Giovani Mambrim Leme, Matheus Ferracciú Scatolin, Anderson Luis Bento Soares, Leonardo Nascimento Ferreira, Ruben Interian, and Zanoni Dias. 2026.
 How Well Do Small Language Models Understand the Brazilian Educational Context? An Evaluation on ENEM, FUVEST and Unicamp. In <em>Anais do XXIII Encontro Nacional de Inteligência Artificial e Computacional</em>, outubro 19, 2026, Cuiabá/MT, Brasil. SBC, Porto Alegre, Brasil, 3274-3285. DOI: <a href="https://doi.org/10.5753/eniac.2026.31313">https://doi.org/10.5753/eniac.2026.31313</a>.