The field of artificial intelligence is moving towards a more nuanced understanding of cultural competence, with a focus on developing models that can reason and understand cultural contexts. This shift is driven by the need for AI systems to be deployed in diverse environments, where cultural awareness and sensitivity are crucial. Researchers are working on creating benchmarks and evaluation metrics that can assess the cultural competence of large language models, moving beyond traditional methods that focus on de-contextualized correctness or forced-choice judgments. The development of frameworks such as Cultural Norm-based Cultural Alignment (CNCA) and benchmarks like TCM-5CEval are notable examples of this trend. Noteworthy papers include CURE, which introduces a set of benchmarks for culturally grounded reasoning, and TCM-5CEval, which provides a comprehensive evaluation of large language models in Traditional Chinese Medicine. These papers highlight the importance of cultural understanding and reasoning in AI systems and demonstrate the potential for reasoning models to better reflect diverse human values through culturally informed alignment strategies.