Adversarial training is widely used to make classifiers robust to a spec...
Self-supervised methods have achieved remarkable success in transfer
lea...
Test-time adaptation harnesses test inputs to improve the accuracy of a ...
The promise of self-supervised learning (SSL) is to leverage large amoun...
Adaptive defenses that use test-time optimization promise to improve
rob...
General perception systems such as Perceivers can process arbitrary
moda...
Domain adaptation seeks to mitigate the shift between training on the
so...
The recently-proposed Perceiver model obtains good results on several do...
Adversarial attacks optimize against models to defeat defenses. Existing...
Anytime inference requires a model to make a progression of predictions ...
We recall that certain common losses are simplified likelihoods and inst...
Faced with new and different data during testing, a model must adapt its...
Given the variety of the visual world there is not one true scale for
re...
The visual world is vast and varied, but its variations divide into
stru...
We propose infinite mixture prototypes to adaptively represent both simp...
The current dominant paradigm for imitation learning relies on strong
su...
Recent years have seen tremendous progress in still-image segmentation;
...
Convolutional networks are powerful visual models that yield hierarchies...
Multiple instance learning (MIL) can reduce the need for costly annotati...
Caffe provides multimedia scientists and practitioners with a clean and
...