Text Generation from Knowledge Graphs with Graph Transformers
By Rik Koncel-Kedziorski, Dhanush Bekal, Yi Luan, Mirella Lapata, Hannaneh Hajishirzi. 2019
363 citations recorded by the source corpus.
Generating texts which express complex ideas spanning multiple sentences requires a structured representation of their content (document plan), but these representations are prohibitively expensive to manually produce. In this work, we address the problem of generating coherent multi-sentence texts from the output of an information extraction system, and in particular a knowledge graph. Graphical knowledge representations are ubiquitous in computing, but pose a significant challenge for text generation techniques due to their non-hierarchical nature, collapsing of long-distance dependencies, and structural variety. We introduce a novel graph transforming encoder which can leverage the relational structure of such knowledge graphs without imposing linearization or hierarchical constraints. Incorporated into an encoder-decoder setup, we provide an end-to-end trainable system for graph-to-text generation that we apply to the domain of scientific text. Automatic and human evaluations show that our technique produces more informative texts which exhibit better document structure than competitive encoder-decoder methods.
Subfields: Artificial Intelligence, 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
- Mohit Bansal
- Jianfeng Gao
- Yejin Choi
- Percy Liang
- Luke S. Zettlemoyer
- Noah A. Smith
- Graham Neubig
- Kyunghyun Cho
In Natural Language Generation
- Leonardo F. R. Ribeiro
- Hinrich Schütze
- Martin Schmitt
- Iryna Gurevych
- Oriol Vinyals
- Mohit Bansal
- Dong Yu
- Yejin Choi
Citations in the corpus
- DART: Open-Domain Structured Data Record to Text Generation (2020, neutral): “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”
- GraDA: Graph Generative Data Augmentation for Commonsense Reasoning (2022, methodology): “We refer the reader to Koncel-Kedziorski et al. (2019)”
- Evaluation of Text Generation: A Survey (2020, methodology): “, 2015) has been used in NLG tasks (Kiritchenko & Mohammad, 2016; Koncel-Kedziorski et al., 2019) to simplify comparative evaluations; best-worst scaling asks participants to choose the best and worst elements from a set of candidates, a simpler task than fully ranking the set that still provides re”
- Graph-to-Tree Learning for Solving Math Word Problems (2020, methodology): “In community of natural language processing, the first graph transformer was introduce in a knowledgegraph-to-text task (Koncel-Kedziorski et al., 2019), where a graph attention Network (Veličković et al., 2018) is used with a transformer style architecture.”
- Knowledge Graph-Augmented Abstractive Summarization with Semantic-Driven Cloze Reward (2020, methodology): “Our graph encoder improves upon Graph Attention Networks (GATs) (Veliˇckovi´c et al., 2018) by adding residual connections between layers as discussed in Koncel-Kedziorski et al. (2019).”
- A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, background): “[183] studied text generation for information extraction systems, and proposed a graph transforming encoder for graph-to-text generation from the knowledge graph.”
- WikiGraphs: A Wikipedia Text - Knowledge Graph Paired Dataset (2021, methodology): “The most recent prior work on graph-to-text generation follows an encoder-decoder architecture (Koncel-Kedziorski et al., 2019; Jin et al., 2020), where the graph part is encoded with a GNN model, e.”
- Neural Machine Translation: A Review (2019, methodology): “Lattices have also been used to augment the input with external knowledge sources such as knowledge graphs (Koncel-Kedziorski et al., 2019; Moussallem et al., 2019) or semantic predicate-argument structures (Marcheggiani et al.”
- A Comprehensive Survey of AI-Generated Content (AIGC): A History of Generative AI from GAN to ChatGPT (2023, background): “[178] encode graph structure for text generation with a transformer-based architecture extended from the graph attention network (GAT) [179].”
- Structural Information Preserving for Graph-to-Text Generation (2020, background): “Previous work for neural graph-to-text generation (Konstas et al., 2017; Song et al., 2018; Beck et al., 2018; Trisedya et al., 2018; Marcheggiani and Perez-Beltrachini, 2018; Xu et al., 2018; Cao and Clark, 2019; Damonte and Cohen, 2019; Hajdik et al., 2019; Koncel-Kedziorski et al., 2019; Hong et ”
- Cooperative Generator-Discriminator Networks for Abstractive Summarization with Narrative Flow (2019, background): “Generation with Narrative Flow Due to the need for accurate understanding of long-distance dependencies and narrative structure, modeling coherent narrative flow has proved to be a major challenge in the field of text generation, particularly for scientific documents (Koncel-Kedziorski et al., 2019;”
- Large Scale Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training (2020, methodology): “, 2018), graph transformers over structured data (Koncel-Kedziorski et al., 2019), and text-to-text generation with T5”
- Investigating Pretrained Language Models for Graph-to-Text Generation (2020, methodology): “While the texts in KGAIA and AGENDA share the domain of scientific abstracts, texts in WebNLG are more general.”
- Text Editing by Command (2020, background): “Thus, several works have explored using various forms of grounding to enable models to generate factually consistent texts (Koncel-Kedziorski et al., 2019; Liu et al., 2018b; Prab-humoye et al., 2019; Liu et al., 2018a; Guu et al., 2020).”
- A Survey of Knowledge-enhanced Text Generation (2020, methodology): “re conceptual flow in conversation/dialogue systems. However, many NLG tasks might not need to follow a reasoning process but to better understand global context under a particular generation process [56, 65, 75]. For instance, the summarization task requires structured representation to facilitate ”
- Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training (2021, background): “datasets, such as graph transformers over structured data (Koncel-Kedziorski et al., 2019), latent templates for interpretability (Wiseman et al.”
- DA-Transformer: Distance-aware Transformer (2020, methodology): “We use Adam (Kingma and Ba, 2015) as the optimization algorithm.”
- Modeling Global and Local Node Contexts for Text Generation from Knowledge Graphs (2020, neutral): “Figure 5 shows the effect of the size of a graph, defined as the number of nodes, on the quality (measured in CHRF++ scores) and length of the generated text (in number of words) in the AGENDA dev set.”