Structural Information Preserving for Graph-to-Text Generation

By Linfeng Song, Ante Wang, Jinsong Su, Yue Zhang, Kun Xu, Yubin Ge, Dong Yu. 2020

58 citations recorded by the source corpus.

The task of graph-to-text generation aims at producing sentences that preserve the meaning of input graphs. As a crucial defect, the current state-of-the-art models may mess up or even drop the core structural information of input graphs when generating outputs. We propose to tackle this problem by leveraging richer training signals that can guide our model for preserving input information. In particular, we introduce two types of autoencoding losses, each individually focusing on different aspects (a.k.a. views) of input graphs. The losses are then back-propagated to better calibrate our model via multi-task training. Experiments on two benchmarks for graph-to-text generation show the effectiveness of our approach over a state-of-the-art baseline.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Graph Learning, Natural Language Generation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Caiming Xiong
2. Mohit Bansal
3. Hinrich Schütze
4. Luke S. Zettlemoyer
5. Percy Liang
6. Richard Socher
7. Iryna Gurevych
8. Dragomir R. Radev

In Computer Vision and Pattern Recognition

1. Caiming Xiong
2. Luc Van Gool
3. Mohit Bansal
4. Trevor Darrell
5. Dacheng Tao
6. Cordelia Schmid
7. Li Fei-Fei
8. Bernt Schiele

In Graph Learning

1. Leonardo F. R. Ribeiro
2. Xi Victoria Lin
3. Hinrich Schütze
4. Iryna Gurevych
5. Richard Socher
6. Caiming Xiong
7. Dragomir R. Radev
8. Mohit Bansal

In Natural Language Generation

1. Hinrich Schütze
2. Martin Schmitt
3. Leonardo F. R. Ribeiro
4. Iryna Gurevych
5. Rui Zhang
6. Richard Socher
7. Caiming Xiong
8. Dragomir R. Radev

Citations in the corpus

• DART: Open-Domain Structured Data Record to Text Generation (2020, background): “Recently, neural encoder-decoder models based on attention and copy mechanisms have shown promising results (Wiseman et al., 2017; Liu et al., 2018; Gehrmann et al., 2018; Puduppully et al., 2018, 2019; Iso et al., 2019; Castro Ferreira et al., 2019; Shen et al., 2019; Roberti et al., 2019; Zhu et a”
• Text2Event: Controllable Sequence-to-Structure Generation for End-to-end Event Extraction (2021, methodology): “…and has been widely used in many NLP tasks, such as machine translation (Kalchbrenner and Blunsom, 2013), semantic parsing (Dong and Lapata, 2016; Song et al., 2020), entity extraction (Strakov´a et al., 2019), relation extraction (Zeng et al., 2018; Zhang et al., 2020b), and aspect term…”
• Investigating Pretrained Language Models for Graph-to-Text Generation (2020, methodology): “Recent methods (Zhu et al., 2019; Cai and Lam, 2020; Wang et al., 2020; Song et al., 2020) employ Transformers to learn node representations injecting the graph structure into the self-attention”
• A Survey of Controllable Text Generation Using Transformer-based Pre-trained Language Models (2022, background): “For instance, the lexical-level overlap between the reference text containing the control elements and the generated text can be used to reflect how well the control conditions are met in the task of table/graph to text [102, 111, 112, 128, 131, 165] and story generation [33, 40].”