AI
ML
NLP
This groundbreaking paper introduces the Transformer architecture, revolutionizing natural language processing by relying entirely on attention mechanisms without recurrence or convolution.
CV
ML
Introduces residual networks (ResNets) that enable training of extremely deep neural networks by using skip connections to address the vanishing gradient problem.
Robotics
AI
Presents a maximum entropy reinforcement learning algorithm that achieves state-of-the-art performance on a range of continuous control benchmark tasks.
NLP
AI
Introduces BERT, a method for pre-training language representations that obtains state-of-the-art results on eleven natural language processing tasks.
CV
ML
Introduces Generative Adversarial Networks, a novel framework for estimating generative models via an adversarial process between two neural networks.
NLP
AI
The original GPT paper that demonstrates how unsupervised pre-training of a language model on a diverse corpus can achieve significant gains on discriminative tasks.