.png){kind=icon}
.png){kind=icon}
Return
Multi-pattern Structure Refinement and Fusion for Heterophilic Graphs
.png){kind=icon}
PRE
AI
DOI：10.1016/j.inffus.2026.104381.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A dual-phase strategy of active refinement and consensus fusion is designed. • The structure refinement refines diverse patterns to enhance latent homophily. • A fusion mechanism is designed to achieve semantic consensus across all patterns. • Experiments show that MSRF handles HGSS and learns robust node representations.
Keywords:
Multi-pattern Structure Refinement
Consensus Fusion
Heterophilic Graphs
Node Representations
Latent Homophily
Journal
IF:
15.5
Papers:
4.1K
Citations:
2.7W