Modern machine learning systems are able to learn extremely complicated relations from data.

T o this end, we propose a simple yet powerful paradigm for seamlessly unifying different human pose and shape-related tasks and datasets. Our formulation is centered on the ability - both at training and test time - to query any arbitrary point of the human volume, and obtain its estimated location in 3D. We achieve this by learning a continuous neural field of body point localizer functions, each of which is a differently parameterized 3D heatmap-based convolutional point localizer (detector).

Recurrent neural networks (RNNs) notoriously struggle to learn long-term memories, primarily due to vanishing and exploding gradients.

Motion is a powerful cue that humans use to detect and segment visual objects.

However, practical model deployment often operates in the gap between these two limit cases, as real-world applications demand adaptation to specific subdomains, tasks or concepts -- spread over the entire, varying life cycle of a model.

Randomized smoothing is a popular certified defense against adversarial attacks.

We study causal effect estimation in a setting where the data are not i.i.d.

Similar ideas can be found in recent developments inspired by attention, such as RWKV [Peng et al.,