Computer vision models have known performance disparities across attribu...
Despite impressive advances in object-recognition, deep learning systems...
We introduce the Segment Anything (SA) project: a new task, model, and
d...
Disaggregated performance metrics across demographic groups are a hallma...
We explore the extent to which zero-shot vision-language models exhibit
...
Recent research suggests that predictions made by machine-learning model...
Model pre-training is a cornerstone of modern visual recognition systems...
Physical reasoning requires forward prediction: the ability to forecast ...
Understanding and reasoning about physics is an important ability of
int...
In complex tasks, such as those with large combinatorial action spaces,
...