PRICOPE T-V. HardML: A Benchmark for Evaluating Data Science and Machine Learning Knowledge and Reasoning in AI. Studia UBB Informatica [Internet]. 2025 Mar. 16 [cited 2025 Apr. 2];69(2):59-76. Available from: https://studia.reviste.ubbcluj.ro/index.php/subbinformatica/article/view/9123