Jordi CASTELLÀ ROCA, Ignacio MIGUEL RODRÍGUEZ

-

Detection of System Compromise through Remote Attestation Assisted by Large Language Models

This project explores a novel approach to detecting system compromise through remote attestation assisted by large language models (LLMs). A low-level component (e.g., eBPF) collects system information, which is analyzed by an LLM via interactive queries. The goal is to identify inconsistencies or anomalies indicative of compromise. The work includes the design, implementation, and evaluation of a prototype in controlled attack scenarios.

01-ABR-2026

Cybersecurity, Remote Attestation, System Compromise Detection, Anomaly Detection, Kernel, Large Language Models

Seguretat en Xarxes, Estructura de Computadors, Estructura de Sistemes Operatius, Fonaments de Sistemes Operatius, Gestió de Sistemes i Xarxes, Sistemes Distribuïts, Intel·ligència Artificial