Sequence Generation
Related Papers
Fast Interleaved Bidirectional Sequence Generation
AbstractIndependence assumptions during sequence generation can speed up inference, but parallel generation of highly inter-dependent tokens comes at a cost in quality. Instead of assuming independence between neighbouring tokens (semi-autoregressive …...
Leveraging WordNet Paths for Neural Hypernym Prediction
AbstractWe formulate the problem of hypernym prediction as a sequence generation task, where the sequences are taxonomy paths in WordNet. Our experiments with encoder-decoder models show that training to generate …...
On Long-Tailed Phenomena in Neural Machine Translation
AbstractState-of-the-art Neural Machine Translation (NMT) models struggle with generating low-frequency tokens, tackling which remains a major challenge. The analysis of long-tailed phenomena in the context of structured prediction tasks is …...
Augmentation, Retrieval, Generation: Event Sequence Prediction with a Three-Stage Sequence-to-Sequence Approach
AbstractBeing able to infer possible events related to a specific target is critical to natural language processing. One challenging task in this line is event sequence prediction, which aims at …...
End-to-end Dense Video Captioning as Sequence Generation
AbstractDense video captioning aims to identify the events of interest in an input video, and generate descriptive captions for each event. Previous approaches usually follow a two-stage generative process, which …...