Can GNN be Good Adapter for LLMs?arxiv.org/abs/2402.12984 Introduction 最近,有大量的研究探索GNN和LLM的耦合结构。本文旨在预训练一个框架,它可以通过利用丰富的文本特性来结合图结构和LLM。因此本文创造地提出利用GNN作为适配器(Adapter),其优势如下: 轻量级:一个GNN适配器引入了一些可训练的参数和较低的...
LLMs. Therefore, we propose GraphAdapter, which uses a graph neural network (GNN) as an efficient adapter in collaboration with LLMs to tackle TAGs. In terms of efficiency, the GNN adapter introduces only a few trainable parameters and can be trained with low computation costs. The entire ...