.png){kind=icon}
.png){kind=icon}
Return
ComCon: Complementary-Contradictory Regularization for Multimodal Knowledge Graph Completion
.png){kind=icon}
PRE
AI
DOI：10.1016/j.inffus.2025.104016.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• ComCon: Novel regularization using complementary-contradictory features for MMKGC. • Resolves modality misalignment via constrained multimodal feature interactions. • Weighted Negative Sampling (WNS) reduces false negative impacts adaptively.
Journal
IF:
15.5
Papers:
4.1K
Citations:
2.7W