.png){kind=icon}
.png){kind=icon}
Return
HIM2A: Hierarchical interactive multi-modal entity alignment with semantic augmentation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.aei.2025.104136.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Tackles multi-modal entity alignment by proposing HIM2A. • Identifies key flaws in early structural fusion and weak textual modeling. • Proposes SIA to enrich entity semantics with external factual descriptions. • Designs TGIS to select representative images via cross-modal similarity.
Journal
IF:
9.9
Papers:
4.0K
Citations:
1.7W