Neural Machine Translation (NMT) models are known to suffer from noisy
i...
This paper presents the JHU-Microsoft joint submission for WMT 2021 qual...
We propose a novel scheme to use the Levenshtein Transformer to perform ...
Saliency methods are widely used to interpret neural network predictions...
We present Espresso, an open-source, modular, extensible end-to-end neur...
Despite their original goal to jointly learn to align and translate, Neu...
Most neural machine translation systems are built upon subword units
ext...
Stack Long Short-Term Memory (StackLSTM) is useful for various applicati...
In recent years, end-to-end models have become popular for application i...
Using pre-trained word embeddings as input layer is a common practice in...
We present a new end-to-end architecture for automatic speech recognitio...