Uncertainty-Aware Training and Computational Scaling for Hybrid Neural Pipeline Models: Heteroscedastic Loss Functions, Mixed Precision, and Multi-GPU Optimisation. Federal University Gusau Faculty of Education Journal, [S. l.], v. 3, n. 2, p. 272–286, 2026. DOI: 10.64348/zije.2026442. Disponível em: https://js.foefugusau.com.ng/index.php/foefujs/article/view/442. Acesso em: 28 oct. 2026.