Modeling Graph Structure via Relative Position for Text Generation from Knowledge Graphs

Abstract

We present Graformer, a novel Transformerbased encoder-decoder architecture for graphto-text generation. With our novel graph selfattention, the encoding of a node relies on all nodes in the input graph - not only direct neighbors - facilitating the detection of global patterns. We represent the relation between two nodes as the length of the shortest path between them. Graformer learns to weight these nodenode relations differently for different attention heads, thus virtually learning differently connected views of the input graph. We evaluate Graformer on two popular graph-to-text generation benchmarks, AGENDA and WebNLG, where it achieves strong performance while using many fewer parameters than other approaches.