Deep Graph Convolutional Encoders for Structured Data to Text Generation
This work addresses text generation from structured data for NLP applications, representing an incremental improvement over existing methods.
The paper tackled the problem of generating text from graph-structured data by proposing a graph convolutional network encoder that directly exploits input structure, achieving improved results on two graph-to-sequence datasets compared to standard sequence-to-sequence methods.
Most previous work on neural text generation from graph-structured data relies on standard sequence-to-sequence methods. These approaches linearise the input graph to be fed to a recurrent neural network. In this paper, we propose an alternative encoder based on graph convolutional networks that directly exploits the input structure. We report results on two graph-to-sequence datasets that empirically show the benefits of explicitly encoding the input graph structure.