Diffusion models have demonstrated excellent potential for generating di...
Autoregressive models for text sometimes generate repetitive and low-qua...
3D-aware image synthesis encompasses a variety of tasks, such as scene
g...
Diffusion models have recently become the de-facto approach for generati...
Training stability is of great importance to Transformers. In this work,...
Recent Self-Supervised Learning (SSL) methods are able to learn feature
...
Diffusion probabilistic models have quickly become a major approach for
...
Novel view synthesis from a single image requires inferring occluded reg...
Diffusion models (DMs) have recently emerged as SoTA tools for generativ...
This paper presents a Progressively-connected Light Field network (ProLi...
Recent work in multilingual translation advances translation quality
sur...
We propose an autoregressive entity linking model, that is trained with ...
We describe a method to jointly pre-train speech and text in an
encoder-...
Prompt tuning is a new, efficient NLP transfer learning paradigm that ad...
Direct speech-to-speech translation (S2ST) models suffer from data scarc...
While the general idea of self-supervised learning is identical across
m...
We present a textless speech-to-speech translation (S2ST) system that ca...
We propose StyleNeRF, a 3D-aware generative model for photo-realistic
hi...
This paper presents fairseq S^2, a fairseq extension for speech synthesi...
We present a direct speech-to-speech translation (S2ST) model that trans...
Neural volume rendering became increasingly popular recently due to its
...
We propose Neural Actor (NA), a new method for high-quality synthesis of...
Adapter modules were recently introduced as an efficient alternative to
...
Fully non-autoregressive neural machine translation (NAT) is proposed to...
Pre-trained language models have proven their unique powers in capturing...
This paper describes Facebook AI's submission to WMT20 shared news
trans...
Neural sequence models can generate highly fluent sentences but recent
s...
We introduce dual-decoder Transformer, a new model architecture that joi...
Recent work demonstrates the potential of multilingual pretraining of
cr...
Simultaneous translation on both text and speech focuses on a real-time ...
Photo-realistic free-viewpoint rendering of real-world scenes using clas...
Arguably one of the top success stories of deep learning is transfer
lea...
End-to-end speech-to-text translation can provide a simpler and smaller
...
Recent studies have demonstrated the cross-lingual alignment ability of
...
Transfer learning from high-resource languages is known to be an efficie...
Spoken language translation has recently witnessed a resurgence in
popul...
This paper demonstrates that multilingual denoising pre-training produce...
State-of-the-art neural machine translation models generate a translatio...
Non-autoregressive machine translation (NAT) systems predict a sequence ...
State of the art sequence-to-sequence models perform a fixed number of
c...
Self-training is one of the earliest and simplest semi-supervised method...
While we live in an increasingly interconnected world, different places ...
Simultaneous machine translation models start generating a target sequen...
Posterior collapse plagues VAEs for text, especially for conditional tex...
For automatic speech translation (AST), end-to-end approaches are
outper...
For automatic speech translation (AST), end-to-end approaches are
outper...
Automatic evaluation of text generation tasks (e.g. machine translation,...
Almost all existing machine translation models are built on top of
chara...
Zero-shot translation, translating between language pairs on which a Neu...
Modern neural sequence generation models are built to either generate to...