LLM-Assisted Programming: Trends, Evaluation Gaps, and Software Assurance

Year : 2026 | Volume : 13 | Issue : 02 | Page : 28 42
By

Jyoti Dabass,

Bhupender Singh Dabass,

  1. Global Professor of Practice, Golden Gate University, California, USA
  2. Researcher and Advocate, District Bar Association, Haryana, India

Abstract

Large language models (LLMs) are increasingly used for code generation, software testing, program repair, and security analysis. However, their evaluation still relies heavily on functional benchmarks that often focus on small, isolated programming tasks. Such benchmarks provide limited insight into security, test quality, repository-level dependencies, and maintainability. This study examines how the evaluation of LLM-assisted programming has developed across these application areas. The literature shows a gradual shift from function-level benchmarks toward repository-level tasks and agent-based development. At the same time, several recurring problems remain difficult to capture with conventional functional metrics. These include incorrect API use, weak test oracles, unreliable fault localization, test suite overfitting, insecure generated code, and difficulties in handling dependencies across files. To address these gaps, a four-stage assurance framework is proposed. It separates evaluation into functional correctness, test and oracle quality, security assessment, and maintainability. The framework is intended to complement existing benchmark measures by examining properties that are important when generated code is used in practical software development. It is a literature-derived framework and has not yet been empirically validated. Future work should evaluate it on real software repositories and establish suitable measures for each stage.

Keywords: Large language models (LLMs), LLM-assisted programming, code generation, automated program repair, unit test generation, software security, code review, evaluation benchmarks, benchmark contamination, software assurance, quantum software engineering, cyber reasoning systems

[This article belongs to Recent Trends in Programming languages ]

How to cite this article: Jyoti Dabass, Bhupender Singh Dabass. LLM-Assisted Programming: Trends, Evaluation Gaps, and Software Assurance. Recent Trends in Programming languages. 2026; 13(02):28-42.
How to cite this URL: Jyoti Dabass, Bhupender Singh Dabass. LLM-Assisted Programming: Trends, Evaluation Gaps, and Software Assurance. Recent Trends in Programming languages. 2026; 13(02):28-42. Available from: https://journals.stmjournals.com/rtpl/article=2026/view=259982

