Enhancing AMR-to-Text Generation with Dual Graph Representations

Abstract : Generating text from graph-based data, such as Abstract Meaning Representation (AMR), is a challenging task due to the inherent difficulty in how to properly encode the structure of a graph with labeled edges. To address this difficulty , we propose a novel graph-to-sequence model that encodes different but complementary perspectives of the structural information contained in the AMR graph. The model learns parallel top-down and bottom-up representations of nodes capturing contrasting views of the graph. We also investigate the use of different node message passing strategies, employing different state-of-the-art graph en-coders to compute node representations based on incoming and outgoing perspectives. In our experiments, we demonstrate that the dual graph representation leads to improvements in AMR-to-text generation, achieving state-of-the-art results on two AMR datasets.
Document type :
Conference papers
Complete list of metadatas

Cited literature [40 references]  Display  Hide  Download

Contributor : Claire Gardent <>
Submitted on : Tuesday, September 3, 2019 - 1:03:09 PM
Last modification on : Thursday, September 5, 2019 - 1:20:39 AM


Files produced by the author(s)


  • HAL Id : hal-02277053, version 1



Leonardo Ribeiro, Claire Gardent, Iryna Gurevych. Enhancing AMR-to-Text Generation with Dual Graph Representations. 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, Nov 2019, Hong Kong, China. ⟨hal-02277053⟩



Record views


Files downloads