Bridging Local Details and Global Context in Text-Attributed Graphs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yaoke, Zhu, Yun, Zhang, Wenqiao, Zhuang, Yueting, Li, Yunfei, Tang, Siliang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916436229750784
author Wang, Yaoke
Zhu, Yun
Zhang, Wenqiao
Zhuang, Yueting
Li, Yunfei
Tang, Siliang
author_facet Wang, Yaoke
Zhu, Yun
Zhang, Wenqiao
Zhuang, Yueting
Li, Yunfei
Tang, Siliang
contents Representation learning on text-attributed graphs (TAGs) is vital for real-world applications, as they combine semantic textual and contextual structural information. Research in this field generally consist of two main perspectives: local-level encoding and global-level aggregating, respectively refer to textual node information unification (e.g., using Language Models) and structure-augmented modeling (e.g., using Graph Neural Networks). Most existing works focus on combining different information levels but overlook the interconnections, i.e., the contextual textual information among nodes, which provides semantic insights to bridge local and global levels. In this paper, we propose GraphBridge, a multi-granularity integration framework that bridges local and global perspectives by leveraging contextual textual information, enhancing fine-grained understanding of TAGs. Besides, to tackle scalability and efficiency challenges, we introduce a graphaware token reduction module. Extensive experiments across various models and datasets show that our method achieves state-of-theart performance, while our graph-aware token reduction module significantly enhances efficiency and solves scalability issues.
format Preprint
id arxiv_https___arxiv_org_abs_2406_12608
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Bridging Local Details and Global Context in Text-Attributed Graphs
Wang, Yaoke
Zhu, Yun
Zhang, Wenqiao
Zhuang, Yueting
Li, Yunfei
Tang, Siliang
Computation and Language
Artificial Intelligence
Representation learning on text-attributed graphs (TAGs) is vital for real-world applications, as they combine semantic textual and contextual structural information. Research in this field generally consist of two main perspectives: local-level encoding and global-level aggregating, respectively refer to textual node information unification (e.g., using Language Models) and structure-augmented modeling (e.g., using Graph Neural Networks). Most existing works focus on combining different information levels but overlook the interconnections, i.e., the contextual textual information among nodes, which provides semantic insights to bridge local and global levels. In this paper, we propose GraphBridge, a multi-granularity integration framework that bridges local and global perspectives by leveraging contextual textual information, enhancing fine-grained understanding of TAGs. Besides, to tackle scalability and efficiency challenges, we introduce a graphaware token reduction module. Extensive experiments across various models and datasets show that our method achieves state-of-theart performance, while our graph-aware token reduction module significantly enhances efficiency and solves scalability issues.
title Bridging Local Details and Global Context in Text-Attributed Graphs
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.12608