Incredibly excited to announce the release of MultiNet v0.2 - a major update to our comprehensive open-source benchmark suite for evaluating Multimodal Models on Action tasks.
Read on for several paper announcements, details on the evaluation harness and platform, and more!
This work was done w/ incredible collaborators including @pranavguru13@Yangyue_Wang and @pliang279.
Twenty years.
426 games for @ManUtd. 138 goals.
Five major trophies.
60 @England caps. Goals in major international tournaments.
A boy who became a man who used his profile to change national policy so children didn’t go hungry.
@MarcusRashford we are all proud of you
♥️👏🏻
Excited to share our new paper "Benchmarking Vision, Language, & Action Models on Robotic Learning Tasks"
We evaluate how well VLM & VLA models can control robots across 20 different real-world tasks.
Experimental details, Models, Links, & Vision Below ⬇️
This work was done with an incredible team @pranavguru13@devjwsong@Yangyue_Wang@pliang279
@findmyke @figuret20 @ylecun@aidan_mclau Why do you expect a research paper for a consumer product? He posted the paper for the underlying model above (the part that actually required research)