LATIDIA · Ciberseguridad

Detección de envenenamiento de datos en LLM de generación de código a través de Black-Box, escaneo orientado a vulnerabilidades

arXiv: 2603.17174v2Announce Type: replace Resumen: Los modelos de lenguaje grande (LLM) de generación de código se integran cada vez más en los flujos de trabajo modernos de desarrollo de software. Trabajos recientes han demostrado que estos modelos son vulnera

La noticia

arXiv:2603.17174v2 Announce Type: replace Abstract: Code generation large language models (LLMs) are increasingly integrated into modern software development workflows. Recent work has shown that these models are vulnerable to backdoor and poisoning attacks that induce the generation of insecure code, yet effective defenses remain limited. Existing scanning approaches rely on token-level generation consistency to invert attack targets, which is ineffective for source code where identical semantics can appear in diverse syntactic forms. We present CodeScan, the first black-box, vulnerability-specific scanning framework for auditing code generation LLMs, assuming that the defender specifies the target vulnerability classes and provides corresponding task-relevant prompts. CodeScan identifies attack targets by analyzing structural similarities across multiple