3260 papers • 126 benchmarks • 313 datasets
This task has no description! Would you like to contribute one?
(Image credit: Papersgraph)
These leaderboards are used to track progress in unsupervised-kg-to-text-generation-2
Use these libraries to find unsupervised-kg-to-text-generation-2 models and implementations
No subtasks available.
Inspired by recent work on unsupervised machine translation, this work serializes a KG as a sequence of facts and frames both tasks as sequence translation, and learns to map both graphs and texts into a joint semantic space and thus generalizes over different surface representations with the same meaning.
This paper proposes a large-scale, general-domain dataset, GenWiki, which has 1.3M text and graph examples, respectively, and provides this new benchmark dataset for future research on unsupervised text generation from knowledge graphs.
Adding a benchmark result helps the community track progress.