UW RAIVN Lab
Objaverse-XL
We introduce Objaverse-XL, an open dataset of 10+ million 3D objects. With it, we train Zero123-XL, a foundation model for 3D, observing incredible 3D generalization abilities.
Latest Research
DataComp
DataComp is the machine learning benchmark where the models are fixed and the challenge is to find the best possible data!
SugarCrepe
Fixing hackable benchmarks for vision-language compositionality by having models choose the correct caption for an image from two options: one that accurately describes the image and another that's very similar but slightly different compositionally.
UW RAIVN Lab
The Reasoning, AI, and VisioN (RAIVN) Lab in the Allen School at the University of Washington is directed by Prof. Ali Farhadi and Prof. Ranjay Krishna.
Our research is at the intersection of Computer Vision, Machine Learning, Natural Language Processing and Robotics and is targetted towards helping computers improve their perception and reasoning.
RAIVN Lab's famous research break-throughs include YOLO, XNOR-Net, Visual Genome, AI2-THOR, VCR, BiDAF, Objaverse, DataComp, OpenCLIP, Merlot, Model Soups & Matryoshka Representations/Models among many other amazing projects.
Sponsors
