JAKUB L. RAO. PathGuard-Med: Interpretable Safety Alignment for Clinical Large Language Models via Multi-Hop Reasoning Intervention. Computational Intelligence Systems, [S. l.], v. 4, n. 1, 2026. Disponível em: https://scivexus.org/index.php/CIS/article/view/313. Acesso em: 27 may. 2026.