about me.
I am a Research Scientist at AMI Labs, working closely with Yann LeCun & Saining Xie on world models, physical intelligence, and robotics. I did my undergrad in CS & Philosophy at Stanford.
Previously, I was at NVIDIA GEAR Lab, where I worked on DreamZero and DreamDojo. We planted the seed of ego-centric human video scaling in world models & created WAMs. Before that, I worked on sparse visual representations at Together AI. I’ve been lucky to be the youngest on every team I’ve been a part of, and I think that is the reason I grew so much in the last few years.
I grew up in Lahore, Pakistan, where I worked on education reform and helped develop Pakistan’s Single National Curriculum, which standardizes education across all schools to eliminate social and educational disparities.
Reach me at ayaannaveed04 [at] gmail [dot] com.
work.
DreamZero: World Action Models are Zero-shot Policies
A World Action Model built on a pretrained video diffusion backbone that jointly predicts future world states and actions, achieving over 2x improvement in generalization to new tasks and environments compared to state-of-the-art VLAs in real-robot experiments, with real-time closed-loop control at 7Hz.
DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
A foundation world model pretrained on 44k hours of diverse human egocentric videos—the largest dataset to date for world model pretraining—that demonstrates strong generalization to diverse objects and environments, with stable real-time interactions at 10 FPS for over 1 minute after distillation.