LATIDIA · Ciberseguridad

Las palabras hablan más alto que el código: investigación de la heurística cognitiva en la detección de vulnerabilidades de código basada en LLM

arXiv: 2606.30587v2Tipo de anuncio: reemplazar Resumen: Los investigadores y los profesionales aplican cada vez más los Modelos de Lenguaje Grande (LLM) para la detección automatizada de vulnerabilidades. Trabajos recientes han demostrado que los LLM son susceptibles a

La noticia

arXiv:2606.30587v2 Announce Type: replace Abstract: Researchers and practitioners increasingly apply Large Language Models (LLMs) for automated vulnerability detection. Recent work has shown that LLMs are susceptible to the same cognitive heuristics that bias human judgment. Yet, no work has investigated whether these heuristics affect a model's assessment of code vulnerabilities. In this paper, we present the first systematic exploration of cognitive heuristics in LLM-driven code vulnerability detection. We introduce a controlled framework that holds the code fixed and only varies the surrounding context to trigger three cognitive heuristics: the halo effect through author attribution, the framing effect through task objectives and consequences, and the anchoring effect through prior analysis results.