JointGT: Graph-Text Joint Representation Learning for Text Generation from Knowledge Graphs

TL;DR

JointGT introduces a novel graph-text joint representation learning framework that explicitly models graph structure and alignment, significantly improving KG-to-text generation performance.

Contribution

The paper proposes a structure-aware semantic aggregation module and three new pre-training tasks to better encode graph structure and enhance graph-text alignment.

Findings

Achieves state-of-the-art results on KG-to-text datasets

Effectively preserves graph structure during encoding

Improves graph-text alignment through novel pre-training tasks

Abstract

Existing pre-trained models for knowledge-graph-to-text (KG-to-text) generation simply fine-tune text-to-text pre-trained models such as BART or T5 on KG-to-text datasets, which largely ignore the graph structure during encoding and lack elaborate pre-training tasks to explicitly model graph-text alignments. To tackle these problems, we propose a graph-text joint representation learning model called JointGT. During encoding, we devise a structure-aware semantic aggregation module which is plugged into each Transformer layer to preserve the graph structure. Furthermore, we propose three new pre-training tasks to explicitly enhance the graph-text alignment including respective text / graph reconstruction, and graph-text alignment in the embedding space via Optimal Transport. Experiments show that JointGT obtains new state-of-the-art performance on various KG-to-text datasets.