Deep Graph Convolutional Encoders for Structured Data to Text Generation

WS 2018 Diego MarcheggianiLaura Perez-Beltrachini

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... (read more)

PDF Abstract

Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Data-to-Text Generation SR11Deep GCN + feat BLEU 0.666 # 1
Data-to-Text Generation WebNLG GCN EC BLEU 0.559 # 1