Many recent advances in natural language generation have been fueled by
...
Semi-parametric models, which augment generation with retrieval, have le...
Machine translation models struggle when translating out-of-domain text,...
Transformers struggle when attending to long contexts, since the amount ...
We introduce GEM, a living benchmark for natural language Generation (NL...
Current state-of-the-art text generators build on powerful language mode...
Visual attention mechanisms are widely used in multimodal tasks, such as...
Named entity recognition (NER) and entity linking (EL) are two fundament...
Using natural language to give instructions to robots is challenging, si...