arXivDaryna Dementieva, Nikolay Babakov, Kathy Hämmerl, Ilseyar Alimova, Jindřich Libovický, Shu Okabe, Miras Baisbay, Lukas Edman, Abrorkhon Inomkhujaev, Antonia Karamolegkou, Mateusz Lango, Volkan Özer, Nikola Selic, Subhankar Swain, Tsedeniya Kinfe Temesgen, Galit Bary Weisberg, Alexander FraserTue, Jul 7, 2026, 1:25 AM PDT
score 16.8
New math benchmark tests AI on 18 underrepresented languages
Original: PluraMath: Extending Mathematical Reasoning Evaluation Beyond High-Resource Languages
Source: arxiv.org ↗
Writing ELI5 summary…