Jianan Zhao, a computer science student, joins to discuss using graphs with LLMs efficiently. They explore graph inductive bias, graph machine learning, limitations of natural language models for graphs, graph text as a preprocessing step, information loss in translation process, and comparison with graph neural networks.