AI RESEARCH

Graph Transductive Sharpening: Leveraging Unlabeled Predictions in Node Classification

arXiv CS.LG

ArXi:2605.20248v1 Announce Type: new In the transductive setting, where the full graph is observed but node labels are only partially available, progress in semi-supervised node classification has largely focused on architectural innovation. In this paper, we revisit an orthogonal axis: the