References

  1. Jiang J, Wang F, Shen J, Kim S, Kim S. A survey on large language models for code generation. ACM Trans Softw Eng Methodol. 2026;35(2):1-72. doi:10.1145/3747588.
  2. Zhang Q, Fang C, Xie Y, Zhang Y, Yu S, Sun W, et al. A survey on large language models for software engineering. Sci China Inf Sci. 2026;69(4):141102. doi:10.1007/s11432-025-4670-0.
  3. Bhati H. Agentic AI in the software development lifecycle: architecture, empirical evidence, and the reshaping of software engineering [Preprint]. 2026. arXiv:2604.26275.
  4. Gulwani S. Dimensions in program synthesis. In: Proceedings of the 12th International ACM SIGPLAN Symposium on Principles and Practice of Declarative Programming. Hagenberg, Austria: Association for Computing Machinery; 2010. p. 13-24. doi:10.1145/1836089.1836091.
  5. Gülmez B. Code generation with large language models: a survey from neural program synthesis to autonomous software development. Appl Intell. 2026;56(6):200. doi:10.1007/s10489-026-07230-0.
  6. Feng Z, Guo D, Tang D, Duan N, Feng X, Gong M, et al. CodeBERT: a pre-trained model for programming and natural languages. In: Findings of the Association for Computational Linguistics: EMNLP 2020. Online: Association for Computational Linguistics; 2020. p. 1536-1547. doi:10.18653/v1/2020.findings-emnlp.139.
  7. Wang Y, Wang W, Joty S, Hoi SCH. CodeT5: identifier-aware unified pre-trained encoder-decoder models for code understanding and generation. In: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Online and Punta Cana, Dominican Republic: Association for Computational Linguistics; 2021. p. 8696-8708. doi:10.18653/v1/2021.emnlp-main.685.
  8. Chen M, Tworek J, Jun H, Yuan Q, Pinto HP, Kaplan J, et al. Evaluating large language models trained on code [Preprint]. 2021. arXiv:2107.03374.
  9. Roziere B, Gehring J, Gloeckle F, Sootla S, Gat I, Tan XE, et al. Code Llama: open foundation models for code [Preprint]. 2023. arXiv:2308.12950.
  10. Guo D, Zhu Q, Yang D, Xie Z, Dong K, Zhang W, et al. DeepSeek-Coder: when the large language model meets programming—the rise of code intelligence [Preprint]. 2024. arXiv:2401.14196.
  11. Zhu Q, Guo D, Shao Z, Yang D, Wang P, Xu R, et al. DeepSeek-Coder-V2: breaking the barrier of closed-source models in code intelligence [Preprint]. 2024. arXiv:2406.11931.
  12. Jain N, Han K, Gu A, Li WD, Yan F, Zhang T, et al. LiveCodeBench: holistic and contamination free evaluation of large language models for code. In: International Conference on Learning Representations. Singapore: International Conference on Learning Representations; 2025. p. 25479-25491.
  13. Kaniewski S, Schmidt F, Enzweiler M, Menth M, Heer T. A systematic literature review on detecting software vulnerabilities with large language models. ACM Trans Softw Eng Methodol. 2026. doi:10.1145/3815425.
  14. Jimenez CE, Yang J, Wettig A, Yao S, Pei K, Press O, et al. SWE-bench: can language models resolve real-world GitHub issues? In: The Twelfth International Conference on Learning Representations. Vienna, Austria: International Conference on Learning Representations; 2024.
  15. Pearce H, Ahmad B, Tan B, Dolan-Gavitt B, Karri R. Asleep at the keyboard? Assessing the security of GitHub Copilot’s code contributions. Commun ACM. 2025;68(2):96-105. doi:10.1145/3610721.
  16. Dou S, Jia H, Wu S, Zheng H, Wu M, Tao Y, et al. What is wrong with your code generated by large language models? An extensive study. Sci China Inf Sci. 2026;69(1):112107. doi:10.1007/s11432-025-4632-8.
  17. Petke J, Martinez M, Kechagia M, Aleti A, Sarro F. The patch overfitting problem in automated program repair: practical magnitude and a baseline for realistic benchmarking. In: Companion Proceedings of the 32nd ACM International Conference on the Foundations of Software Engineering. Porto de Galinhas, Brazil: Association for Computing Machinery; 2024. p. 452-456. doi:10.1145/3663529.3663776.
