Progressive Feature Completion Framework Improves Node Clustering on Incomplete Real-World Graphs
Computer scientists designed a progressive completion and multi-level contrastive framework that preserves graph clustering accuracy when node attributes are missing.

