DOUGLAS ASHCROFT; WILLIAM PRESCOTT. Synergizing Symbolic Logic and Reinforcement Learning for Provably Correct Reasoning in Large Language Model Decision Pipelines. Computational Intelligence Systems, [S. l.], v. 4, n. 1, 2026. DOI: 10.66280/cis.v1i1.199. Disponível em: https://scivexus.org/index.php/CIS/article/view/199. Acesso em: 27 may. 2026.