  18. Hou X, Zhao Y, Liu Y, Yang Z, Wang K, Li L, et al. Large language models for software engineering: a systematic literature review. ACM Trans Softw Eng Methodol. 2024;33(8):1-79. doi:10.1145/3695988.
  19. Campos V, Shariffdeen R, Ulges A, Noller Y. Exploring generalizable automated program repair with large language models. IEEE Access. 2026;14:110543-110558. doi:10.1109/ACCESS.2026.3714292.
  20. Wei J, Wang X, Schuurmans D, Bosma M, Ichter B, Xia F, et al. Chain-of-thought prompting elicits reasoning in large language models. In: Advances in Neural Information Processing Systems 35. New Orleans, LA, USA: Neural Information Processing Systems Foundation, Inc.; 2022. p. 24824-24837. doi:10.52202/068431-1800.
  21. Li J, Rabbi F, Cheng C, Sangalay A, Tian Y, Yang J. An exploratory study on fine-tuning large language models for secure code generation. Empir Softw Eng. 2026;31(4):81. doi:10.1007/s10664-026-10803-9.
  22. Hu EJ, Shen Y, Wallis P, Allen-Zhu Z, Li Y, Wang S, et al. LoRA: low-rank adaptation of large language models [Preprint]. 2021. arXiv:2106.09685.
  23. He J, Vechev M. Large language models for code: security hardening and adversarial testing. In: Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security. Copenhagen, Denmark: Association for Computing Machinery; 2023. p. 1865-1879. doi:10.1145/3576915.3623175.
  24. Gao Y. Retrieval-augmented generation for large language models: a survey [Preprint]. 2023. arXiv:2312.10997.
  25. Tao Y, Li Y, Qin Y, Liu Y. Retrieval-augmented code generation: a survey with focus on repository-level approaches [Preprint]. 2025. arXiv:2510.04905.
  26. Zhang F, Chen B, Zhang Y, Keung J, Liu J, Zan D, et al. RepoCoder: repository-level code completion through iterative retrieval and generation. In: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. Singapore: Association for Computational Linguistics; 2023. p. 2471-2484. doi:10.18653/v1/2023.emnlp-main.151.
  27. Zhang Y, Ruan H, Fan Z, Roychoudhury A. AutoCodeRover: autonomous program improvement. In: Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis. Vienna, Austria: Association for Computing Machinery; 2024. p. 1592-1604. doi:10.1145/3650212.3680384.
  28. Ruan H, Zhang Y, Roychoudhury A. SpecRover: code intent extraction via LLMs. In: 2025 IEEE/ACM 47th International Conference on Software Engineering (ICSE). Ottawa, ON, Canada: IEEE; 2025. p. 963-974. doi:10.1109/ICSE55347.2025.00080.
  29. Austin J, Odena A, Nye M, Bosma M, Michalewski H, Dohan D, et al. Program synthesis with large language models [Preprint]. 2021. arXiv:2108.07732.
  30. Luo Z, Xu C, Zhao P, Sun Q, Geng X, Hu W, et al. WizardCoder: empowering code large language models with Evol-Instruct. In: Proceedings of the Twelfth International Conference on Learning Representations, ICLR 2024. Vienna, Austria: International Conference on Learning Representations; 2024. p. 1-21.
  31. Gunasekar S, Zhang Y, Aneja J, Mendes CC, Del Giorno A, Gopi S, et al. Textbooks are all you need [Preprint]. 2023. arXiv:2306.11644.
  32. Huang D, Zhang JM, Luck M, Bu Q, Qing Y, Cui H. AgentCoder: multi-agent-based code generation with iterative testing and optimisation [Preprint]. 2023. arXiv:2312.13010.
  33. Zhuo TY, Vu MC, Chim J, Hu H, Yu W, Widyasari R, et al. BigCodeBench: benchmarking code generation with diverse function calls and complex instructions. In: International Conference on Learning Representations. 2025.
  34. Guo X, Zhao J, Zhao P. On repairing quantum programs using ChatGPT. In: Proceedings of the 5th ACM/IEEE International Workshop on Quantum Software Engineering, Q-SE 2024. Lisbon, Portugal: Association for Computing Machinery; 2024. p. 9-16. doi:10.1145/3643667.3648223.
  35. Jin H, Chen H. Are LLMs reliable code reviewers? Systematic overcorrection in requirement conformance judgement. Autom Softw Eng. 2026;33:90. doi:10.1007/s10515-026-00638-5.
  36. Wolff D, Mirchev M, Roychoudhury A. Large language models in software security analysis. Commun ACM. 2026;69:60-67. doi:10.1145/3770922.
