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