(2025). Correction: GATmath and GATLc: Comprehensive benchmarks for evaluating Arabic large language models. PLoS ONE. https://doi.org/10.1371/journal.pone.0338465