2014). Defects4J: A database of existing faults to enable controlled studies for Java programs. ISSTA. https://dl.acm.org/doi/10.1145/2610384.2628055
-
Gazzola, L., et al. (2019). Automatic Software Repair: A Survey. IEEE Transactions on Software Engineering. https://www.computer.org/csdl/journal/ts/2019/01/08089448/17D45WaTknO
-
Madaan, A., et al. (2023). Self-Refine: Iterative Refinement with Self-Feedback. arXiv preprint. https://arxiv.org/abs/2303.17651
-
Ajay Varma, (2025). Infrastructure as code: A paradigm shifts in cloud resource management and deployment automation. https://doi.org/10.30574/wjaets.2025.15.1.0478
-
Renicius Pagotto. (2025). The Importance of CI/CD in Modern Software Development. https://dev.to/reniciuspagotto/the-importance-of-cicd-in-modern-software-development-3d9k
-
Yang, J., et al. (2024). SWE-agent: Agent-Computer Interfaces Enable Software Engineering Agents. Research Report. https://arxiv.org/abs/2405.15793
-
Jimenez, C. E., et al. (2024). SWE-bench: Can Language Models Resolve Real-World GitHub Issues? ICLR. https://arxiv.org/abs/2310.06770
-
Allamanis, M., et al. (2018). Learning to Represent Programs with Graphs. ICLR. https://arxiv.org/abs/1711.00740
Mualliflar haqida
Litsenziya
This work is licensed under a Creative Commons Attribution 4.0 International License.