PRICOPE, Tidor-Vlad. HardML: A Benchmark for Evaluating Data Science and Machine Learning Knowledge and Reasoning in AI. Studia Universitatis Babeș-Bolyai Informatica, [S. l.], v. 69, n. 2, p. 59–76, 2025. DOI: 10.24193/subbi.2024.2.04. Disponível em: https://studia.reviste.ubbcluj.ro/index.php/subbinformatica/article/view/9123. Acesso em: 2 apr. 2025.