Newest papers

CAT-RAG: Improving Secure Code Generation with Category-Aware Agentic Retrieval

Mahdieh Abazar, Lorenzo De Carli, Benjamin Tan · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.24907 · cited by 0 works (OpenAlex) · open-access PDF

Arabic

وعلى الرغم من التقدم السريع في إنتاج الكود على أساس LLM، لا تزال مشاكل الأمان متاحة على نطاق واسع، ويمكن إنتاج الكود المرتفعة عن بعد (RACG) أن يزيد من المخاطر عندما يكون المعرفة المرتفعة مجهزًا أو ضعيفًا على التسمم. تتمكن التقنيات الأمنية الحالية من إلقاء المعرفة الأمنية على مستوى منخفض من المخاطر، مما يجعلك تشعر بالقلق عندما تتطلب طلبًا مزيدًا من المشاكل ذات الصلة. ونحن نقترح CAT-RAG، وهو نموذجًا أمنيًا منخفضًا عن مستوى المخاطر. ونحن نقترح CAT-RAG إجراء اختبارًا متكاملًا من القطاعات الأمنية عبر

Machine translation (Meta M2M100); the English original follows.

Abstract

Despite rapid progress in LLM-based code generation, security vulnerabilities remain widespread, and retrieval-augmented code generation (RACG) can further amplify risk when retrieved knowledge is verbose or vulnerable to poisoning. Existing secure RACG approaches inject security knowledge at a fine-grained vulnerability level, causing prompt overhead and diluted model attention when a query implicates multiple related weaknesses. We propose CAT-RAG, an agentic security category-aware retrieval-augmented code generation framework. CAT-RAG performs adaptive category selection via an LLM as a categorizer agent and operates either as a standalone security layer or as a drop-in add-on to RACG pipelines, requiring no modification to the underlying model. We replace CWE-level retrieval with compact, language-specific security categories. These categories provide a higher-level abstraction over individual CWE weaknesses, organizing them by the security concern they share, such as Input Validation and Injection Attacks, rather than merely grouping related entries. Each category consolidates best practices and secure code examples into a single, prompt-efficient representation. We perform a comprehensive evaluation across CyberSecEval, CodeGuard+, and HumanEval. On CyberSecEval, CAT-RAG nearly doubles the secure code generation improvement over the state of the art (23.57% vs. 12.11%) while reducing token overhead by 41.71%.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Code (set theory), Vulnerability (computing), Abstraction, Overhead (engineering), Computer security

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange