Minh-Thang Luong
Research Scientist at
We present a combined scaling method called BASIC that achieves 85.7
zer...
Despite their recent successes in tackling many NLP tasks, large-scale
p...
We introduce Electric, an energy-based cloze model for representation
le...
Despite recent success, most contrastive self-supervised learning method...
Masked language modeling (MLM) pre-training methods such as BERT corrupt...
We present Meena, a multi-turn open-domain chatbot trained end-to-end on...
We propose a language-independent approach for improving statistical mac...
We present a simple self-training method that achieves 87.4
on ImageNet,...
This document describes the findings of the Third Workshop on Neural
Gen...
It can be challenging to train multi-task neural networks that outperfor...
We introduce a pretraining technique called Selfie, which stands for
SEL...
Despite its success, deep learning still needs large labeled datasets to...
Unsupervised representation learning algorithms such as word2vec and ELM...
Latent variable models have been a preferred choice in conversational
mo...
This document describes the findings of the Second Workshop on Neural Ma...
Current end-to-end machine reading and question answering (Q&A) models a...
Neural networks have excelled at many NLP tasks, but there remain open
q...
The standard content-based attention mechanism typically used in
sequenc...
Recurrent neural network models with an attention mechanism have proven ...
Neural Machine Translation (NMT) has shown remarkable progress over the ...
Neural Machine Translation (NMT), like many other deep learning domains,...
Nearly all previous work on neural machine translation (NMT) has used qu...
Sequence to sequence learning has recently emerged as a new paradigm in
...
An attentional mechanism has lately been used to improve neural machine
...
Recursive neural models, which use syntactic parse trees to recursively
...
Neural Machine Translation (NMT) is a new approach to machine translatio...