Semantic Patch Learning for Vulnerability Repair in LLM-Generated Code
Global Media and Social Sciences Research Journal · 15.02.2026 · 7 (1) · 161-167
DOI: 10.71465/gmssrj160 ↗Kurz gefasst
Fachartikel in der Zeitschrift Global Media and Social Sciences Research Journal, online erschienen am 15.02.2026. Verfasst von Mikael Saarinen und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
We introduce a semantic patch learning model that generalizes repair templates from historical CVE patches. Extracted semantic patterns guide automatic correction of insecure LLM outputs. Using 28,000 real vulnerability patches, the model learns transformations for memory bugs, crypto misuse, and authorization bypasses. On a benchmark with 900 insecure LLM-generated samples, the repair tool fixes 68% of vulnerabilities automatically and partially mitigates another 21%. This work demonstrates that semantic patch abstraction is a powerful mechanism for improving LLM code security.
Zitieren
APA
Saarinen, M., Korhonen, E., & Lehtola, J. (2026). Semantic Patch Learning for Vulnerability Repair in LLM-Generated Code. Global Media and Social Sciences Research Journal, 7(1), 161-167. https://doi.org/10.71465/gmssrj160
Vancouver
Saarinen M, Korhonen E, Lehtola J. Semantic Patch Learning for Vulnerability Repair in LLM-Generated Code. Global Media and Social Sciences Research Journal. 2026;7(1):161-167. doi:10.71465/gmssrj160
BibTeX
@article{Saarinen2026,
author = {Saarinen, Mikael and Korhonen, Elina and Lehtola, Jukka},
title = {Semantic Patch Learning for Vulnerability Repair in LLM-Generated Code},
journal = {Global Media and Social Sciences Research Journal},
year = {2026},
volume = {7},
number = {1},
pages = {161--167},
doi = {10.71465/gmssrj160}
}
EUVIMED