Re-evaluating the Explainability-Performance Trade-Off Paradigm in Natural Language Processing Models: A Quantitative Meta-Analysis of Transformer Architectures (2019-2023). Global Journal of Computer Science and Technology, [S. l.], v. 26, n. D1, p. 1–14, 2026. DOI: 10.34257/GJCSTD256108. Disponível em: https://computerresearch.org/index.php/computer/article/view/102490. Acesso em: 8 jul. 2026.