Massively Multilingual Neural Machine Translation
BitTrain: Sparse Bitmap Compression for Memory-Efficient Training on the Edge
PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning
Res2Net: A New Multi-scale Backbone Architecture
CE-Net: Context Encoder Network for 2D Medical Image Segmentation
Mixed Precision Training
Denoising Diffusion Implicit Models
Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters
Deep Directed Generative Autoencoders
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
Webly Supervised Joint Embedding for Cross-Modal Image-Text Retrieval
Gaussian Error Linear Units (GELUs)
Making Pre-trained Language Models Better Few-shot Learners
Deep Fragment Embeddings for Bidirectional Image Sentence Mapping
Robust Optimization for Multilingual Translation with Imbalanced Data
Whitening Sentence Representations for Better Semantics and Faster Retrieval
Neural Natural Language Inference Models Enhanced with External Knowledge
Adversarial Representation Learning for Robust Privacy Preservation in\n Audio
Ecological Consequences of Trophic Cascades: A Global Perspective
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
