2025. Enhancing Multimodal Models with Vision-Centric Feedback Loops. International Journal of Computational Health & Machine Learning. 3, 1 (Sep. 2025).