keyword
text-attributed graphs
A text-attributed graph is a data structure in which graph elements, primarily nodes and sometimes edges, are associated with natural language text such as documents, titles, or descriptions. Unlike standard graphs that rely exclusively on topological connectivity or basic numerical feature vectors, text-attributed graphs combine structural relational information with rich, unstructured semantic content. Typical examples include academic citation networks where research papers contain titles and abstracts connected by citations, e-commerce networks where items with product descriptions are linked by purchasing behaviors, and social networks connecting user profiles and posts. Learning tasks on text-attributed graphs, such as node classification and link prediction, commonly require combining language models to interpret textual semantics with graph neural networks to model relational dependencies across the network topology.
2 items
Position: Graph Foundation Models Are Already Here
Haitao Mao, Zhikai Chen, Wenzhuo Tang, Jianan Zhao, Yao Ma, Tong Zhao, Neil Shah, Mikhail Galkin, Jiliang Tang
Why you should read this
Presents a unifying graph vocabulary perspective that explains how current primitive models achieve cross-dataset transferability and provides concrete principles to guide the design of general-purpose Graph Foundation Models.
Added
2026-10-02
Label-free Node Classification on Graphs with Large Language Models (LLMs)
Zhikai Chen, Haitao Mao, Hongzhi Wen, Haoyu Han, Wei Jin, Haiyang Zhang, Hui Liu, Jiliang Tang
Why you should read this
Proposes LLM-GNN, an active annotation framework that prompts large language models on an informative subset of graph nodes to train graph neural networks without human supervision, achieving high classification accuracy on massive text-attributed graphs for under one dollar.
In recent years, there have been remarkable advancements in node classification achieved by Graph Neural Networks (GNNs). However, they necessitate abundant high-quality labels to ensure promising performance. In contrast, Large Language Models (LLMs) exhibit impressive zero-shot proficiency on text-attributed graphs. Yet, they face challenges in efficiently processing structural data and suffer from high inference costs. In light of these observations, this work introduces a label-free node classification on graphs with LLMs pipeline, LLM-GNN. It amalgamates the strengths of both GNNs and LLMs while mitigating their limitations. Specifically, LLMs are leveraged to annotate a small portion of nodes and then GNNs are trained on LLMs' annotations to make predictions for the remaining large portion of nodes. The implementation of LLM-GNN faces a unique challenge: how can we actively select nodes for LLMs to annotate and consequently enhance the GNN training? How can we leverage LLMs to obtain annotations of high quality, representativeness, and diversity, thereby enhancing GNN performance with less cost? To tackle this challenge, we develop an annotation quality heuristic and leverage the confidence scores derived from LLMs to advanced node selection. Comprehensive experimental results validate the effectiveness of LLM-GNN. In particular, LLM-GNN can achieve an accuracy of 74.9% on a vast-scale dataset \products with a cost less than 1 dollar.
Added
2026-09-26