A Robust Two-Stage Retrieval-Augmented Vision-Language Framework for Knowledge-Intensive Multimodal Reasoning and Alignment. Computational Discovery and Intelligent Systems (CDIS), [S. l.], v. 2, n. 2, p. 42–52, 2026. Disponível em: https://pub.scientificirg.com/index.php/CDIS/article/view/40. Acesso em: 22 feb. 2026.