logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

GraphProbe: Knowledge Probing for Graph Representation Learning

delete2025-10-04
delete0
PRE
AI
M
Mingyu Zhao
X
Xingyu Huang
S
Songming Zhang
刘安齐 cover
刘安齐 (Anqi Liu)
Z
Ziyu Lyu
Y
Yanlin Wang
崔丽欣 cover
崔丽欣 (Lixin Cui)
白
白璐 (Lu Bai)
DOI：10.1016/j.patcog.2025.112518delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Propose GraphProbe, a novel framework for knowledge probing in graph representation learning. • Design node-/path-/structural-level probes to explore inherent graph property encoding. • Construct a benchmark (10 GNN models, 6 datasets) to verify GraphProbe’s effectiveness.

Journal

Pattern Recognition cover
Pattern Recognition
IF:
7.6
Papers:
1.3W
Citations:
4.5W

Organization

C
Central University of Finance and Economics
Scholars:
560
Papers: 399
Citations: 2.8K
B
Beijing Normal University
Scholars:
3.3W
Papers: 2.7W
Citations: 4.2W
S
sun yat-sen university
Scholars:
1.9W
Papers: 6.4K
Citations: 14
S
Shenzhen Institute of Advanced Technology
Scholars:
362
Papers: 203
Citations: 1
researcher View more organizations