06-08 论文阅读——Harnessing explanations Llm-to-lm interpreter for enhanced text-attributed graph representation learning