Search papers, labs, and topics across Lattice.
Tianjin University
2
0
4
6
Overcome the heterogeneity hurdle: MUG pre-trains a single graph encoder that transfers across diverse heterogeneous graphs by unifying node/relation types and learning consistent structural patterns across meta-paths.
Achieve SOTA in cross-domain graph few-shot learning by adaptively aligning diverse graph features into a shared semantic space.
