Study of the robustness of temporal factual knowledge in large language models through the evaluation of their capacity to distinguish between correct and incorrect temporal contexts.