PETROV, Nikolai; ANDERSSON, Sofia. Sparse Experts Scale Better in Efficient Mixture Architectures for Trillion Parameter Models. Computer Life, [S. l.], v. 14, n. 2, p. 16–22, 2026. DOI: 10.54097/baczzj49. Disponível em: https://computer-life.org/index.php/ojs/article/view/41. Acesso em: 19 may. 2026.