How can AI models be made more interpretable in their mathematical reasoning processes?Answer not yet generated.