google-deepmind / google-deepmind/deepmind-research
examples of the Wikigraps generated papers?
Open
- Dominant language
- Jupyter Notebook
- Stars
- 15.2k
- Forks
- 2.9k
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
Any plans to publish examples of the generated "test" generated papers? It would be interesting to have human assessment of these. An undressed question is why the BLEU are so close between the BoW and GNN models. The similar results were obtained in http://arxiv.org/abs/2007.08426 : "perform well even when the input graph representation is reduced to a simple bag of node and edge labels"
Contributor guide
Assessment
This issue has not been assessed yet.