Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools

Гаипназаров , Рустам, Кузнецова , Виктория, Рахимов , Гиёс

Al-Farg'oniy avlodlari · 2026-yil

Annotatsiya

LLMs accelerate software development but often generate insecure code. This study proposes the Security Information Budget (SIB), a real-time metric measuring mutual information between model hidden representations and security context. Research suggests that a decrease in SIB precedes the generation of vulnerabilities. Tested on DeepSeek-Coder-33B and Qwen2.5-Coder-32B using CWE-Bench and HumanEval-Security, SIB monitoring achieved an F1 score of 0.81. It detected risks an average of 23 tokens before a vulnerability appeared, effectively complementing existing static analysis tools.

Maqola ma’lumotlari
MualliflarГаипназаров , Рустам, Кузнецова , Виктория, Рахимов , Гиёс
JurnalAl-Farg'oniy avlodlari
NashriyotТАТУ Фарғона филиали
ISSN2181-4252
Nashr sanasi2026-04-28
Son2
Betlar37-45
TilIngliz
Ilmiy sohaTexnika
Indeksga qo‘shilgan2026-08-26

Maqolaning asl sahifasi

Kalit so‘zlar

large language models, code security, mutual information, information bottleneck, vulnerability detection, CWE., большие языковые модели, безопасность кода, взаимная информация, информационное узкое место, обнаружение уязвимостей, CWE., katta til modellari, kod xavfsizligi, o‘zaro axborot, axborot torayishi (information bottleneck), zaifliklarni aniqlash, CWE.

Ilmiy soha

Texnika

Iqtibos keltirish

APAГаипназаров , Рустам, Кузнецова , Виктория, & Рахимов , Гиёс (2026). Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools. Al-Farg'oniy avlodlari, (2), 37-45. https://al-fargoniy.uz/index.php/journal/article/view/1124
GOST (OAK)Гаипназаров , Рустам, Кузнецова , Виктория, Рахимов , Гиёс Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools // Al-Farg'oniy avlodlari. – 2026. – № 2. – P. 37-45.
MLAГаипназаров , Рустам, Кузнецова , Виктория, and Рахимов , Гиёс. “Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools.” Al-Farg'oniy avlodlari, (2), 37-45, 2026. https://al-fargoniy.uz/index.php/journal/article/view/1124
ChicagoГаипназаров , Рустам, Кузнецова , Виктория, and Рахимов , Гиёс. “Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools.” Al-Farg'oniy avlodlari, (2), 37-45 (2026). https://al-fargoniy.uz/index.php/journal/article/view/1124
HarvardГаипназаров , Рустам, Кузнецова , Виктория, and Рахимов , Гиёс (2026) ‘Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools’, Al-Farg'oniy avlodlari, (2), 37-45. https://al-fargoniy.uz/index.php/journal/article/view/1124
BibTeX
@article{article2026,
  title = {Information-theoretical security budget for code generation using language models: identifying hidden vulnerability drift using mutual information monitoring tools},
  author = {Гаипназаров , Рустам and Кузнецова , Виктория and Рахимов , Гиёс},
  journal = {Al-Farg'oniy avlodlari},
  year = {2026},
  number = {2},
  pages = {37-45},
  url = {https://al-fargoniy.uz/index.php/journal/article/view/1124}
}

Boshqa maqola uchun iqtibos tayyorlash

Al-Farg'oniy avlodlari jurnalidan boshqa maqolalar

Al-Farg'oniy avlodlari — barcha maqolalar · 2026-yil arxivi

Гаипназаров , Рустам — boshqa maqolalari

Texnika sohasidagi boshqa maqolalar

Texnika sohasidagi OAK jurnallari