  37. Du X, Liu M, Wang K, Wang H, Liu J, Chen Y, et al. Evaluating large language models in class-level code generation. In: Proceedings of the 46th IEEE/ACM International Conference on Software Engineering (ICSE 2024). Lisbon, Portugal: Association for Computing Machinery; 2024. p. 982-994. doi:10.1145/3597503.3639219.
  38. Zhang J, Hu X, Gao C, Xia X, Li S. Enhancing automated unit test generation with large language models: a systematic literature review. ACM Trans Softw Eng Methodol. 2026;35:1-67. doi:10.1145/3802827.
  39. Celik A, Mahmoud QH. A review of large language models for automated test case generation. Mach Learn Knowl Extr. 2025;7:97. doi:10.3390/make7030097.
  40. Tasarsu M, Tokmak AV, Catal C. Test case generation using large language models: a systematic literature review. Cluster Comput. 2026;29:227. doi:10.1007/s10586-026-06021-z.
  41. Lemieux C, Inala JP, Lahiri SK, Sen S. CodaMosa: escaping coverage plateaus in test generation with pre-trained large language models. In: 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE). Melbourne, Australia: IEEE; 2023. p. 919-931. doi:10.1109/ICSE48619.2023.00085.
  42. Chu B, Feng Y, Liu K, Guo Z, Zhang Y, Shi H, et al. Large language models for unit test generation: achievements, challenges, and opportunities [Preprint]. 2025. arXiv:2511.21382.
  43. Xia CS, Zhang L. Less training, more repairing please: revisiting automated program repair via zero-shot learning. In: Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering. Singapore: Association for Computing Machinery; 2022. p. 959-971. doi:10.1145/3540250.3549101.
  44. Xia CS, Wei Y, Zhang L. Automated program repair in the era of large pre-trained language models. In: 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE). Melbourne, Australia: IEEE Computer Society; 2023. p. 1482-1494. doi:10.1109/ICSE48619.2023.00129.
  45. Hossain SB, Jiang N, Zhou Q, Li X, Chiang WH, Lyu Y, et al. A deep dive into large language models for automated bug localization and repair. Proc ACM Softw Eng. 2024;1:1471-1493. doi:10.1145/3660773.
  46. Wu F, Liu X, Deceptprompt XC. Exploiting LLM-driven code generation via adversarial natural language instructions [Preprint]. 2023. arXiv:2312.04730.
  47. Schuster R, Song C, Tromer E, Shmatikov V. You autocomplete me: poisoning vulnerabilities in neural code completion. In: 30th USENIX Security Symposium (USENIX Security 21). Berkeley, CA, USA: USENIX Association; 2021. p. 1559-1575.
  48. Croft R, Babar MA, Kholoosi MM. Data quality for software vulnerability datasets. In: 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE). Melbourne, Australia: IEEE; 2023. p. 121-133. doi:10.1109/ICSE48619.2023.00022.
  49. Steenhoek B, Rahman MM, Roy MK, Alam MS, Tong H, Das S, et al. To err is machine: vulnerability detection challenges LLM reasoning [Preprint]. 2024. arXiv:2403.17218.
  50. Li Z, Lu S, Guo D, Duan N, Jannu S, Jenks G, et al. Automating code review activities by large-scale pre-training. In: Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering. New York, NY, USA: Association for Computing Machinery; 2022. p. 1035-1047. doi:10.1145/3540250.3549081.
  51. Liu F, Liu S, Zhu Y, Lian X, Zhang L. SecureReviewer: enhancing large language models for secure code review through secure-aware fine-tuning. In: Proceedings of the 2026 IEEE/ACM 48th International Conference on Software Engineering. 2026. p. 310-322. doi:10.1145/3744916.3773191.
  52. Dobslaw F, Feldt R, Yoon J, Yoo S. Challenges in testing large language model-based software: a faceted taxonomy. ACM Trans Softw Eng Methodol. 2026;35:1-24. doi:10.1145/3806396.
  53. Ye J, Hu Q, Zhang F, Xia S, Guo X, Wu X, et al. Large language models for quantum software engineering: opportunities and challenges in modeling, generation, testing, and repair. ACM Trans Softw Eng Methodol. 2026;35:1-42. doi:10.1145/3828167.
Regular Issue Subscription Original Research
Volume 13
Issue 02
Received 24/06/2026
Accepted 25/08/2026
Published 28/09/2026
Publication Time 96 Days

Login

My IP

PlumX Metrics

Support