Due to the versatile nature of the GNN modeling formalism, many fundamentally different tasks are studied in the research area, and it should be noted that good results on one task often do not transfer over to other tasks.
We presented a framework for extending sequence encoders with a graph component that can leverage rich additional structure. In an evaluation on three different summarization tasks, we have shown that this augmentation improves the performance of a …
We presented graph partition neural networks, which extend graph neural networks. Relying on graph partitions, our model alternates between locally propagating information between nodes in small subgraphs and globally propagating information between…
Learning tasks on source code (i.e., formal languages) have been considered recently, but most work has tried to transfer natural language methods and does not capitalize on the unique opportunities offered by code's known syntax. For example, long-…
Credentials & highlights
- Project Views