Why do large language models sometimes output factual inaccuracies and
e...
Good data augmentation is one of the key factors that lead to the empiri...
Algorithmic reasoning requires capabilities which are most naturally
und...
The vast majority of work in self-supervised learning, both theoretical ...
Noise-contrastive estimation (NCE) is a statistically consistent method ...
Contrastive learning is a family of self-supervised methods where a mode...
Our goal is to predict future video frames given a sequence of input fra...