Chain-of-Verification for Secure Code Generation Using Iterative Self-Review
American Journal Of Big Data · 2026-01-31 · 7 (1) · 142-150
DOI: 10.71465/ajbd3634 ↗At a glance
Journal article in American Journal Of Big Data, published online on 2026-01-31. Written by Laura Fernández and 2 co-authors. 24 references. The full text is freely available (open access, CC BY 4.0).
Abstract (original language)
We develop an iterative self-verification pipeline in which LLMs generate code, analyze potential vulnerabilities, revise the code, and re-verify in cycles. This “chain-of-verification” approach improves vulnerability detection by 51% and secure-code accuracy by 38% over single-pass generation. Tested on 1,200 prompts involving web, crypto, and system programming tasks, the method excels in catching missing checks and unsafe fallbacks. This demonstrates that structured self-review can transform LLMs into more reliable secure-code generators.
Cite
APA
Fernández, L., Morales, J., & Castillo, M. (2026). Chain-of-Verification for Secure Code Generation Using Iterative Self-Review. American Journal Of Big Data, 7(1), 142-150. https://doi.org/10.71465/ajbd3634
Vancouver
Fernández L, Morales J, Castillo M. Chain-of-Verification for Secure Code Generation Using Iterative Self-Review. American Journal Of Big Data. 2026;7(1):142-150. doi:10.71465/ajbd3634
BibTeX
@article{Fernndez2026,
author = {Fernández, Laura and Morales, Javier and Castillo, María},
title = {Chain-of-Verification for Secure Code Generation Using Iterative Self-Review},
journal = {American Journal Of Big Data},
year = {2026},
volume = {7},
number = {1},
pages = {142--150},
doi = {10.71465/ajbd3634}
}
EUVIMED