Marco Mellia

The paper proposes a multi-modal contrastive learning framework to improve the generalization of machine learning models in cybersecurity by transferring knowledge from rich textual vulnerability descriptions to data-scarce payload classification.

The paper re-evaluates LLM agents on CTFs, finding that while general-purpose agents like claude-code are strong baselines, specialized, modular architectures significantly improve performance and consistency by systematically orchestrating specialized roles.