An Unsupervised Joint System for Text Generation from Knowledge Graphs and Semantic Parsing
This addresses the challenge of adapting knowledge graph and text conversion systems across diverse domains with little data overlap, offering a more flexible solution for NLP applications.
The paper tackles the problem of domain-specific data scarcity for graph-to-text generation and text-to-graph semantic parsing by proposing an unsupervised joint system that eliminates the need for annotated data and domain adaptation. It outperforms strong baselines on WebNLG v2.1 and a new Visual Genome benchmark without manual adaptation between datasets.
Knowledge graphs (KGs) can vary greatly from one domain to another. Therefore supervised approaches to both graph-to-text generation and text-to-graph knowledge extraction (semantic parsing) will always suffer from a shortage of domain-specific parallel graph-text data; at the same time, adapting a model trained on a different domain is often impossible due to little or no overlap in entities and relations. This situation calls for an approach that (1) does not need large amounts of annotated data and thus (2) does not need to rely on domain adaptation techniques to work well in different domains. To this end, we present the first approach to unsupervised text generation from KGs and show simultaneously how it can be used for unsupervised semantic parsing. We evaluate our approach on WebNLG v2.1 and a new benchmark leveraging scene graphs from Visual Genome. Our system outperforms strong baselines for both text$\leftrightarrow$graph conversion tasks without any manual adaptation from one dataset to the other. In additional experiments, we investigate the impact of using different unsupervised objectives.