AI RESEARCH
OCN: Effectively Utilizing Higher-Order Common Neighbors for Better Link Prediction
arXiv CS.LG
•
ArXi:2505.19719v2 Announce Type: replace Common Neighbors (CNs) and their higher-order variants are important pairwise features widely used in state-of-the-art link prediction methods. However, existing methods often struggle with the repetition across different orders of CNs and fail to fully leverage their potential. We identify that these limitations stem from two key issues: redundancy and over-smoothing in high-order common neighbors.